Review study on Vacuum cleaners

Final report

Viegand Maagøe A/S

Van Holsteijn en Kemna B.V.

June 2019

The information and views set out in this study are

those of the author(s) and do not necessarily reflect

the official opinion of the European Commission

Prepared by

Study team:

Mette Rames, Peter Martin Skov Hansen, Annette Gydesen, Baijia Huang, Michelle Peled

and Larisa Maya-Drysdale (Viegand Maagøe A/S)

René Kemna and Roy van den Boorn (Van Holsteijn en Kemna B.V.)

Quality assurance:

Annette Gydesen (Viegand Maagøe A/S)

Contract managers:

Viegand Maagøe A/S

Project website: https://www.review-vacuumcleaners.eu/

Implements Framework Contract: ENER/C3/2015-619-LOT 2

Specific contract no.: ENER/C3/SER/FV 2017-438/03/FWC 2015-619 LOT2/05/SI2.757436

This study was ordered and paid for by the European Commission, Directorate-General for

Energy.

The information and views set out in this study are those of the author(s) and do not

necessarily reflect the official opinion of the Commission. The Commission does not

guarantee the accuracy of the data included in this study. Neither the Commission nor

any person acting on the Commission’s behalf may be held responsible for the use which

may be made of the information contained therein.

This report has been prepared by the authors to the best of their ability and knowledge.

The authors do not assume liability for any damage, material or immaterial, that may

arise from the use of the report or the information contained therein.

Reproduction is authorised provided the source is acknowledged.

More information on the European Union is available on the internet (http://europa.eu).

1. Preface

This Final report for the review study of the Ecodesign Regulation

and the annulled Energy

Labelling Regulation

for vacuum cleaners is the final delivery of the specific contract. As

specified in the contract the Final report concerns all tasks of the MEErP methodology and

includes recommendations for the revision of these regulations.

Task 1 outlines the scope of the regulations and of the review study as well as the relevant

standards and legislation related to vacuum cleaner energy consumption, durability and

resource efficiency.

Task 2 gives an overview of the vacuum cleaner market including sales, stock and base

data on consumer costs, as well as an overview of market development trends and

production structures.

Task 3 regards the user behaviour, especially looking at robot and cordless vacuum

cleaners in order to suggest representative testing and energy consumption calculation at

later stages of the study. Furthermore, the end-user relevance of the current test standards

is discussed.

Task 4 reviews the technical aspects of vacuum cleaners as products, and outlines the

current technology levels in terms of average and best available technologies, on both

component and product level. Besides the energy consumption effect, the technologies are

also reviewed in terms of resource efficiency.

Task 5 defines the base cases and the environmental and economic impact of each of them.

The environmental impact is both the energy consumption in the use phase as well as the

material consumption and impact categories are given in the EcoReport tool. The

environmental impact is calculated as the product life cycle cost for the end-user for each

base case.

Task 6 outlines the design options for improving the environmental performance of the

base cases without causing excessive costs for the end-users. Design options are outlined

for both energy and resource efficiency improvements.

Task 7 defines policy options for each base case based on the viable design options and

presents the results on the scenario analyses that estimates the environmental and

economic impact of each of the policy options.

OJ L 192, 13.07.2013, p. 24-34, available at: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32013R0666

OJ L 192, 13.07.2013, p. 1-23, available at: https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32013R0665

The specific aspects to review according to article 7 of the Ecodesign Regulation and of

the annulled Energy Labelling Regulation are:

• The review of the ecodesign regulation in light of technological progress;

• The review of verification tolerances to be used by Member State authorities for

market surveillance purposes;

• whether full size battery operated vacuum cleaners should be included in the scope

and

• whether it is feasible to use measurement methods for annual energy

consumption, dust pick-up and dust re-emission that are based on a partly loaded

rather than an empty receptacle.

It should be noted that this review study was begun before the General Court decision to

annul the Energy Labelling Regulation 665/2013

on November 8, 2018, which took effect

on 18 January 2019. The report therefore includes a review of this regulation, including

the evaluation of its effect according to the better regulation principles. Furthermore the

available market data and development observed in energy efficiency reflects the situation

as it has been with the energy label. Even though the Energy Label is now annulled, it was

in force from 2014 to January 2019, which is of course reflected in the data. Therefore the

text refers to “Regulations” in the plural, because it was in force during the period that was

studied.

https://curia.europa.eu/jcms/upload/docs/application/pdf/2018-11/cp180168en.pdf

 
 
5 
 
2.  Table of Contents 
1.  Preface .......................................................................................................... 3 
2.  Table of Contents ............................................................................................ 5 
3.  List of tables ..................................................................................................11 
4.  List of figures .................................................................................................16 
5.  List of abbreviations .......................................................................................20 
6.  Summary ......................................................................................................24 
6.1  Background .............................................................................................24 
6.2  Scope .....................................................................................................25 
6.3  Standardisation and legislation...................................................................27 
6.4  Market data .............................................................................................28 
  Energy and performance .....................................................................29 
  Product prices ....................................................................................33 
6.5  Use patterns ............................................................................................34 
  Cordless vacuum cleaners ...................................................................35 
  Robot vacuum cleaners .......................................................................35 
  End of Life behaviour ..........................................................................36 
  Consumer relevant testing ..................................................................37 
  Uncertainties of test methods ..............................................................38 
6.6  Technology overview ................................................................................38 
  Mains-operated household vacuum cleaners ..........................................39 
  Commercial vacuum cleaners...............................................................40 
  Cordless vacuum cleaners ...................................................................41 
  Robot vacuum cleaners .......................................................................42 
6.7  Environmental and economic impacts .........................................................44 
6.8  Design options .........................................................................................44 
6.9  Scenarios  ...............................................................................................45 
  Energy efficiency scenarios ..................................................................45 
  Energy label ......................................................................................48 
  Resource efficiency scenarios ...............................................................49 
7.  Task 1: Scope ................................................................................................52 
7.1  Product scope ..........................................................................................52 
  Definitions from the regulations ...........................................................52 
  Definitions from preparatory study .......................................................54 
  Definitions from standards ..................................................................55 

 
 
6 
 
  Description of products .......................................................................55 
  Bagged vs bagless vacuum cleaners .....................................................60 
  Alignment of definitions ......................................................................61 
  Recommendations ..............................................................................64 
7.2  Review of relevant regulations ...................................................................66 
  Legislation and agreements at EU level .................................................66 
  Voluntary agreements at Member State level .........................................73 
  Legislation and agreements at third country level ...................................74 
7.3  Review of relevant standards .....................................................................77 
  Mandate 540 .....................................................................................77 
  Safety standards ................................................................................78 
  Material efficiency standards ................................................................79 
  WEEE and RoHS standards ..................................................................80 
  Other relevant standards .....................................................................82 
  Consumer organizations ......................................................................89 
8.  Task 2: Market data: sales and stock ................................................................91 
8.1  Production and trade ................................................................................91 
8.2  Sales data ...............................................................................................94 
  Market values ....................................................................................97 
8.3  Lifespan ..................................................................................................98 
8.4  Stock .................................................................................................... 100 
8.5  Energy and performance ......................................................................... 101 
  Energy ............................................................................................ 102 
  Cleaning performance ....................................................................... 104 
  Dust re-emission .............................................................................. 106 
  Sound power ................................................................................... 107 
  Cordless vacuum cleaners ................................................................. 108 
  Robot vacuum cleaners ..................................................................... 109 
8.6  Market structure and -actors.................................................................... 110 
  Industry .......................................................................................... 110 
  Distribution structure ........................................................................ 112 
  Other actors .................................................................................... 112 

 
 
7 
 
8.7  Consumer expenditure base data ............................................................. 112 
  Interest and inflation rates ................................................................ 113 
  Consumer purchase price .................................................................. 113 
  Electricity cost ................................................................................. 114 
  Repair & maintenance costs ............................................................... 115 
  End of life costs ............................................................................... 116 
9.  Task 3: Users............................................................................................... 117 
9.1  Use pattern of mains-operated household cleaners ..................................... 117 
  Formula  for  calculating  annual  energy  consumption  for  mains-operated 
cleaners 118 
9.2  Use patterns for commercial vacuum cleaners ............................................ 121 
  Formula for calculating annual energy consumption for commercial cleaners
  121 
9.3  Use pattern of cordless vacuum cleaners ................................................... 124 
  Formula  for  calculating  annual  energy  consumption  for  cordless  vacuum 
cleaners 125 
9.4  Use pattern of robot vacuum cleaners ....................................................... 126 
  Formula for calculating annual energy consumption for robot vacuum cleaners
  127 
9.5  Alternative calculations methods .............................................................. 130 
9.6  Consumer relevance – consumer survey results ......................................... 132 
  Ranking of important parameters ....................................................... 133 
  Floor types ...................................................................................... 134 
  Vacuum cleaner settings ................................................................... 135 
9.7  Consumer relevance – testing .................................................................. 136 
  Carpet test ...................................................................................... 136 
  Hard floor test ................................................................................. 140 
  Specialised nozzles ........................................................................... 141 
  Commercial vacuum cleaner test ........................................................ 142 
  Definition of rated power input ........................................................... 142 
  Cordless and robot vacuum cleaner tests ............................................ 144 
9.8  Testing with part load ............................................................................. 144 
  Dyson vs European Commission ......................................................... 145 
  Definition of part load ....................................................................... 146 

 
 
8 
 
  Current part load definition ................................................................ 148 
  Part load of bagged vs bagless vacuum cleaners .................................. 149 
  Available data for part load testing ..................................................... 151 
  Possible options for considering part load testing ................................. 155 
9.9  Verification tolerances ............................................................................. 158 
9.10  Local infra-structure ............................................................................... 161 
  Electricity ........................................................................................ 161 
9.11  Use of auxiliary products ......................................................................... 163 
9.12  Repair practice ....................................................................................... 165 
9.13  End of life behaviour ............................................................................... 168 
  Estimated second-hand use ............................................................... 168 
  Recyclability of vacuum cleaners ........................................................ 169 
10.  Task 4: Technical analysis ............................................................................. 171 
10.1  Components .......................................................................................... 171 
  Motor.............................................................................................. 172 
  Fan ................................................................................................ 173 
  Receptacle ...................................................................................... 178 
  Filters ............................................................................................. 180 
  Hose ............................................................................................... 181 
  Nozzle ............................................................................................ 182 
  Batteries ......................................................................................... 183 
  Plug and power cord ......................................................................... 187 
10.2  Materials and resource level .................................................................... 187 
  Material consumption in vacuum cleaners ............................................ 187 
  Critical materials and components ...................................................... 191 
  Manufacturing and distribution ........................................................... 193 
  Recycled content .............................................................................. 193 
  Use phase ....................................................................................... 194 
  End of life ....................................................................................... 194 
  Blue Angel requirements ................................................................... 197 
10.3  Products ................................................................................................ 198 
  Mains-operated household vacuum cleaners ........................................ 199 

 
 
9 
 
  Commercial vacuum cleaners............................................................. 201 
  Cordless handstick vacuum cleaners ................................................... 204 
  Robot vacuum cleaners ..................................................................... 207 
11.  Task 5: Environmental and economic impact ................................................... 217 
5.1  Inputs for baseline calculations ................................................................ 217 
11.1  Outputs from baseline calculations ........................................................... 220 
  Mains-operated household vacuum cleaners ........................................ 221 
  Commercial vacuum cleaners............................................................. 223 
  Cordless vacuum cleaners ................................................................. 224 
  Robot vacuum cleaners ..................................................................... 224 
  EU Totals – Environmental impacts ..................................................... 225 
11.2  Consumption of critical raw materials and other materials of high importance 227 
11.3  Life cycle cost ........................................................................................ 229 
12.  Task 6: Design options .................................................................................. 231 
12.1  Household mains-operated vacuum cleaners (BC1) .................................... 231 
  Option 1: More stringent energy requirements ..................................... 232 
  Option 2: More realistic performance, indirectly better energy efficiency . 232 
  Option 3: Recycled content and/or light-weighting ............................... 234 
  Option 4: Increase product life ........................................................... 238 
  Option 5: Recycling .......................................................................... 243 
12.2  Commercial mains-operated vacuum cleaners (BC2)................................... 244 
12.3  Cordless vacuum cleaners (BC3) .............................................................. 244 
12.4  Household robot vacuum cleaners ............................................................ 245 
13.  Task 7: Scenarios ......................................................................................... 246 
13.1  Better Regulation evaluation .................................................................... 246 
  Description of the current regulations and their objectives ..................... 247 
  Baseline and point of comparison ....................................................... 248 
  Effectiveness ................................................................................... 250 
  Efficiency ........................................................................................ 257 
  Relevance ....................................................................................... 264 
13.2  Policy analysis........................................................................................ 268 
  Stakeholders consultation ................................................................. 268 
  Policy measures ............................................................................... 268 
13.3  Baseline scenario - BAU .......................................................................... 270 

 
 
10 
 
13.4  Policy scenarios for energy efficiency and performance ............................... 274 
  Requirements .................................................................................. 275 
  Energy saving potentials ................................................................... 280 
  Total consumer expenditure .............................................................. 283 
  Consumer health potentials ............................................................... 285 
  Conclusions ..................................................................................... 287 
  Label classes ................................................................................... 288 
13.5  Policy scenario for resource efficiency ....................................................... 290 
  Material energy saving potentials ....................................................... 294 
13.6  Parameters on the energy label ................................................................ 297 
13.7  Sensitivity analysis ................................................................................. 298 
13.8  Conclusions and recommendations ........................................................... 300 
14.  Annexes ...................................................................................................... 303 
I.  Annex A – Elaboration of standards ................................................................ 303 
1.  Durability of the hose and operational lifetime of the motor ......................... 304 
2.  Water filter vacuum cleaners ................................................................... 307 
3.  Full size battery operated vacuum cleaners................................................ 307 
4.  Robot vacuum cleaners ........................................................................... 307 
5.  Measurement with market-representative carpet(s) and hard floor(s) ........... 308 
6.  Consumer organization tests .................................................................... 311 
II.  Annex B – GfK data coverage ......................................................................... 315 
III.  Annex C - Sales and stock data ...................................................................... 316 
IV.  Annex D - Calculated collection rate ................................................................ 318 
V.  Annex E– Test results ................................................................................... 320 
VI.  Annex F - Impacts over a lifetime of vacuum cleaners calculated in the EcoReport Tool
  327 
 
   

 
 
11 
 
3.  List of tables 
Table 1: Derived sales of each vacuum cleaner type from 1990 to 2030 ......................28 
Table 2: Stock of different vacuum cleaner types in the EU ........................................29 
Table 3: Unit retail prices in EUR vacuum cleaners, in 2018-prices for EU28 ................33 
Table 4: Use pattern for cordless vacuum cleaners ...................................................35 
Table 5: Use pattern for robot vacuum cleaners .......................................................36 
Table  6:  The  top  fault  rates  (above  10%)  and  causes  for  upright  and  cylinder  vacuum 
cleaners. .............................................................................................................36 
Table 7: BAU, BAT and BNAT of household mains-operated vacuum cleaners in terms of 
energy and performance (2018) .............................................................................39 
Table 8: Household mains-operated vacuum cleaners’ materials (product-life 8 years) ..39 
Table  9:  BAU,  BAT  and  BNAT  of  commercial  vacuum  cleaners  in  terms  of  energy  and 
performance ........................................................................................................40 
Table 10: Commercial vacuum cleaners’ materials (product-life 5 years) .....................40 
Table  11:  BAU,  BAT  and  BNAT  of  cordless  vacuum  cleaners  in  terms  of  energy  and 
performance ........................................................................................................41 
Table 12: Cordless vacuum cleaners’ materials (product-life 6 years, package 0.05 m³) 41 
Table  13:  BAU,  BAT  and  BNAT  of  Robot  vacuum  cleaners  in  terms  of  energy  and 
performance ........................................................................................................43 
Table 14: Robot vacuum cleaners’ materials (product-life 6 years, package 0.05 m³) ...43 
Table 15: Policy Option 1, 2 and 3: Energy and performance related requirements. ......45 
Table 16: 2030 energy consumption and savings in PO1, PO2 and PO3 .......................47 
Table 17: Expected market distribution of energy label classes with the rescaled label ..48 
Table 18: Suggested performance classes ...............................................................49 
Table 19: Requirements in Policy Options 4 .............................................................49 
Table 20: Material energy savings for each base case in 2030 for PO4 and PO5 ............50 
Table 21: Advantages and disadvantages for bagged and bagless vacuum cleaners ......60 
Table 22: Vacuum cleaner product types from different sources .................................61 
Table 23: Outline of Ecodesign requirements ...........................................................66 
Table 24: Vacuum cleaner - the previous, annulled energy label classifications .............68 
Table 25: PRODCOM and HS6 product codes and nomenclature ..................................91 
Table 26: Eurostat, PRODCOM, Total vacuum cleaners with self-contained motor - codes 
27512123+27512125. Trade data relates to extra-EU only ........................................92 
Table 27: Value of EU production and selected Extra-EU trade data 2011-2017 in million 
euros ..................................................................................................................93 
Table 28: Market shares of household vacuum cleaners ............................................95 
Table 29: Derived vacuum cleaner sales from 1990 to 2030 ......................................96 
Table 30: Vacuum cleaner market values ................................................................97 

 
 
12 
 
Table 31: Average unit price for vacuum cleaner in EU according to GfK and Prodcom ..98 
Table 32: Average expected lifetimes and assumed variations used in the stock model, in 
years ..................................................................................................................99 
Table 33: Stock of vacuum cleaners in EU 28 from 2005 to 2030.............................. 100 
Table 34: APPLIA Database 2015-2016, Model count, average energy, power and sound 
power ............................................................................................................... 101 
Table 35:  Average power (in W) of mains-operated household VCs EU  in the year 2016
 ........................................................................................................................ 103 
Table 36: Average power (in W) of mains-operated household VCs EU in the year 2018, 
after tier 2 Ecodesign .......................................................................................... 104 
Table 37: Sound power mains-operated household vacuum cleaners EU 2016 ............ 107 
Table 38: Performance of cordless vacuum cleaners. Test  results from GTT Laboratories
 ........................................................................................................................ 108 
Table  39:  Performance  of  robot  vacuum  cleaners  than  from  separate  sources,  such  as 
consumer test organisations and products for sale online, ....................................... 109 
Table 40: Unit retail prices in EUR for household vacuum cleaners, in 2016-prices for EU28
 ........................................................................................................................ 113 
Table 41: Electricity prices with 2016 as base year will be used ................................ 114 
Table 42: Vacuum cleaner spare part retail prices ................................................... 115 
Table 43: Average total labour costs for repair services in euro per hour, in fixed 2016-
prices ................................................................................................................ 116 
Table 44: Use pattern for mains-operated household vacuum cleaners ...................... 118 
Table 45: Use pattern for commercial vacuum cleaners ........................................... 121 
Table 46: use pattern for cordless vacuum cleaners ................................................ 125 
Table 47: Average annual running hours in different modes for cordless vacuum cleaners.
 ........................................................................................................................ 125 
Table 48: use pattern for robot vacuum cleaners .................................................... 127 
Table  49:  Average  annual  running  hours  in  different  modes  for  robot  vacuum  cleaners
 ........................................................................................................................ 127 
Table  50:  Percentage  of  consumers  rating  parameters  important/very  important  in  a 
purchase situation .............................................................................................. 133 
Table 51: Uncertainty of measuring MUV, results from RRT by CENELC TC59X WG6 ... 152 
Table  52:  Results  on  variation  in  DMT8  filling  according  to  each  of  the  three  “bag  full” 
criteria. Range indicating largest minus lowest measured value ................................ 152 
Table 53: suction power uncertainty for vacuum cleaner no. 1 (bagless, upright vacuum 
cleaner) ............................................................................................................ 153 
Table 54: suction power uncertainty for vacuum cleaner no. 2 (bagged, cylinder/barrel with 
large bag) ......................................................................................................... 153 

 
 
13 
 
Table 55: suction power uncertainty for vacuum cleaner no. 3 (bagged, cylinder with small 
bag) ................................................................................................................. 153 
Table 56: Effect on dust pick-up (carpet) at part load (200g/25g) and full load (400g/50g) 
compared to empty ............................................................................................ 154 
Table 57: Effect on input power at part load (200g/25g) and full load (400g/50g) compared 
to empty ........................................................................................................... 154 
Table 58:  Verification tolerances  set out in the regulations and preliminary indication of 
expanded uncertainties ....................................................................................... 159 
Table 59: Global Energy Architecture Performance Index report – best performing countries
 ........................................................................................................................ 162 
Table 60: Faults experienced with upright vacuum cleaners and cylinder vacuum cleaners
 ........................................................................................................................ 166 
Table 61: Re-use, recycling, heat recovery, incineration and landfill rates assumed for the 
End of life handling of vacuum cleaners ................................................................. 170 
Table 62: Filter classes according to EN 1822:2009 ................................................ 180 
Table 63: Comparison properties of Li-ion battery types (L =Low, M=Moderate, H=high)
 ........................................................................................................................ 186 
Table 64: Cycle life of LI-ion batteries as a function of DoD. .................................... 186 
Table 65: Bill-of-materials, Cylinder Vacuum Cleaner (source: JRC-IES 2015) ............ 188 
Table 66: The assumed material composition in the current study. ........................... 190 
Table 67: List of critical raw materials ................................................................... 191 
Table 68. Base case 1: Household mains-operated vacuum cleaners’ energy, performance, 
price ................................................................................................................. 201 
Table 69.  Base Case 1:  Household mains-operated vacuum cleaners’ materials (product 
life 8 years, package 0.08 m³) ............................................................................. 201 
Table 70. Nilfisk commercial cylinder vacuum cleaner examples (source: Nilfisk.com, Sept. 
2018)................................................................................................................ 202 
Table 71. Base case 2: Commercial mains-operated vacuum cleaners (BC2) .............. 203 
Table 72. Base Case 2: Commercial mains-operated vacuum cleaner materials (product-
life 5 years, package 0.1 m³) ............................................................................... 204 
Table 73: Average data for cordless handstick cleaners collected from online retailers for 
27 models from 16 brands. .................................................................................. 206 
Table 74. Base case 3: Cordless vacuum cleaners’ energy, performance, price, 2018 data
 ........................................................................................................................ 206 
Table 75. Base Case 3: Cordless vacuum cleaners’ materials (product-life 6 years, package 
0.05 m³, dock/charger included) .......................................................................... 206 
Table  76:  characteristics  of  6  robot  vacuum  cleaner  models (source  Stiftung  Warentest 
2017)................................................................................................................ 212 

 
 
14 
 
Table 77: Measurements of robot vacuum cleaner energy consumption when in use, energy 
from battery ...................................................................................................... 214 
Table 78: Measurements of energy consumption from electricity grid ........................ 214 
Table 79. Base Case 4: Robot vacuum cleaners’ Energy and performance .................. 215 
Table  80.  Base Case  4: Robot  vacuum  cleaner  materials  (product-life 6  years,  package 
0.05 m³, dock/charger included) .......................................................................... 215 
Table 81: Base case economic and market data for EcoReport, from task 2. All data is for 
2016. ................................................................................................................ 218 
Table 82: Average annual energy consumption (based on AE values) for each base case in 
2016. ................................................................................................................ 219 
Table  83:  Inputs to calculate  the  environmental impacts  and  where they  are  presented
 ........................................................................................................................ 219 
Table 84: Environmental impacts during the entire lifetime of vacuum cleaners sold in 2016
 ........................................................................................................................ 225 
Table 85: Annual environmental impacts of vacuum cleaners (EU-28 stock) ............... 226 
Table 86: The amount of cobalt, gold and copper and the derived impacts regarding energy, 
emission of CO2-eq and market value in euros per product ...................................... 227 
Table 87: The amount of cobalt, gold and copper and the derived impacts regarding energy, 
emission of CO2-eq and market value in euros for the total stock of vacuum cleaners . 228 
Table 88: The combined impact and value of gold and copper in all air conditioners (stock)
 ........................................................................................................................ 229 
Table 89: Life cycle costs of the three base cases (VAT included) ............................. 230 
Table 90: Annual consumer expenditure in EU28 .................................................... 230 
Table 91 . Prices of plastic injection moulding grades .............................................. 235 
Table 92: Comparison of results of this study to results from the 2013 Impact Assessment 
regarding cumulative savings of key parameters .................................................... 250 
Table 93: Coverage of the previous, annulled energy label data for each vacuum cleaner 
type in scope of the regulations ............................................................................ 253 
Table  94:  Percentage  of  consumers  rating  parameters  important/very  important  in  a 
purchase situation (Source: APPLiA 2018 consumer survey) .................................... 256 
Table  95:  Development  of  average  AE  values  for  household  mains-operated  and 
commercial vacuum cleaners 2020-2030 ............................................................... 270 
Table 96: Policy Option 1, 2 and 3: Energy and performance related requirements. .... 274 
Table 97: Energy savings for each base case in 2030 for PO1, PO2 and PO3 in EU28 .. 282 
Table  98:  Energy consumption of cordless and  robot  vacuum  cleaners in BAU and  PO2, 
kWH/year .......................................................................................................... 283 
Table 99: EU User expenditure for each base case .................................................. 285 

 
 
15 
 
Table 100: Average noise levels of each vacuum cleaner type in 2018, 2025 and 2030 in 
the policy scenarios ............................................................................................ 286 
Table 101: Average dust re-emission levels of each vacuum cleaner type in 2018, 2025 
and 2030 in the policy scenarios ........................................................................... 287 
Table 102: Rescaling of the energy label and assumed distributions .......................... 289 
Table 103: Suggested label classes for the performance parameters on the energy label
 ........................................................................................................................ 289 
Table 104: Policy Option 4 and 5: resource efficiency requirements .......................... 290 
Table 105: Material energy savings for each base case in 2030 for PO4 and PO5 in EU28
 ........................................................................................................................ 296 
Table 106: EU Material end-user expenditures for each base case ............................ 297 
Table 107: parameters suggested for the energy label in PO1, PO2 and PO4 .............. 298 
Table 108: Change in robot vacuum cleaner sales and the effect in BAU, PO1, PO2 and PO3
 ........................................................................................................................ 298 
Table 109: Change in cordless vacuum cleaner sales and the effect in BAU, PO1, PO2 and 
PO3 .................................................................................................................. 299 
Table 110: Change in the expected increase in lifetime in policy option 4 .................. 300 
Table 111: CENELEC TC 59X WG 6 sub-working groups ........................................... 303 
Table 112: IEC TC 59 SC 59F Working groups and advisory groups .......................... 303 
Table 113: Calculated collection rate in EU 2014 .................................................... 318 
Table 114: All impact categories for mains-operated household vacuum cleaners. The life 
cycle phase with the highest impact for each of the categories is highlighted with red text.
 ........................................................................................................................ 327 
Table 115: All impact categories for commercial vacuum cleaners. The life cycle phase with 
the highest impact for each of the categories is highlighted with red text. ................. 327 
Table 116: All impact categories for cordless vacuum cleaners. The life cycle phase with 
the highest impact for each of the categories is highlighted with red text. ................. 328 
Table 117: All impact categories for robot vacuum cleaners. The life cycle phase with the 
highest impact for each of the categories is highlighted with red text. ....................... 329 
 
 
 
 
 

 
 
16 
 
4.  List of figures 
Figure 1: main types of vacuum cleaners included in the scope of the review study ......26 
Figure 2: .............................................................................................................26 
Figure 3: Annual sales and stock numbers for the total vacuum cleaner market 2005-2029
 ..........................................................................................................................29 
Figure 4: The previous, annulled Energy Label classification energy 2015-2016 (sources: 
APPLiA and GfK) ...................................................................................................30 
Figure 5: The previous, annulled Energy Label classification hard floor cleaning 2015-2016 
(sources: APPLiA and GfK) .....................................................................................31 
Figure  6:  The  previous,  annulled  Energy  Label  classification  carpet  cleaning  2015-2016 
(sources: APPLiA and GfK) .....................................................................................32 
Figure 7: The previous, annulled Energy Label classification dust-re-emission 2015-2016 
(sources: APPLiA and GfK) .....................................................................................33 
Figure 8: Dust pick-up for an average cylinder cleaner and the three best robot cleaners 
(source: Stiftung Warentest 2017)..........................................................................42 
Figure 9: Annual energy consumption in each of the three policy scenarios compared to 
BAU ....................................................................................................................47 
Figure 10: Annual consumer costs in each of the three policy scenarios compared to BA
 ..........................................................................................................................48 
Figure 11: GHG emissions in PO4 compared to BAU from 2018 to 2030 .......................50 
Figure 12: End-user expenditure for all vacuum cleaners in EU each year from 2018-2030.
 ..........................................................................................................................51 
Figure 13: Left: Barrel or tub form factor. Right: Sledge form factor ...........................56 
Figure  14:  Upright  or  Beat  &  Brush  vacuum  cleaner  form  factor  (left)  and  roller  brush 
(right) .................................................................................................................57 
Figure 15: Battery operated handstick vacuum cleaners ............................................58 
Figure 16: two examples of 2-in-1 handstick vacuum cleaners and the detached handheld 
vacuum cleaner ....................................................................................................59 
Figure 17: Example of a robot vacuum cleaner .........................................................60 
Figure 18: Overview of vacuum cleaner categories and the level to which they are defined
 ..........................................................................................................................62 
Figure 19: main types of vacuum cleaners included in the scope of the review study ....63 
Figure 20: scenario for sub-categorisation of the cordless vacuum cleaner category ......65 
Figure  21:  The  previous,  annulled  Energy  Label  1  (left)  and  label  2  (right)  for vacuum 
cleaners ..............................................................................................................68 
Figure 22: Floor plan of test-box for cleaning, according to section 5 ..........................86 
Figure 23: Floor plan of straight-line cleaning test according to section 6.....................86 
Figure 24: Floor plan for testing autonomous coverage .............................................87 

 
 
17 
 
Figure  25:  Apparent  VC  consumption 2010-2017  according  to Eurostat  PRODCOM,  with 
estimated fractions of products out of scope of the regulation ....................................93 
Figure 26: Vacuum cleaner ≤1500W and <20L receptacle, EU 2017 imports by origin and 
EU 2017 exports by destination ..............................................................................94 
Figure 27: Vacuum cleaner ≤1500W and <20L receptacle, EU 2017 imports by origin and 
EU 2017 exports by destination ..............................................................................97 
Figure 28: Total annual sales and stock of all vacuum cleaner types in EU-28 ............ 101 
Figure 29: The  annulled Energy  Label classification energy  2015-2016 (sources: APPLiA 
and GfK) ........................................................................................................... 102 
Figure  30:  The  previous,  annulled  Energy  Label  classification  hard  floor  cleaning  2015-
2016 (sources: APPLiA and GfK) ........................................................................... 105 
Figure 31: The previous, annulled Energy Label classification carpet cleaning 2015-2016 
(sources: APPLiA and GfK) ................................................................................... 106 
Figure 32: The previous, annulled Energy Label classification dust-re-emission 2015-2016 
(sources: APPLiA and GfK) ................................................................................... 107 
Figure 33: Types of rooms that more than 50% of the respondents in the APPLiA survey 
have in their homes ............................................................................................ 134 
Figure 34: Flooring types in the five most commonly occurring room types ................ 135 
Figure 35: typical dirt types in the five most commonly occurring room types ............ 135 
Figure 36: User behaviour regarding power settings, according to APPLiA consumer survey.
 ........................................................................................................................ 136 
Figure 37: Typical bag-full indicator on bagged vacuum cleaner (left) and bagless vacuum 
cleaner (right) .................................................................................................... 147 
Figure 38: Net electricity generation, EU-28, 2015 (% of total, based on GWh) .......... 161 
Figure 39: Hourly load values a random day in March ............................................. 163 
Figure  40:  Consumer  habits  regarding  changing  bags  and  filter  of  their  main  vacuum 
cleaner, according to the APPLiA consumer survey .................................................. 164 
Figure 41: Hourly labour cost in €, 2016 for European countries ............................... 166 
Figure 42: Expected reprocessing of vacuum cleaners at End of life .......................... 169 
Figure 43: Key components in a mains-operated vacuum cleaner ............................. 171 
Figure 44: Sankey-diagram of energy flows in a mains-operated cylinder vacuum cleaner 
(source: VHK 2017 graph on the basis of AEA Ricardo 2009 data) ............................ 172 
Figure 45: Backwards curved centrifugal fan (left) and fan definitions using the centrifugal
 ........................................................................................................................ 175 
Figure 46: Cordier diagram (Eurovent/EVIA 2016 citing Eck 2003) ........................... 175 
Figure 47: Fan efficiency as a function of specific speed for industrial centrifugal fans in the 
range up to 10 kW (source: Eurovent, EVIA. pers. comm.) ...................................... 177 

 
 
18 
 
Figure 48: The volume of the receptacle is between 1.3 and 3.4 litres. Average size in the 
most recent tests is 2.2 litres ............................................................................... 179 
Figure 49 The principle of a dry vacuum cleaner with a water filter (picture source: Kärcher 
2018)................................................................................................................ 180 
Figure 50: Example of an exploded drawing and spare parts listing for the canister (left) 
and the nozzle plate (right) .................................................................................. 188 
Figure 51. Commercial, cordless, backpack vacuum cleaner (source: Hoover) ............ 202 
Figure 52. Examples of form factors for cordless stick models .................................. 205 
Figure 53: Robot vacuum cleaner (illustrative only, VHK 2018) ................................ 209 
Figure 54: Robot cleaner using a random bounce pattern to cover the surface ........... 210 
Figure 55: Robot cleaner using a random + spiralling pattern to cover the surface ..... 210 
Figure 56: Robot cleaner using SLAM technology to map the room ........................... 211 
Figure 57: Dust pick-up for an average cylinder cleaner and the three best robot cleaners 
on flat floor without crevice (source: Stiftung Warentest 2017). ............................... 212 
Figure 58: Total energy consumption and emission of CO
2
-eq of mains-operated vacuum 
cleaners – the impact of one vacuum cleaner over a lifetime .................................... 222 
Figure  59:  Total  energy  consumption  and  emission  of  CO
2
-eq  of  commercial  vacuum 
cleaners – the impact of one vacuum cleaner over a lifetime .................................... 223 
Figure 60: Total energy consumption and emission of CO
2
-eq of cordless vacuum cleaners 
– the impact of one vacuum cleaner over a lifetime ................................................ 224 
Figure 61: Total energy consumption and emission of CO2-eq of robot vacuum cleaners – 
the impact of one vacuum cleaner over a lifetime ................................................... 225 
Figure 62: Pricing history of recycled injection grade PP (above) versus virgin PP (below). 
Source: www.plasticsnews.com , extract 2018)................................................................. 236 
Figure 63: Conceptual drawing of a recycling sign................................................... 238 
Figure 64: LCC of the base-case (first column) and the durable scenario (second column) 
(source: JRC-IES 2015) ....................................................................................... 241 
Figure 65: Comparison of stock in 2013 Impact Assessment (IA) and the stock estimates 
used in this study ............................................................................................... 249 
Figure 66: Total energy consumption for various scenarios (based on stock) .............. 251 
Figure 67: Greenhouse gas emissions related to electricity consumption in the use phase
 ........................................................................................................................ 251 
Figure 68: Average annual energy consumption of household VC in stock and impact  of 
Ecodesign and Energy Labelling Regulations .......................................................... 252 
Figure  69:  Share  of  energy  savings  due  to  the  Ecodesign  regulation  and  the  previous, 
annulled Energy Labelling Regulation, based on average AE value of sales each year .. 253 
Figure 70: percentage distribution of energy classes for each vacuum cleaner type in 2013, 
label coverage 6% .............................................................................................. 254 

 
 
19 
 
Figure 71: Percentage distribution of energy classes for each vacuum cleaner type in 2016, 
label coverage 85% ............................................................................................ 254 
Figure  72:  Share  of  people  finding  areas  of  the  annulled  label  unclear,  out of  the  70% 
finding at least one parameter unclear (source: APPLiA 2018 consumer survey) ......... 256 
Figure 73: Average total costs of ownership for household users .............................. 258 
Figure 74: Average total costs of ownership for commercial users ............................ 259 
Figure  75:  Manufacturers  turnover  without  regulations  (BAU0)  and  with  the  current 
regulations (BAU). .............................................................................................. 260 
Figure 76: Retailers turnover without regulations (BAU0) and with the current regulations 
(BAU). .............................................................................................................. 261 
Figure 77: Importance of the energy label for future vacuum cleaner purchases ......... 267 
Figure  78:  Expected energy  consumption  development  in  the BAU  scenario, 2015-2030
 ........................................................................................................................ 272 
Figure  79:  Expected  annual  greenhouse  gas  emissions  in  the  BAU  scenario  2015-2030
 ........................................................................................................................ 272 
Figure 80: Expected development in consumer life cycle costs in the BAU scenario from 
2016 to 2030 ..................................................................................................... 273 
Figure 81: Energy consumption in PO1, PO2 and PO3 compared to BAU from 2018 to 2030
 ........................................................................................................................ 281 
Figure 82: GHG emissions in PO1, PO2 and PO3 compared to BAU from 2018 to 2030 281 
Figure  83:  Total  end-user  expenditure  for  all  vacuum  cleaners  in  EU28  each  year  from 
2018-2030. ....................................................................................................... 284 
Figure 84: Conceptual drawing of a recycling sign................................................... 294 
Figure 85: Material energy in PO4 compared to BAU from 2018 to 2030 in EU 28 ....... 295 
Figure 86: GHG emissions in PO4 and PO5 compared to BAU from 2018 to 2030 ........ 295 
Figure 87: Material end-user expenditures for all vacuum cleaners in EU each year from 
2018-2030. ....................................................................................................... 296 
Figure 88: Total EU market for floor coverings in 2015, equalling 1900 million m2 and 15% 
of global market ................................................................................................. 309 
Figure 89: left: domestic loop pole, right: domestic cut pile ..................................... 309 
Figure 90: left: Allura Vinyl Tile, right: Viva Cushion vinyl ........................................ 310 
 
   

5. List of abbreviations

Abbreviation

Full name

ABS

Acrylic Butadiene Styrene

AC/DC

Alternating Current/Direct Current

ACD

Approved for Committee Draft

Annual Energy Consumption (kWh/year)

ANEC

European consumer voice in

standardisation

ASE

Average Specific Energy (Wh/m

)

B2B

Business to Business

BAT

Best Available Technology

BAU

Business as Usual

Base Case

BEP

Best Efficiency Point

BEUC

Bureau Européen des Unions de

Consommateurs

BNAT

Best Not Available Technology

GDP

Gross Domestic Product

BOM

Bill-of-Material

Brushless DC (BLDC)

Brushless Direct Current [motor]

Committe Draft

CEN

European Committee for Standardization

CENELEC

European Committee for Electrotechnical

Standardization

CLC/TC

Technical Committe

Cobalt

-eq

Carbon Dioxide Equivalent

CPU

Central Processing Unit

CRM

Critical Raw Material

Decibel

dB(A)

Decibel (Average)

Direct Current

dm³

Cubic Decimetre

DoD

Depth of Discharge

dpu

Dust Pickup

dpu

Dust Pickup (carpet)

dpu

Dust Pickup (Hard Floor)

dre

Dust Re-Emission

European Commission

Electronically Communicated [motors]

ECCP

European Climate Change Programme

ECOS

European Environmental Citizens

Organisation

EEB

European Environmental Bureau

EEE

Electrical and Electronic Equipment

Energy Index

EMC

Electromagnetic Compatibility Directive

EoL

End of Life

EPA

Efficiency Particulate Air filter

EPS

Expanded Polystyrene

EPS

External Power Supply

ErP

Energy-related Product

European Union

EuP

Energy-using Product

EUR

Euro

Eurostat

European Statistical Office

GfK

Growth from Knowledge

GHG

Greenhouse Gas

General Purpose [50% c + 50% hf]

GPSD

General Product Safety Directive

GWP

Global Warming Potential

HEPA

High Efficiency Particulate Air filter

HPLV

High Pressure Low Volume

HREE

Heavy Rare Earth Elements

Harmonized Commodity Description and

Coding Systems

HVAC

Heating, Ventilation, and Air Conditioning

Impact Assessment

Integrated Circuit

ICRT

The Consumer Test Institute

IEC

International Electrotechnical Commission

Infrared

ISO

International Organization for

Standardization

JRC-IES

Joint Research Centre - Institute for

Environment and Sustainability

Kilogram

kPa

Kilopascal

Kiloton

L/s

Liters per Second

LCA

Life Cycle Assessment

LCC

Life Cycle Cost

LCI

Labour Cost Index

LD-PE

Low Density Polyethylene

Li-ion

Lithium-ion

LLCC

Least Life Cycle Cost

LREE

Light Rare Earth Elements

LVD

Low Voltage Directive

Meter

Mandatory

m/s

Meters per Seconnd

Square Meter

m³

Cubic Meter

MEErP

Methodology for Ecodesign of Energy-

related products

This European Standard is currently under development and deals with methods for the

assessment of the ability to repair, reuse and upgrade energy related products. This

standard suggests a horizontal approach for all energy related products. The standard is

described as generic and general in nature which means that it is not intended to be applied

directly but may be cited in relation with product specific or product group harmonised

standards.

Vacuum cleaner that is connected to a ducting system installed in the building

http://ec.europa.eu/growth/tools-databases/mandates/index.cfm?fuseaction=search.detail&id=564

http://ecostandard.org/work-on-material-efficiency-standards-for-ecodesign-finally-kicks-off/,

https://docs.wixstatic.com/ugd/39a2f0_75eb06c438494c8ea0bb578f5b2f6ef0.pdf

The standard provides a general methodology for:

• the ability to repair products

• the ability to reuse products, or parts thereof,

• the ability to upgrade products, excluding remanufacturing.

Furthermore, this standard provides a common framework for future vertical/product

specific standards.

WEEE and RoHS standards

ISO 11469:2016 - Plastics - Generic identification and marking of plastics products

The EN ISO 11469 standard identifies specifies a system of uniform plastic material

marking system. The standard does not cover every aspect of marking (e.g. the marking

process, the minimum size of the item to be marked, the size of the lettering or the

appropriate location of the marking) but the marking system described is intended to help

identify plastics products for subsequent decisions concerning handling, waste recovery or

disposal. The standard refers to ISO 1043-1 for generic identification of the plastics.

EN ISO 1043-2:2011 - Plastics. Symbols and abbreviated terms. Fillers and reinforcing materials

The EN ISO 1043 standard defines abbreviated terms for the basic polymers used in

plastics, symbols for components of these terms, and symbols for special characteristics

of plastics.

IEC TR 62635:2012 - Guidelines for end of life information provided by manufacturers and

recyclers and for recyclability rate calculation of electrical and electronic equipment

IEC/TR 62635:2012(E) provides a methodology for information exchange involving

electronic and electrical equipment manufacturers and recyclers. The standard also

provides a methodology enabling calculation of the recyclability and recoverability rates of

to facilitate optimized end of life treatment operations.

EN 50419:2006 - Marking of electrical and electronic equipment in accordance with Article

11(2) of Directive 2002/96/EC (WEEE)

EN 50419 contains the product marking requirements needed to ensure compliance with

the WEEE Directive. EN 50419 also contains additional information relating to the marking

requirements, including positioning, visibility, dimensions, location and referenced

documents. The marking requirements are applicable to all manufacturers and producers

of electrical and electronic equipment placing products on the EU market.

EN 50625-1:2014 Collection, logistics & treatment requirements for WEEE - Part 1: General

treatment requirements

EN 50625 was prepared as part of a series of standards requested in Commission mandate

518 which aim to support implementation and effectiveness of Directive 2012/19/EU

(WEEE). The standard contains requirements applicable to the treatment of all types of

WEEE and addresses all operators involved in the treatment (including related handling,

sorting, and storage) of WEEE. In particular, the standard addresses the following issue

areas:

• Management principles

o Technical and infrastructural pre-conditions

o Training

o Monitoring

o Shipments

• Technical requirements

o General

o Receiving of WEEE at treatment facility

o Handling of WEEE

o Storage of WEEE prior to treatment

o De-pollution (including Annex A normative requirements)

o De-pollution monitoring (including Annex B normative requirements)

o Treatment of non-de-polluted WEEE and fractions

o Storage of different fractions of waste (e.g. plastics, metals etc.)

o Recycling and recovery targets (including Annex C & D normative

requirements)

o Recovery and disposal of fractions

• Documentation

The standard applies to the treatment of WEEE until end-of-waste status is fulfilled, or until

the WEEE is prepared for re-use, recycled, recovered, or final disposal.

EN 50574 on the collection, logistics & treatment requirements

EN 50574 on the collection, logistics & treatment requirements for end of life household

appliances containing volatile fluorocarbons or volatile hydrocarbons.

EN 62321 series - Determination of certain substances in electrotechnical products

The purpose of the harmonized EN 62321/IEC 62321 series of standards is to provide test

methods that will allow determination of the levels of certain substances of concern in

electrotechnical products on a consistent global basis.

EN 50581:2012 - Technical documentation for the evaluation of electrical and electronic

products with respect to restriction of hazardous substances

The EN 50581 standard specifies the technical documentation a producer of EEE has to

collect for applicable substance restrictions in order to demonstrate compliance with

Directive 2011/65/EU of the European Parliament and of the Council of 8 June 2011 on the

restriction of the use of certain hazardous substances in electrical and electronic equipment

(RoHS). The technical documentation required to meet the standard includes:

• A general product description

• Documentation of materials, parts and/or sub-assemblies

• Information showing the relationship between the technical documents and respective

materials, parts and/or sub-assemblies

A list of harmonized standards and/or technical specifications used to prepare the technical

documents.

Other relevant standards

This paragraph is intended to give an overview of other standards used to test vacuum

cleaners. These can standards for dry, wet or commercial vacuum cleaners, it is a non-

exhaustive list that is included to show the big diversity in test standards related to vacuum

cleaners.

IEC 62885-8 ED1: Surface cleaning appliances - Part 8: Dry vacuum cleaners for commercial use

- Methods for measuring the performance

This standard was developed particularly for commercial vacuum cleaners to better

simulate the use hereof. The standard includes measurement of vacuum cleaner

performance in terms of debris pickup on hard floor simulated by vacuuming M3 brass nuts

laid out in a specific pattern. Furthermore, the standard includes measurement of the

push/pull forces, or motion resistance, on carpet, which is a safety criterion under the

machinery directive. The standard applies to commercial vacuum cleaners, meaning

vacuum cleaners compliant with the Machinery Directive rather than the Low Voltage

Directive.

EN 60704-2-1:2015 “Household and similar electrical appliances. Test code for the

determination of airborne acoustical noise. Particular requirements for vacuum cleaners”.

Note that this standard does not apply to commercial vacuum cleaners, for which noise is

measured according to EN 60335-2-69 as a safety criterion under the machinery directive.

This standard applies to electrical vacuum cleaners (including their accessories and their

component parts) for household use, or under conditions similar to those in households.

This part of IEC 60704 applies as it is to electrical vacuum cleaners operating in dry

conditions.

IEC 60704-2-17 “Household and similar electrical appliances - Test code for the determination

of airborne acoustical noise - Part 2-17: Particular requirements for dry cleaning robots for

household use”.

This standard is being developed by IEC SC 59F WG 2 to test airborne acoustical noise for

dry cleaning robots and is currently in the ACD stage

Approved for Committee draft.

http://www.iec.ch/dyn/www/f?p=103:38:4477692311473::::FSP_ORG_ID,FSP_APEX_PAGE,FSP_PROJECT_ID:1395,20,23534

EN 60704-3:2006 "Household and similar electrical appliances - Test code for the

determination of airborne acoustical noise - Part 3: Procedure for determining and verifying

declared noise emission values".

This part of IEC 60704 describes procedures for determining and verifying the declared

values of the noise emitted by household and similar appliances. It applies to all categories

of household and similar electrical appliances covered by IEC 60704-1 and IEC 60704-2

dealing with particular requirements for special categories of appliances. It applies to

appliances being produced in quantity (in series, batches, lots) manufactured to the same

technical specification and characterized by the same labelled value of noise emission.

EN 62826:2014 “Surface cleaning appliances - Floor treatment machines with or without

traction drive, for commercial use - Methods of measuring the performance”.

This International Standard lists the characteristic performance parameters for walk-

behind and ride-on floor scrubbers and sweepers and other floor cleaning machines

according to IEC 60335-2-72

. This standard does not apply to IEC 60312 series.

The intent is to serve the manufacturers in describing parameters that fit in their manuals,

and in their literature. This may include all or some of the parameters listed in this definition

document. When any of the parameters listed in this document are used, they are noted

as being measurements made in accordance with this document.

EN 62929:2014 “Cleaning robots for household use - Dry cleaning: Methods of measuring

performance”.

This International Standard is applicable to dry cleaning robots for household use in or

under conditions similar to those in households. The purpose of this standard is to specify

the essential performance characteristics of dry cleaning robots and to describe methods

for measuring these characteristics. The standard describes several tests:

• Measuring the dust removal in a box (hard floor and carpets):

• Measuring dust removal in a straight line (hard floor and carpets):

• Autonomous navigation/coverage test

• Average robot speed

EN 61960-3:2017 “Secondary cells and batteries containing alkaline or other non-acid

electrolytes.

Secondary lithium cells and batteries for portable applications. Prismatic and cylindrical

lithium secondary cells, and batteries made from them”. Includes measurement methods

Household and similar electrical appliances – Safety – Part 2-72: Particular requirements for floor treatment machines with or

without traction drive, for commercial use

for battery performance, including electrical measurements, charge measurements and

endurance testing in terms of cycle times the battery can withstand.

IEC 62885-2:2016 “Surface cleaning appliances - Part 2: Dry vacuum cleaners for household or

similar use - Methods for measuring the performance”.

IEC 62885-2:2016 is applicable for measurements of the performance of dry vacuum

cleaners for household use in or under conditions similar to those in households. The

purpose of this standard is to specify essential performance characteristics of dry vacuum

cleaners which are of interest to users and to describe methods for measuring these

characteristics. This standard is not intended for cordless vacuum cleaners.

A new edition is currently under preparation which will incorporate the new content of EN

60312-1:2017 (like amended durability tests, water filter vacuum cleaners etc.). It should

be highlighted that the draft new edition also adopts new tests reflecting better real life

and being more consumer relevant. As an example the debris pick-up test from hard floor

can be mentioned, this is without a predecessor test.

IEC 62885-4:2016 “Surface cleaning appliances - Part 4: Cordless dry vacuum cleaners for

household or similar use - Methods for measuring the performance”.

A standard for cordless (= battery operated) vacuum cleaners is currently under

development at IEC SC 59F WG 7. The designation of this new standard will be IEC 62885-

4 ED1 Surface cleaning appliances - Part 7: Cordless dry vacuum cleaners for household

or similar use - Methods for measuring the performance which is at CD (Committee Draft)

level. Publication is expected for 2019-09

The purpose of this standard is to specify the essential performance characteristics of

cordless dry vacuum cleaners which are of interest to users and to describe methods for

measuring these characteristics. This standard is not intended for mains-operated vacuum

cleaners or cleaning robots. For safety requirements, reference is made to IEC 60335-1

and IEC 60335-2-2. This is still a draft standard and the expected date of publication is

July 2020

. The IEC standard will be submitted for parallel voting at CENELEC.

This standard will be a fragmented standard based on the standard for mains-operated

vacuum cleaners IEC 62885-2. That means that the standard for cordless vacuum cleaners

only contains the deviations from the standard for mains-operated vacuum cleaners. Most

of the tests remain unchanged.

Important changes are:

https://www.iec.ch/dyn/www/f?p=103:23:23463396680231::::FSP_ORG_ID,FSP_LANG_ID:1395,25

https://www.iec.ch/dyn/www/f?p=103:38:1857809424266::::FSP_ORG_ID,FSP_APEX_PAGE,FSP_PROJECT_ID:1395,23,232

• All tests were checked and amended where applicable regarding the duration of a

test with respect to the limited runtime of a cordless vacuum cleaner (e.g. time for

conditioning, running-in procedures, waiting time and alike).

• As a new test the (effective) runtime of a cordless VC introduced which is the time

it takes to go from an original vacuum (negative pressure versus ambient) realised

by a fully charged cordless VC, operating in accordance with the manufacturer's

instructions for the cleaning performance, to a vacuum that is 40% of the original

vacuum. The test shall be performed on both hard floor and carpet. This is presumed

to reflect real-life runtime.

• The test cycle for measurement of the energy consumption is adapted to cordless

vacuum cleaners. The outcome of this test gives the energy used to clean an area

of 10 m².

IEC 62885-4 ED1 also contains a first tentative definition of a 'Non-full size battery

operated vacuum cleaner', i.e. a 'handheld' that is not typically used for floor cleaning, as

'a battery operated vacuum cleaner which when fully charged, cannot clean 15 m2 of floor

area by applying 2 double strokes to each part of the floor without recharge'. It is

mentioned that this definition is not clear enough. Thus it should be extended/amended.

EN 62929:2014 Cleaning robots for household use - Dry cleaning: Methods of measuring

performance

The purpose of this standard is to specify the essential performance characteristics of robot

vacuum cleaners which are of interest to users and to describe methods for measuring

these characteristics.

The standard describes several tests:

• Measuring the dust removal in a box (hard floor and carpets):

Section 5 describes a test with a rectangular dust area of 1300 mm x 500 mm in

the middle of a rectangular box of 2000 mm x 1150 mm where the robot has to

find its own way in picking up the dust during a test run of 15 minutes. There are

two test runs, each with a different starting position of the robot.

This test is designed to give indicative data on the dust removal capability of a

robotic cleaner, while allowing it to function and move in an autonomous way in an

open area with no obstacles. Navigation strategies differ, so the dust removal result

shall always be reported with time taken to deliver that score, to allow for relative

comparison between different products.

Figure 22: Floor plan of test-box for cleaning, according to section 5

• Measuring dust removal in a straight line (hard floor and carpets):

Section 6 describes a straight-line cleaning test, similar to that of a mains-operated

vacuum cleaner, using a dust area of 700 mm x (Nozzle width -20 mm) and

appropriate acceleration and deceleration zones before and after the 700 mm long

test area to ensure a constant speed.

This test is designed to isolate the dust removal system of the robot from the

autonomous movement, in order to assess only the ability to remove dust. This

facilitates direct comparison between robotic cleaners.

Figure 23: Floor plan of straight-line cleaning test according to section 6

• Autonomous navigation/coverage test

Section 7 describes the determination of how well the robot covers a typical room

area (in cumulative percentage floor area traversed), also measuring multiple floor

area passes. The standardised test room configuration has a floor plan of 4 m x 5 m

with full height walls, furniture and other obstacles, carpet-areas, etc, described in

great detail. In three test runs (each with a different starting position) the robot will

get typically half an hour from each starting position to cover the area efficiently

and effectively. The robot's movements are measured with a Visual Tracking

System (VTS).

The purpose of the autonomous navigation/coverage test is to measure the ability

of floor cleaning robots, as defined within this standard, to cover the available floor

space against a standardised room configuration. The measure of performance for

this test is the cumulative percent floor space traversed during a period of time.

Multiple passes of the robot over the same floor space is also measured in this test.

Figure 24: Floor plan for testing autonomous coverage

This standard is neither concerned with safety nor with performance requirements.

A new edition of this standard is currently under development at IEC SC 59F WG 5. The

designation of this new edition will be IEC 62885-7 ED1 Surface cleaning appliances - Part

7: Dry-cleaning robots for household use - Methods of measuring performance, which is at

CD (Committee Draft) level. Publication is expected for 2019-09

100

This draft standard contains new tests like mobility, debris pick-up, fibre removal from

carpet and energy consumption while the box test will be removed. Tests like corner/edge

cleaning and emissions will be considered for a future edition.

The draft standard describes these new tests:

• Mobility

100

https://www.iec.ch/dyn/www/f?p=103:23:23463396680231::::FSP_ORG_ID,FSP_LANG_ID:1395,25

Section 9 describes a variety of obstacles which can be found in a real environment

at home. These include a minimum gap to go through or to go under, a maximum

transition (floor height offset) and a maximum threshold to go over.

The purpose of these tests is to quantify the capability of a cleaning robot to

overcome various standardised obstacles in defined configurations.

• Debris pick-up

Sections 10 to 12 describe the determination of the capability of the cleaning robot

to pick up debris of various size. Debris that can be found in households is often

organic material. For the sake of repeatability and reproducibility this organic

material is represented by synthetic material of similar size and weight which is

available in defined dimensions. In addition a pre-defined amount of set screws

101

screws and nuts are proposed which are distributed on the carpet

102

• Fibre removal

Section 13 describes the determination of the capability of the cleaning robot to

remove fibre from carpets. Fibres are distributed and embedded on a certain area

of the carpet. Fibre removal performance is evaluated based on visual inspection

Illustrative picture of fibre distribution before and after the test

101

A set screw is a type of screw generally used to secure an object within or against another object, normally not using a nut

102

The plastic (PA6.6) nuts and screws according to ISO 4032 (nuts, M3, weight approx. 0.5 g/piece) and ISO 4766 (screws,

M3 x 6, approx. 0.35 g/piece) are roughly similar in shape as rice and lentils. They are distributed on the test carpet at a

density of 15 g/m² each.

• Energy consumption of a cleaning robot

Section 14 describes the energy consumption of a cleaning robot in different states.

These states are:

1.) Docking station without the cleaning robot

Considers the energy consumption of the docking station in “stand-by”.

2.) Cleaning robot is charged after operating in the navigation test room

Determines the energy consumption for one operation in the navigation room

which is typically half an hour.

3.) Fully charged robot at docking station

Determines the energy consumption of a cleaning robot waiting for the next

cleaning task.

State 3.) is an important part of the robot use but details about this state are not

yet agreed and are under further consideration.

This test is a general method for measuring and calculating the energy consumption

of cleaning robots. This method should be the basis for further definitions of annual

energy consumption for cleaning robots and also mobile household robots.

Note that the CD for IEC 62885-7 ED1 is a preliminary draft that still has to go through

several stages of comments and approvals and thus can be changed considerably before

publication.

IEC/PAS 62611:2009 “Vacuum cleaners for commercial use - Methods for measuring

performance”

These test methods are applicable to vacuum cleaners for commercial use. The purpose of

this PAS is to specify essential performance characteristics of vacuum cleaners being of

interest to the users and to describe methods for measuring these characteristics. For

safety requirements, refer to IEC 60335-1, IEC 60335-2-2 and IEC 60335-2-69.

Work recently started to replace this PAS by a new performance standard for commercial

vacuum cleaners: IEC 62885-8 ED1 Surface cleaning appliances - Part 8: Dry vacuum

cleaners for commercial use - Methods for measuring the performance.

Consumer organizations

Besides industry or Market surveillance testing, consumer organisations also do product

testing. The harmonised standard EN 60312-1 has been used for many years but they also

deviate sometimes and test different aspects. A detailed overview of test performed by

consumer organizations is given in Annex A.

Consumentenbond is a Dutch independent consumer organization who have tested cylinder

vacuum cleaners

103

. Which? is an independent consumer organization based in the UK.

Every year they test over 3600 products and cover the essential features of a product.

They perform tests performed on cylindrical and upright vacuum cleaners

104

, robot vacuum

cleaners

105

and Cordless vacuum cleaners

106

. Stiftung Warentest is an independent

German consumer organization who tests products and services according to scientific

methods in independent institutes and publishes the results in their publications. The

Stiftung Warentest tested corded vacuum cleaners, battery and robot vacuum cleaners.

The Belgian consumer association Test Achats tested cylinder vacuum cleaners in 2017.

107

103

https://www.consumentenbond.nl/stofzuiger/hoe-wij-testen

104

http://www.which.co.uk/reviews/vacuum-cleaners/article/how-we-test-vacuum-cleaners

105

http://www.which.co.uk/reviews/robot-vacuum-cleaners/article/how-we-test-robot-vacuum-cleaners

106

http://www.which.co.uk/reviews/cordless-vacuum-cleaners/article/how-we-test-cordless-vacuums

107

Test Achats, Test d'aspirateurs, juin 2017 - No. 620. www.test-achats.be

8. Task 2: Market data: sales and stock

In the following sections the market for vacuum cleaners is analysed in terms of sales,

stock and prices. The analyses are based on data purchased from GfK on household

vacuum cleaners, supplemented with data from stakeholders. Furthermore, assumptions

from the preparatory study and the impact assessment are applied where necessary and

appropriate.

8.1 Production and trade

The official source of market and stock data is the Eurostat PRODCOM database

108

, in which

data is collected from Member States each year. There are a number of PRODCOM codes

that relate to vacuum cleaners and associated products and are relevant for the study.

However, these product categories, shown in Table 25, group the vacuum cleaner market

in a different way than the regulations. This poses an issue, since the PRODCOM data

encompasses more products than the scope of the regulations, for instance battery

operated vacuum cleaners or wet and dry vacuum cleaners are included in the scope of

the PRODCOM data, but not in that of the regulations.

Table 25: PRODCOM and HS6 product codes and nomenclature

PRODCOM code

PRODCOM Nomenclature (NACE Rev. 1.1, until 2006)

29.71.21.13

Domestic vacuum cleaners with self-contained electric motor for a

voltage >= 110V

29.71.21.15

Domestic vacuum cleaners with self-contained electric motor for a

voltage < 110V

PRODCOM Nomenclature – (NACE Rev 2, from 2007)

29.71.21.23

Vacuum cleaners with a self-contained motor of a power <= 1500 watt

and having a dust bag or receptacle <=20 litres

29.71.21.25

Other vacuum cleaners

29.71.30.10

Parts for vacuum cleaners

New categories were introduced in the PRODCOM database 2007, changing the grouping

of the market data, but not the total number of vacuum cleaners included in the database.

One of the most important changes in light of this study, is that in revision 1.1 the

categories specified that they covered only domestic vacuum cleaners, whereas after 2007

this distinction is not made. For both revisions, however, it is not possible to exclude

vacuum cleaners that are not covered by the scope of the regulation, and thus difficult to

use these categories or the data directly for this study. PRODCOM data will therefore in

this study mainly be used for comparison with data from other sources, and not used for

estimating sales in specific product categories.

108

http://ec.europa.eu/eurostat/web/prodcom/data/database

The production and extra-EU trade data for the total of the two NACE Rev 2 categories is

given in Table 26. Note that the values and prices relate to the manufacturer selling price.

Table 26: Eurostat, PRODCOM, Total vacuum cleaners with self-contained motor - codes

27512123+27512125. Trade data relates to extra-EU only

Exports

Imports

Production

Apparent

consumption

Qty

value

price

Qty

value

price

Qty

value

price

Qty

value

price

mill#

Mill

€

mill#

Mill

€

mill#

mill

€

mill#

mill

€

2010

770

68.8

1910

12.9

1034

72.7

2175

2011

9.7

820

67.5

2044

13.9

1135

71.7

2360

2012

9.6

864

64.6

2259

13.1

1098

68.1

2493

2013

10.2

933

67.4

2330

13.8

1193

71.0

2590

2014

9.9

892

70.6

2434

14.0

1200

74.7

2742

2015

8.8

862

74.6

3113

14.1

1151

79.9

3401

2016

10.3

954

77.3

3070

14.4

1158

81.4

3275

2017

6.4

591

43.9

1773

15.9

1263

53.4

2445

The table shows an apparent EU consumption of around 70 million units in the period 2010

to 2014. In 2015 and 2016, the apparent consumption jumps to approximately 80 million

units, possibly because wholesalers and retailers are stocking up before the second tier of

the Ecodesign measure in 2017. Then in 2017, the apparent consumption drops to 53

million units, possibly because retailers selling their stock from the previous two years.

Based on this, it is concluded that 70 million units constitutes a plausible long term average

apparent consumption for the relevant PRODCOM categories.

Only around half of the quantities in the table relate to vacuum cleaners in scope of the

current regulation. The other half includes vacuum cleaners that are explicitly out of scope,

such as wet, wet & dry, industrial and central vacuum cleaners. Based on the apparent

consumption of vacuum cleaners >1500W in 2017

109

this fraction is estimated at 5.7

million units. Also out-of-scope, by definition, are small handhelds not for floors (see

Chapter 1), USB- or car-battery driven gadgets with a small suction motor, etc. Based on

the GfK figures of approximately 40 million products in scope

110

in recent years and the 5

million out-of-scope products mentioned above, it is estimated that the fraction of small

out-of-scope items is in the order of 25 million units.

Based on these estimates the graph in Figure 25 gives a split of the apparent consumption

for both categories over the period 2010-2017. Despite the large uncertainty in the

109

the year in which the regulation certainly has eliminated all products >1500W in scope

110

This is 35 million units in the current scope and 5 million for cordless and robot products that could possibly be in the new

scope

numbers, it is reassuring that the impact of the Ecodesign Regulation and the annulled

Energy Labelling Regulation is clearly visible even from the Eurostat numbers.

Figure 25: Apparent VC consumption 2010-2017 according to Eurostat PRODCOM, with

estimated fractions of products out of scope of the regulation

Regarding production per country the PRODCOM country-specific data show many gaps,

probably for reasons of confidentiality. For vacuum cleaners ≤1500W the EU28 production

in 2016 was 11.7 million units, of which Germany 3.2 million, Hungary 3.5 million, Italy

0.7 million. The production data for FR, NL, UK, SV, PL, RO and SI were withheld. For

vacuum cleaners >1500W production of 2.7 million units is reported for 2016, of which

Italy 0.48 million, Hungary 0.31 and the UK 0.06 million. Other data is zero or withheld.

Imports, not only of these out-of-scope items, play an important role. PRODCOM statistics

for EU trade (according to HS6) are the only source to estimate the origin of EU imports

and the destination of EU exports. The table and graphs below give the most important EU

trade partners for vacuum cleaners in that respect.

Table 27: Value of EU production and selected Extra-EU trade data 2011-2017 in million

euros

111

Year

Production

Extra-EU import

Extra-EU Export

Apparent

consumption

EU28

Total

China

Malaysia

USA

Total

USA

2011

669

561

346

148

147

1083

2012

636

673

408

194

189

1120

2013

710

697

392

231

222

1185

2014

808

867

569

189

240

1435

111

Eurostat, production data: Prodcom, code 27512123 –vacuum cleaners ≤1500W, ≤20 L receptacle; trade data: EU trade

since 1988 by HS6, code 850811; extracted Sept. 2018

2010 2011 2012 2013 2014 2015 2016 2017

million units

EU-28 Vacuum cleaner apparent consumption 2010-2017

(Eurostat Prodcom, codes 27512123 & 27512125)

out-of-scope>1500W

VC>1500W

VC<=1500W

out-of-scope<=1500W

Ecodesign T1

Sept. 2014

Sept. 2017

2015

811

1395

937

298

274

1932

2016

799

1370

953

267

309

1860

2017

849

1605

1197

258

381

2073

*= There are deviations between different Eurostat codes, which may lead to deviations max. ±5%

Figure 26: Vacuum cleaner ≤1500W and <20L receptacle, EU 2017 imports by origin and EU

2017 exports by destination

8.2 Sales data

As mentioned, around half of quantities in the PRODCOM totals table relate to vacuum

cleaner products that are not in the scope of the regulations. It was thus imperative to find

a more robust source for the EU market. For that reason the sales volumes used for models

and calculations in this study are based on market data purchased from GfK. The GfK data

is based point of sales data on household vacuum cleaners for 22 countries (see Annex B)

with an average market coverage of 87% in these countries.

The GfK data has a high coverage of the European market and only six

112

of the EU-28

countries, representing a total of 3% of the EU population and 1.1% of the GDP

113

, are not

included. Considering the coverage in each of the countries included in the dataset, the

GfK vacuum cleaner data coverage on EU-level is 84% based on population and 85% based

on BNP. This makes the GfK data highly reliable in the sense that it is both precise

(collecting specific data from retailers) and accurate (covers a large number of retailers in

each country).

112

Bulgaria, Cyprus, Latvia, Lithuania, Estonia, and Malta

113

https://europa.eu/european-union/about-eu/countries_en

China (PR)

1197

75%

Malaysia

258

16%

USA

Vietnam

other

vacuum cleaner ≤1500W import

value 2017 (total 1606 Meuro)

USA

17%

Switzerland

15%

Norway

Turkey

Russia

other

125

vacuum cleaner ≤1500W export

value 2017 (total 380 Meuro)

The remaining 15% of sales in the EU, not included in the GfK data, was scaled based on

the coverage % for each country. This means that for each country with tracked data and

e.g. 87% coverage, the remaining 13% was scaled based on the average values for that

specific country. For the six Member States not included in the data (i.e. the remaining 3%

of the population), the average data for all other countries was used to scale to 100%

coverage based on population.

The GfK product categories are very much in line with those in the regulations and include

the mains-operated ‘Cylinder’, ‘Upright’ and ‘Handstick mains’ dry vacuum cleaners as well

as the battery operated categories that are considered in the study (‘Robot’ and ‘Cordless’).

Aggregated EU sales are provided by GfK for the period 2006-2016

114

. For the years 2013-

2016 GfK gives a split per product category. GfK does not give figures on commercial

vacuum cleaners, but based on corrected data from the 2009 preparatory study and inputs

from manufacturers, a share of 12% commercial compared to domestic vacuum cleaners

for cylinder and upright types is assumed

115

Future sales are based on the yearly sales growth rates calculated from the GfK data from

2006 to 2016, which was 1.6% per year for the entire market including both commercial

and domestic products. To make a conservative estimate of future sales, it is estimated

that the 1.6% growth in total sales per year, moved toward 0% per year in 2030. However,

the growth will be different for different product types.

The data shows that cylinder vacuum cleaners are the prevalent type in the EU with a

market share of 68% in 2016, and upright vacuum cleaners are only sold to a lesser extent

(7%). The increase in total sales primarily results from the increased sales of handstick

vacuum cleaners and to a lesser extent the robot vacuum cleaners, which still make up the

smallest market share (4%) despite the increasing sales trend of this category.

Since the market shares of the different vacuum cleaner types are only available for the

years 2013 to 2016, the market split was extrapolated to 2030. Assumptions were made

for the continued development of the market shares for 2025 and 2030 based on

stakeholder inputs, with linear interpolation of market shares in the years between. This

yielded the market shares shown in Table 28. The 2005 market split was calculated from

the preparatory study data, and is assumed unchanged for all years prior to 2005.

Table 28: Market shares of household vacuum cleaners

2005

2010

2015

2018

2020

2025

2030

Cylinder

82%

80%

73%

65%

61%

48%

35%

Upright

14%

11%

114

In the scenario calculations in Task 7 a scaled down version of PRODCOM data will be used, for lack of better data

115

The preparatory study assumed a 6% share of PRODCOM data. Because PRODCOM data are too high, it is now assumed that

this translates into 11% of the GfK data.

Robot

11%

Handstick mains

Handstick cordless

11%

18%

22%

31%

42%

Total

100%

According to vacuum cleaner manufacturers more and more people buy cordless vacuum

cleaners. According to industry most users buy them with the intention of using them for

small cleaning jobs, but end up using them as their main vacuum cleaner. This is also

reflected in the sales, where the market for cordless cleaners is expected to pick up speed

as it becomes more accepted by users. The newest GfK data (YTD April 2017-2018) shows

an accelerating trend with an 11% decrease in cylinder cleaners and a simultaneous

increase in of 24% in cordless sales. Based on these data, it is expected that sales of

cordless cleaners will exceed that of cylinders in around 2028.

The robot market is not increasing as fast as the cordless market, but is expected to pick

up speed as well. This is, however, more uncertain, and a more conservative forecast has

been made for robot sales. Based on the above, the following assessments and projections

were made for household vacuum cleaners.

Since robot vacuum cleaners were not included in the preparatory study because it was a

new technology at the time, the market share was assumed to be 0% for robot vacuum

cleaners in 2005. This is consistent with the fact that the first robot vacuum cleaner was

introduced to the market first in 1996 and then in 2001, but phased out each time due to

poor functionality and high cost, respectively

116

. The first robot vacuum cleaner with

commercial success was the Roomba, introduced in 2002

117

. It is thus assumed that the

market share of robot vacuum cleaners remained in the sub-1% range for approximately

five years. The market split shown in Table 28 together with the total market size result in

the sales figures (shown as million units) in Table 29. Sales for all years can be seen in

Annex C.

Table 29: Derived vacuum cleaner sales from 1990 to 2030

Sales in millions

1990

2000

2005

2010

2015

2018

2020

2025

2030

Cylinder household

14.81

16.92

25.01

25.28

25.07

23.43

22.06

17.88

12.07

Cylinder commercial

1.78

2.03

3.00

3.03

3.01

2.95

Upright Household

2.61

2.99

4.41

3.44

2.91

2.60

2.56

2.38

2.01

Upright Commercial

0.31

0.36

0.53

0.41

0.35

0.31

Handstick mains

0.30

0.34

0.50

0.91

1.25

1.66

1.87

2.38

3.22

Handstick cordless

0.51

0.59

0.87

1.56

4.24

7.39

9.11

13.51

18.10

116

The Electrolux Trilobite in 1996 and the Dyson DC06 in 2001: http://www.vacuumcleanerhistory.com/vacuum-cleaner-

development/history-of-robotic-vacuum-cleaner/

117

http://www.irobot.dk/About-iRobot/About-iRobot

Robot

0.00

0.79

1.45

2.00

2.45

3.58

4.83

Total

20.32

23.22

34.33

35.43

38.28

40.35

41.32

43.00

43.49

The graph below combines the PRODCOM data for vacuum cleaners out of scope with the

GfK sales data.

Figure 27: Vacuum cleaner ≤1500W and <20L receptacle, EU 2017 imports by origin and EU

2017 exports by destination

No split of sales per VC type is available per EU country. Based on anecdotal data it is

known that some of the product types, for example stick vacuum cleaners and upright

vacuum cleaners, are sold to specific countries. The upright cleaners are primarily sold in

the UK, while the handsticks are primarily sold in Italy. According to Euromonitor

118

, 54%

of the handsticks sold in 2016, where sold on the Italian market, followed by 12% in France

and 11% in Germany.

Market values

The purchased data from GfK provides data on value of the EU vacuum cleaner market,

based on point of sales data, i.e. the end-user prices. The data is shown in Table 30.

Table 30: Vacuum cleaner market values

Market values, million EUR

2006

2007

2010

2015

2016

2018*

118

Bissell, presentation by Ken Lee

cylinder domestic

26.51

upright domestic

3.08

stick mains

1.32

stick cordless

4.48

robot

1.53

cylinder commercial

3.18

upright commercial

0.37

handhelds,

gadgets, etc.: 25

wet, wet&dry,

industrial,

central

EU28 vacuum cleaner market 2015

(total ~70 M units, ~40 M units in scope)

GfK market value

2 735

2 894

2 865

4 200

4 367

5 018

*Projected data from first half of 2018

PRODOM collects production data and thus corresponds to manufacturer selling price,

import and export prices, whereas GfK collects point of sales data, i.e. the end-user prices.

The difference in data collection point means that the calculated average unit price is

available for both production and point-of-sales, leading to an estimation of the average

mark-up factor, as seen in Table 31. The mark-up factor is defined as the difference in

manufacturer selling price and the end-user purchase price, and are used in economic

calculations.

Table 31: Average unit price for vacuum cleaner in EU according to GfK and Prodcom

Unit prices, EUR

2006

2007

2010

2015

2016

2018*

GfK unit price

109

112

147

148

139

PRODCOM unit price

n/a

Mark-up factor

2.57

2.24

3.73

3.4

3.7

* Projected data from first half of 2018

8.3 Lifespan

The lifespan of the different product categories is used to determine how long they are in

use after purchase, and thus for how long they are a part of the energy-consuming stock.

In the preparatory study, it was determined that the lifetime of household vacuum cleaners

ranged between 6.3 and 10 years according to various sources and an average lifetime of

8 years was used in that study

119

. A lifetime of 8 years on average is backed up by a 2016-

survey made by consumer reports, but with emphasis on the variation in lifetime between

brands

120

. According to an Austrian study from AK Wien in 2015 the average expected

lifetime of vacuum cleaners by consumers is 10.3 years

121

. This does not reflect the actual

lifetime, but shows that consumer might expect products to last longer than they actually

do. Based on these sources an average of 8 years lifetime with a standard deviation of 2

years is used for mains-operated household vacuum cleaners in this study.

For commercial vacuum cleaners no sources were found that reported the lifespan. It is

therefore assumed that the three times higher amount of use hours per year compared to

household vacuum cleaners

122

will decrease the lifespan in years. One third of the lifespan

would be 2.7 years, however it is also assumed that the cleaners are built more robustly

119

Preparatory Studies for Eco-Design Requirements of EuPs (II), Lot 17 Vacuum cleaners, TREN/D3/390-2006, Final Report

February 2009, carried out by AEA Energy & Environment, Intertek, and Consumer Research Associates between November

2007 and January 2009. https://www.eceee.org/static/media/uploads/site-2/ecodesign/products/vacuum-cleaners/vacuum-

cleaners-ecodesign-study-final-report-eup-lot-17-final-report.pdf

120

https://www.consumerreports.org/vacuum-cleaners/how-long-do-vacuum-cleaners-last/

121

https://www.arbeiterkammer.at/infopool/wien/Bericht_Produktnutzungsdauer.pdf

(https://wien.arbeiterkammer.at/service/studien/Konsument/index.html)

122

Preparatory Studies for Eco-Design Requirements of EuPs (II), Lot 17 Vacuum cleaners, TREN/D3/390-2006, Final Report

February 2009, carried out by AEA Energy & Environment, Intertek, and Consumer Research Associates between November

2007 and January 2009. table 13, page 43. https://www.eceee.org/static/media/uploads/site-2/ecodesign/products/vacuum-

cleaners/vacuum-cleaners-ecodesign-study-final-report-eup-lot-17-final-report.pdf

than household cleaners and thus that they can withstand a larger number of use hours

over their lifetime. As no specific sources could be found, a rough estimate is that

commercial vacuum cleaners can withstand around double the use hours of household

cleaners (on average), thus leading to a lifespan of around 5 years. Since this number is

based on uncertain assumptions, a standard variation of 2 years will still be used.

The lifetime for robot and cordless vacuum cleaners was more difficult to determine

because these categories are relatively new in the market. The preparatory study

suggested a 5-year lifetime for cordless vacuum cleaners but none for robots. A shorter

life expectancy is very likely for both vacuum cleaner types, as they are dependent on a

battery as power source, which will not last for a full 8 years. Depending on the battery

type, they will last between 300-1000 charging cycles, which again depends on the use

frequency and general maintenance. Furthermore, especially robot vacuum cleaners are

complex and use many small parts and advanced technologies (sensors, cameras, etc.),

which might decrease the life expectancy.

The predominant battery type in cordless and robot vacuum cleaners are NiMH (Nickel

metal hydride) and Li-ion (Lithium-ion) batteries

123

. NiMH batteries usually last between

300-500 charging cycles

124

, which might limit the vacuum cleaner lifetime, but on the other

hand replacement batteries are readily available for almost all robot and most cordless

vacuum cleaners. Searching the internet for user experience on robot vacuum cleaners, 4-

7 years’ service life is not unusual, even without replacing batteries. It is assumed that the

technology has been improved since the preparatory study, also in terms of lifespans, and

a 6-year lifetime is therefore used for cordless and robot vacuum cleaners in this study,

but with a standard deviation of 3 years, as this is a quite uncertain approximation.

The lifespans and standard deviations (with presumed normal distribution of lifespans)

used in this study shown in Table 32.

Table 32: Average expected lifetimes and assumed variations used in the stock model, in

years

Vacuum cleaner type

Average lifespan

(Years)

Standard variation

(Years)

Cylinder Household

Upright Household

Cylinder Commercial

Upright Commercial

Cordless

Robot

123

http://www.pickvacuumcleaner.com/vacuum-cleaner-battery-types.html

124

https://www.canstarblue.com.au/appliances/cleaning/vacuum-cleaners/robot-vacuum-cleaners-buying-guide/ ,

https://www.batteribyen.dk/batterityper-og-teknologier

100

8.4 Stock

The stock of vacuum cleaners in the EU-28 is calculated based on the sales figures

described in section 8.2, and the expected lifespans described in section 8.3. Normal

distribution of the lifetime was applied to the sales volume for each vacuum cleaner type

each year, which yielded the total EU stock shown in Table 33. Stock for all years can be

seen in Annex C.

Table 33: Stock of vacuum cleaners in EU 28 from 2005 to 2030

Stock, million units

2005

2010

2015

2020

2025

2030

Cylinder household

209.97

217.34

213.00

206.71

179.59

140.38

Cylinder commercial

16.72

16.94

16.58

16.38

16.25

Upright Household

34.02

28.54

25.08

23.59

21.45

19.42

Upright Commercial

2.61

2.07

1.85

1.78

1.74

2.14

Handstick mains

5.40

8.36

10.66

12.32

16.77

22.37

Handstick Cordless

7.55

14.19

28.01

39.19

68.58

98.07

Robot

2.21

6.71

9.48

11.69

18.38

27.82

Total

278.48

294.15

304.66

311.65

322.75

326.44

When looking at the sales and the stock in a compiled graph (Figure 28), it is seen that

the sales (and thus the stock) increases over time, resulting in a total stock of 325 million

vacuum cleaners by 2030. The stock based on collected data is thus a little lower than

calculated in the preparatory study

125

and Impact assessment

126

. The sales and stock

figures will be used in subsequent tasks to estimate annual energy consumption.

Assuming a total of 220 million households in EU in 2016, the penetration rate of all types

of household vacuum cleaners in 2018 was on average 1.4 vacuum cleaners per

household

127

. The second vacuum cleaners of most households with more than one is

expected to mostly be robot vacuum cleaners or cordless handstick vacuum cleaners. This

fits partly with the specified stock numbers in Table 33, which shows that cordless

handstick and robot vacuum cleaners make up approximately 14% of the stock in 2016.

125

Preparatory study: 342 million in 2005 (Table 15, Page 44)

126

Impact Assessment: 288 million units in 2005, 355 million in 2010 (domestic only), (Table 2 Page 19) COMMISSION STAFF

WORKING DOCUMENT IMPACT ASSESSMENT (2013) with regard to Ecodesign requirements for vacuum cleaners and the

Energy Labelling of vacuum cleaners. http://ec.europa.eu/smart-

regulation/impact/ia_carried_out/docs/ia_2013/swd_2013_0240_en.pdf

127

Total 220 million house holds, http://ec.europa.eu/eurostat/statistics-

explained/index.php/File:Private_households_by_household_composition,_2006-

2016_(number_of_households_in_1_000_and_%25_of_household_types)_new.png

101

Figure 28: Total annual sales and stock of all vacuum cleaner types in EU-28

8.5 Energy and performance

For energy and performance data the study can draw on surveys by GfK 2013-2016 and

the new APPLiA-database with models from 2015 and 2016. Furthermore, a confirmation

of these data can be found in test-results from consumer associations like Stiftung

Warentest (DE), Consumentenbond (NL) and Test Achats (BE) presented in Annex E. As

far as energy is concerned, also the PRODCOM (Eurostat) findings in section 8.1 give an

order of magnitude of the impact.

The GfK data is sales weighted and gives a good coverage of 80-85% of sales for 2016,

but for the first year of the previous, annulled energy labelling (2013) less than 10% of

sales is covered and thus strongly biased. GfK covers cylinders (‘barrel’ and ‘sledge’ form

factor), uprights and mains handsticks. The APPLIA data is based on model count, with a

representative population of almost 1600 models for cylinders, but a clear

underrepresentation of upright and stick models with only a few dozen models. Main

characteristics of the APPLIA database are given in Table 34. In the following sections the

label classifications of APPLiA for 2015-2016 and GfK 2016 are presented side by side.

Table 34: APPLIA Database 2015-2016, Model count, average energy, power and sound power

Cylinders

Upright

Stick

Others

2015

2016

2015

2016

2015

2016

Model count

1536

1557

Energy kWh/yr

35.3

33.4

29.6

30.9

32.7

29.5

29.9

Power W

812

774

754

767

868

730

789

Sound Power dB

77.1

76.5

88.3

86.0

81.3

80.0

83.1

102

Energy

In 2015 there were three cylinder models in the APPLiA database with an energy use of 20

kWh/year and that, starting from September 2017, would be in the A+ class (ranging from

16-22 kWh/year). Their max power is 600 W, carpet cleaning performance C, hard-floor

cleaning and dust-re-emission are class A. The best upright has an energy use of 27

kWh/year, which just puts it in the energy class A (ranging from 22-28 kWh/year). The

best stick model is 23 kWh/year.

Figure 29: The annulled Energy Label classification energy 2015-2016 (sources: APPLiA and

GfK)

In 2016 the most energy efficient vacuum cleaner was a 485 W hard-floor-only model with

annual energy use of 15.8 kWh/year, available in cylinder and in stick version. After

1.9.2017 such a model would be classified as energy label class "A++"

128

, with hard-floor

cleaning and dust re-emission classes both A

128

. The most efficient general-purpose

vacuum cleaner is a 550 W model with AE of 19.5 kWh/year (A+

128

) with carpet cleaning

performance class B

128

and with hard-floor cleaning and dust re-emission classes both class

128

. The most efficient upright vacuum cleaner is a 700 W universal model using 23.3

kWh/year, with carpet cleaning performance class C

129

, hard-floor cleaning performance

class A

129

, and dust re-emission class F

129

128

According to the previous, annulled Energy Labelling Regulation

129

According to the previous, annulled Energy Labelling Regulation

53%

63%

55%

57%

50%

87%

68%

83%

14%

12%

20%

33%

10%

14%

10%

17%

11%

-20%

20%

40%

60%

80%

100%

120%

2015

APPLiA

2016

APPLiA

2016

GfK

2015

APPLiA

2016

APPLiA

2016

GfK

2016

APPLiA

2016

GfK

Energy Class Vacuum Cleaners 2015-2016

cylinder

(APPLiA n=1536-1557

GfK n=86% of sales)

upright

(APPLiA n=21-12

GfK n=75% of sales)

mains stick

(APPLiA n=22

GfK n= 77% of sales)

36.4 34.3 34.9 31.7 33.0 29.1 32.1 29.8 avg. kWh/a

103

Comparing the APPLIA database figures with the outcomes of the GfK research database

it has to be remembered that GfK reported the residential canister/cylinder type to be

dominant with 68% of 2016 unit sales (17.5 million units). Uprights held an 8% share (2

million units), robots 4% (1.1 million units), handstick mains 4% (1 million units) and the

fast growing handstick cordless 16% (4.1 million units). In the APPLIA database the

cylinder types represent 96% of all models. Sticks and uprights are clearly

underrepresented at each 2% and have a very small sample size.

According to GfK, 55% of cylinder types sold in 2016 scored an A

129

in energy efficiency

(APPLIA 63%), 18% had a B

129

(APPLIA 12%) and over 25% were in the lower classes

(APPLIA 23%). In other words, possibly with a delay of one year, there is a fair

compatibility between GfK and APPLIA data for cylinder vacuum cleaners. For uprights and

sticks, where e.g. GfK reports 87% A’s

129

for energy efficiency of uprights and 83% A’s

129

for sticks, the data do not match in any plausible way. This is probably due to the small

sample size of the APPLIA database for these types.

According to GfK, calculated by multiplying sales with the lower energy label class limits

129

the average annual energy consumption of cylinders in 2016 was 34 kWh/a, of uprights 29

kWh/a and of handstick mains 29 kWh/a. According to APPLIA the cylinders in the 2016

model database scored 34 kWh/a, uprights 33 kWh/a and sticks 32 kWh/a.

Both data-sources are incomplete. E.g. the GfK data covers 86% of sales and for the APPLiA

database that fraction will not be different. Assuming instead that the missing 14-15%

represents the least efficient models (rather than following the distribution of the 85% that

is covered), the average for the total sales of mains-operated vacuum cleaners would be

10% higher, i.e. a value of 38 kWh/year.

GfK and APPLiA also give an assessment of the electric power input, as is shown in

Table 35. Here the two data sources are further apart, with an overall sales weighted

average for all three types of 909 W according to GfK and a model count average of 771

W for APPLiA. In this case also the consistency with the energy consumption average of 38

kWh/year has to be taken into account and thus, as will be demonstrated in section 10.3.1

the GfK figure of 909 W makes more sense. The sales weighted average price of mains-

operated vacuum cleaners is 122 € (see section 8.7.2 hereafter).

Table 35: Average power (in W) of mains-operated household VCs EU in the year 2016

GfK power class (assumed avg. W)

Cylinder

Upright

Mains

handstick

Average

> 0 <= 600W

(550)

35%

> 600 <= 700W

(650)

25%

22%

19%

104

> 700 <= 800W

(750)

28%

19%

28%

> 800 <= 899W

(850)

22%

> 899 <= 1400W

(1150)

30%

31%

16%

> 1400 <= 1600W

(1500)

> 1600W

(1700)

GFK average

936

718

721

909

APPLiA average

774

767

730

771

Unit sales covered GfK, in mln.

19.59

2.09

1.16

22.84

Based on the 2016 distribution of power it is possible to estimate the average power after

Ecodesign Tier 2 comes into application (Sept. 2017). It is assumed that the models>899W

will disappear from the population and will return according to the distribution of the

remaining classes. The table below shows the results, which gives an average power of

704, i.e. 23% lower than in 2016. Comparing this e.g. to the 2018 consumer tests in Annex

E (693W in NL, 709W in DE) this seems reliable.

Table 36: Average power (in W) of mains-operated household VCs EU in the year 2018, after

tier 2 Ecodesign

GfK power class (assumed avg. W)

Cylinder

Upright

Stick

Average

> 0 <= 600W

550

4.6%

4.4%

43.0%

6.5%

> 600 <= 700W

650

42.6%

33.5%

22.9%

40.7%

> 700 <= 800W

750

47.7%

28.7%

34.2%

45.3%

> 800 <= 899W

850

5.2%

33.4%

0.0%

7.5%

GFK average in W

703

741

641

704

Unit sales covered GfK, in mln.

19.59

2.09

1.16

22.84

Cleaning performance

The GfK-picture for hard floor cleaning performance is similar to the one for energy: 52-

56% of vacuum cleaners scored an A

130

, 15-18% a B

, for the uprights 28% featured a

while for the canister it was only 14-15% with still a significant number in lower classes

in 2016. This gives a reasonable match with the APPLIA data as seen in Figure 30. The

sales-weighted average dpu

for mains-operated VCs, all types, is 1.08-1.09.

130

According to the previous, annulled Energy Labelling Regulation

105

Figure 30: The previous, annulled Energy Label classification hard floor cleaning 2015-2016

(sources: APPLiA and GfK)

For carpet cleaning the situation is different from hard-floor cleaning: According to GfK

only 3% of cylinder and mains-powered handstick achieved an A-class

131

rating versus

33% of the uprights in 2016. Especially taking into account the small sample size of

uprights these results are similar to those in the APPLIA data-base

Overall, according to GfK the uprights did better in carpet cleaning, with 27% in B

and

33% in C

. The canister and mains-powered handsticks scored respectively 2 or 5% in

, 32 or 25% in C

and the most populated class was D

with 37% and 47%,

respectively.

Nonetheless, given that 85-90% of mains-operated VC sales are cylinder types, the overall

average dpu

for all types is 0.81 in 2016.

Having said that, the 2016 APPLIA database features 56 ‘AAAA’ cylinder models (A

energy and in all performance classes) and only 1 upright vacuum cleaner and 1 handstick

vacuum cleaner with ‘AAAA’.

131

According to the previous, annulled Energy Labelling Regulation

57%

65%

58%

19%

17%

53%

77%

52%

15%

13%

17%

24%

17%

16%

15%

10%

13%

10%

17%

26%

14%

11%

10%

11%

43%

50%

20%

40%

60%

80%

100%

120%

2015

APPLiA

2016

APPLiA

2016

GfK

2015

APPLiA

2016

APPLiA

2016

GfK

2016

APPLiA

2016

GfK

Hardfloor Cleaning Class Vacuum Cleaners 2015-2016

cylinder

upright

mains stick

1.08 1.09 1.09 1.04 1.04 1.09 1.09 1.08 avg. dpu

106

Figure 31: The previous, annulled Energy Label classification carpet cleaning 2015-2016

(sources: APPLiA and GfK)

Dust re-emission

For dust re-emission the classification of cylinders and sticks by APPLiA is similar to that

found by GfK, but for uprights it is completely different. In fact, GfK finds that more than

70% of uprights have a class A

132

dust re-emission score, whereas only a few (8%) of

upright vacuum cleaners in the APPLIA database have an A

It is difficult from these data to find a convergent value for dust re-emission of all types,

but –giving more weight to the more conservative GfK data— a dre value of 0.3% for the

average mains-operated VC in 2016 is believed to be representative.

132

According to the previous, annulled Energy Labelling Regulation

19%

25%

34%

12%

43%

33%

28%

40%

38%

33%

42%

34%

23%

32%

29%

36%

64%

43%

14%

19%

21%

14%

21%

20%

40%

60%

80%

100%

120%

2015

APPLiA

2016

APPLiA

2016

GfK

2015

APPLiA

2016

APPLiA

2016

GfK

2016

APPLiA

2016

GfK

Carpet Cleaning Class Vacuum Cleaners 2015-2016

cylinder

upright

mains stick

0.81 0.81 0.80 0.86 0.86 0.87 0.77 0.78 avg. dpu

107

Figure 32: The previous, annulled Energy Label classification dust-re-emission 2015-2016

(sources: APPLiA and GfK)

Sound power

The following table gives the sound power data for the EU 2016 from GfK and APPLiA. The

values between the data sources converge, with the cylinder type the most silent and the

uprights the noisiest. Taking the more conservative data from GfK as a yardstick it is

estimated that the overall sales-weighted average for the EU 2016 is 79 dB(A).

Table 37: Sound power mains-operated household vacuum cleaners EU 2016

Noise power classes GfK

Cylinder

Upright

Mains

handstick

< 70 db

70-75 db

10%

75-80 db

40%

21%

>=80 db

42%

96%

76%

Coverage (% of the whole

population)

92%

75%

81%

GfK Linear average*

APPLiA linear average

*= at class values of assumed 67, 73, 78, 83 dB(A)

52%

60%

39%

55%

27%

22%

19%

17%

10%

17%

25%

15%

14%

13%

41%

22%

16%

11%

42%

23%

31%

20%

40%

60%

80%

100%

120%

2015

APPLiA

2016

APPLiA

2016

GfK

2015

APPLiA

2016

APPLiA

2016

GfK

2016

APPLiA

2016

GfK

Dust Re-emission Class Vacuum Cleaners 2015-2016

cylinder upright mains stick

0.17% 0.16% 0.28% n.a 0.74% 0.41% 0.32% 0.77% avg. dre

108

Cordless vacuum cleaners

Since cordless vacuum cleaners are not included in scope of the current regulations, there

is no available data about the energy label. While Market data was available from GfK

(sales volume and value), the performance data is not available, other than from separate

sources, such as consumer test organisations and products for sale online, where some

information is gathered.

In order to obtain data for cordless vacuum cleaners in accordance with the current daft

test standards, GTT Laboratories

133

located in Suzhou China, offered to perform testing

of 13 cordless vacuum cleaner models. The cordless cleaners were tested according to the

IEC Cordless Draft Standard (IEC 62885-4 CDV). Vacuum settings to be based on

manufacturers recommendations for each floor surface, and these settings to be employed

for all relevant tests on the specified floor surface.

• Dust pickup on carpet (all samples)

• Dust pickup on hard floor crevice (all samples)

• Dust re-emissions (all samples)

• Noise level on hard floor (all samples)

• Runtime on hard floor surface (all samples)

• Max air data (all samples)

• Energy consumption – (1 representative sample from each price segment)

• Max motor power (all samples - based on motor name plate. Note: units will

probably need to be disassembled to obtain this information.)

In total 13 cordless cleaners were tested from different price segments, the models were

chosen based on sales reported by Amazon in Germany, Italy, Spain and France in 2018.

The total list of vacuum cleaners and results of the tests can be seen in Annex H. In Table

38 the results (currently only 5 models) are presented:

Table 38: Performance of cordless vacuum cleaners. Test results from GTT Laboratories

134

Average

Highest

Lowest

Motor Rated Power (W)

212.50

525.00

95.00

Motor Power measured carpet (W)

244.28

590.88

126.36

Motor Power measured hard floor (W)

223.05

520.24

114.47

Annual Energy carpet (kWh)

15.92

25.46

7.70

Annual Energy hard floor (kWh)

10.01

18.53

3.47

Air data at the end of nozzle (W)

19.35

81.50

4.90

Dust pick up carpet (%)

65.72

91.50

43.50

Dust pick up hard floor (%)

58.13

106.50

3.40

133

An accredited Laboratory founded in 2013, which performs test for several vacuum cleaner manufacturers, including SEB

(France), Vax (UK), Dirt devil (Germany), Hoover (Italy), Arcelik(Turkey), Euro-Pro (USA) and Panasonic (Japan).

134

An accredited Laboratory founded in 2013, which performs test for several vacuum cleaner manufacturers, including SEB

(France), Vax (UK), Dirt devil (Germany), Hoover (Italy), Arcelik(Turkey), Euro-Pro (USA) and Panasonic (Japan).

109

Dust re-emissions (%)

3.98

8.65

0.001

Noise carpet dB(A) Brush ON

81.68

86.30

77.20

Noise hard floor dB(A) Brush ON

82.81

86.30

78.50

Runtime on carpet t90%rt (min:s)

11:28

22:53

04:47

Runtime on carpet t40%rt (min:s)

16:34

23:24

07:58

Runtime on hard floor t90%rt (min:s)

12:43

23:51

05:08

Runtime on hard floor t40%rt (min:s)

16:51

23:51

08:43

t90%rt : time until vacuum is fully discharged or the vacuum has dropped to 90% of the original.

T40%rt : time until vacuum is fully discharged or the vacuum has dropped to 40% of the original.

This shows that the average cordless cleaner has lower performance than the standard

mains-operated vacuum cleaner. However, the best performing cordless vacuum cleaner

(also one of the most expensive cordless vacuum cleaner) has a performance on par or

close to a cylinder vacuum cleaner. It should be noted that the corded vacuum cleaners

have a higher noise level (all of the tested appliances) than the current requirement of

maximum 80 dB(A) for mains-operated vacuum cleaners. Also, the dust reemission seems

very high for some of the tested cordless vacuum cleaners and the worst performing

vacuum cleaner has a dust reemission of more than 8%.

Robot vacuum cleaners

Since robot vacuum cleaners are not included in the scope of the current regulations, there

is no available data from the energy label. While Market data was available from GfK (sales

volume and value), the performance data is not available, other than from separate

sources, such as consumer test organisations and products for sale online, where some

information is gathered. In Table 39 the results (currently only 5 models) are presented.

Table 39: Performance of robot vacuum cleaners than from separate sources, such as

consumer test organisations and products for sale online,

Average

Highest

Lowest

Standby, dock only (W)

0.99

3.51

0.18

Standby, robot in charger (W)

3.70

8.10

0.40

Fully recharging (0% to 100%) (Wh)

68.18

125.00

30.00

Noise dB - 1.6 m (A)

60.75

70.10

52.60

Battery size (mAh)

2,810

5,800

1,800

Coverage (%)

83%

95%

58%

Runtime (min)

83.03

240.00

25.00

Charging time (min)

168.13

390.00

54.50

DPU Hard Floor (first pass)

0.63

0.95

0.07

DPU Hard Floor (fifth pass)

0.79

0.96

0.23

DPU Carpet (first pass)

0.16

0.53

0.01

DPU Carpet (fifth pass)

0.27

0.57

0.04

110

The current average robot vacuum cleaner has a high energy consumption in standby, a

battery size of a mobile phone (mAh), and a cleaning performance below mains-operated

and cordless vacuum cleaners. However, robots clean with a high degree of autonomy, and

cleans more often and at a lower noise level. The best robot vacuum cleaners have a high

dust pickup on hard floor, but the performance is lower on carpet in general.

8.6 Market structure and -actors

Industry

The household vacuum cleaner market is characterised by a large number of

manufacturers, with the main players being Dyson, TTI group (VAX, Hoover and more),

Electrolux (including AEG), Miele, Bosch/Siemens, and Philips as well as far east brands

such as LG, Panasonic, and Samsung

135

. The cordless vacuum cleaner market is largely

dominated by the same brands.

The robot vacuum cleaner market is to a larger extend dominated by specialised

manufacturers such as iRobot, Neato and Eufy RoboVac, even though many of the above-

mentioned brands today have a robot model.

The European industry association for household vacuum cleaners is APPLiA

136

. Consumers

associations are represented at EU-level by ANEC/BEUC. Other NGOs include ECOS, EEB,

TopTen and CLASP.

The commercial vacuum cleaner market is characterised by fewer large manufacturers.

The main players are Nilfisk, Kärcher and Numatic, but Hako, Tennant and FIMAP also

produce commercial vacuum cleaners, even though most are wet/dry cleaners, which are

not covered by the regulations. The European industry association for commercial cleaning

is EUnited cleaning.

137

As mentioned in the preparatory study

138

, the majority of vacuum cleaners are

manufactured in China or other far east countries, and this has not changed. Many of the

large manufacturers have their own Chinese-based production facilities, while others

purchase from OEM (Original Equipment Manufacturer) companies. The most significant

production companies located in Western Europe is Numatic

139

, who continues to produce

135

Preparatory Studies for Eco-Design Requirements of EuPs (II), Lot 17 Vacuum cleaners, TREN/D3/390-2006, Final Report

February 2009, carried out by AEA Energy & Environment, Intertek, and Consumer Research Associates between November

2007 and January 2009. https://www.eceee.org/static/media/uploads/site-2/ecodesign/products/vacuum-cleaners/vacuum-

cleaners-ecodesign-study-final-report-eup-lot-17-final-report.pdf + http://www.grandviewresearch.com/industry-

analysis/household-vacuum-cleaners-market

136

www.applia-europe.eu

137

https://www.eu-nited.net/cleaning/

138

Preparatory Studies for Eco-Design Requirements of EuPs (II), Lot 17 Vacuum cleaners, TREN/D3/390-2006, Final Report

February 2009, carried out by AEA Energy & Environment, Intertek, and Consumer Research Associates between November

2007 and January 2009. https://www.eceee.org/static/media/uploads/site-2/ecodesign/products/vacuum-cleaners/vacuum-

cleaners-ecodesign-study-final-report-eup-lot-17-final-report.pdf

139

https://www.numatic.co.uk/about.aspx

111

in the UK, and Miele who has a large automatized vacuum cleaner production site in

Germany. Some brands including VAX, Electrolux and Nilfisk also have production in the

US and Mexico.

There are no SMEs making domestic vacuum cleaners. There are two smaller companies,

SEBO and Fimap, that have some commercial dry cleaning vacuum cleaner models in their

catalogue, probably as a distributor and not a manufacturer

The following gives an (incomplete) overview of vacuum cleaner companies, headquarters

(HQ), most recently published revenue and number of employees as well as brand-names

where they differ from the company name:

• TTI (Hong Kong HQ, 6 bn turnover, 22,000 staff, power tools & floor care, VC brands

Hoover, Dirt Devil, Oreck, etc.; power tools brands Ryobi, AEG)

• Midea group (China HQ, home appliances & lighting, 100,000 employees, VC brand

Eureka since 2016)

• Nilfisk (Denmark HQ, 5,800 employees, >1 bn euros)

• Electrolux (Sweden HQ, 82,000 employees, VC production in Hungary; residential

and commercial VC brands AEG, Electrolux, Sanitaire; industrial VC brand

Husqvarna)

• Bissel (US, 2,500 employees, $800 million, market leader US)

• Kärcher (Germany HQ, 12,304 employees, >turnover? 2.5 bn euros)

• Miele (Germany HQ, turnover 3.9 bn euros, 19,500 employees)

• Dyson (UK HQ, turnover £3.5bn (US$4.82bn), >8500 employees)

• BSHG (Germany HQ, turnover 13.8 bn, 60,000 employees, vacuum cleaners brands

Bosch, Siemens)

• SEB (France HQ, turnover 6.5 bn euros, 33,600 employees, vacuum cleaners

brands: Rowenta,

• Fakir (Turkey HQ, site in Germany, vacuum cleaners brands: Fakir, Nilco)

• SEBO (Germany HQ, commercial VCs, small)

• Arçelik (Turkey HQ, 30,000 employees, turnover 4.57 bn euros, vacuum cleaners

brand: Grundig)

• Vorwerk (Germany HQ, 12,000 employees, turnover 2,9 bn euros, vacuum cleaners

brands: Kobold, Neato Robotics)

• Philips (Netherlands HQ, 74,000 employees, turnover 17,8 bn euros)

• LGE (S-Korea HQ, turnover 56 bn euros, 77,000 employees)

• Samsung (S-Korea HQ, turnover ca. 180 bn euros, 320,000 employees)

• Numatic (UK HQ, 885 employees, turnover 124 million GBP (2013), vacuum

cleaners brand: Henry)

• Hako (Germany HQ, only one commercial dry cleaner model in scope)

112

• Tennant (US HQ, turnover 1 bn $, 4297 employees, some commercial dry cleaning

vacuum cleaners s in scope)

• Fimap (Italy HQ, 100-250 employees, some commercial VC models)

• ECOVACS (China HQ, 5,000 employees, turnover? $270 mln)

• iRobot (US HQ, 920 employees, revenue $883.9 mln)

• Distributor brands, amongst others: Clatronic (DE), Inventum (NL), Princess (NL),

Bestron (NL)

Distribution structure

Vacuum cleaners are sold through traditional retail channels, the internet and door-to-

door. In the traditional retail sector the position of larger retail chains such as Metro (Media

Markt), Carrefour, etc. is increasing. The European trade sector is represented by

Eurocommerce. According to GfK the internet sales of ‘small domestic appliances’ (SDA),

including (robot) vacuum cleaners, is increasing rapidly. In 2015 the SDA-internet sales

value rose 22.8% compared to 2014, whereas traditional retail sales value increased only

2.8%

140

. Vorwerk employs the services of 633,000 independent (door-to-door) advisors

to sell its products.

Other actors

Consumers associations are represented at EU-level by ANEC/BEUC. Other NGOs include

ECOS, EEB, TopTen and CLASP.

8.7 Consumer expenditure base data

The average consumer prices and costs experienced by the end-user throughout the

product lifetime are determined by unit prices in the following categories:

• Purchase price

• Repair and maintenance costs

• Electricity costs

• End of life cost

As there are no installation costs for the types of vacuum cleaners included in the study

scope, this was not be included. Each of the other costs are explained in the following sub-

sections. The costs are shown as unit prices for each product, maintenance event, kWh

electricity and so on. The total life cycle costs, which also depend on use patterns and

frequency of events, is discussed in task 5.

140

GfK, ONLINE VS. TRADITIONAL SALES: KEY FACTS FOR TECHNICAL CONSUMER GOODS (TCG) IN EUROPE, Infographic,

2016.

113

Interest and inflation rates

All economic calculations was made with 2016 as base year, as this is the latest whole year

for which data is available. HICP inflation rates from Eurostat

141

will be used to scale

purchase price, electricity prices etc. to 2016-prices. Furthermore, a discount rate of 4%

will be used in accordance with the MEErP methodology.

Consumer purchase price

The consumer price including VAT was calculated from the data on unit sales and total

market value collected by GfK. The data was available for the years 2013-2016 and was

extrapolated back to 2005 based on the total average. The unit prices reported in the

preparatory study were 110 € for all household vacuum cleaner types in 2005 (excluding

robots. The average unit price for each vacuum cleaner type, corrected for inflation

142

be in 2016-prices, are shown in Table 40.

Table 40: Unit retail prices in EUR for household vacuum cleaners, in 2016-prices for EU28

Unit prices, EUR

2005

2010

2013

2014

2015

2016

2018

Cylinder

133

119

110

112

121

119

120

Upright

210

184

169

177

196

171

168

Handstick mains

114

Sales weighted average of mains-

operated vacuum cleaners

145

126

116

118

128

123

Commercial

143

302

269

250

255

274

271

320

Handstick cordless

216

193

180

200

225

220

221

Robot

323

288

268

284

317

344

221

*Projected data

As seen from the table, prices decreased from 2005 to 2010 (actually the decrease

happened from 2006 to 2009), which is assumed to be due to the economic crisis. From

2013 to 2015 the price increased, however in 2016 the prices actually decreased for

cylinder, upright and handstick vacuum cleaners.

The increase in price from 2013 to 2015 is likely to be a result of implementing the

Ecodesign Regulation and the annulled Energy Labelling Regulation, which caused a shift

in design criteria from high wattage to low wattage and high dust pickup

144

. Such a shift is

likely to cause a price increase due to increases in R&D costs, using higher efficiency

electric motors and other parts. The small decrease in prices from 2015 to 2016 could be

141

http://ec.europa.eu/eurostat/statistics-explained/index.php/File:HICP_all-items,_annual_average_inflation_rates,_2006-

2016_(%25)_YB17.png

142

Using the HICP index from Eurostat

143

Based on an online survey and prices from 58 different commercial vacuum cleaners.

144

According to the preparatory study, ”manufacturers have developed products with higher and higher input wattage. These

have been marketed to consumers on the basis that the higher the wattage the better the product cleans to the point that

consumers now associate power rating with cleaning efficiency.”

114

the result of saturation of the market with high-efficiency products and maturation of

technologies allowing for lower manufacturing costs and thus increased competition on

price in the market.

It should be noted that the prices in Table 40 are the sales weighted averages of the entire

EU, but that the average price covers a larger price variety. An example of this price

variation is seen from the prices of the vacuum cleaners tested by the consumer

organisations Consumentenbond, Test Achat and Stiftung Warentest, which are listed in

Annex E, and range from 69 € to 335 €

145

. It is thus reasonable to assume that the average

price for each country differs significantly, and that on average, the cheaper vacuum

cleaners are often chosen by end-users.

For the commercial cleaners, the numbers are more uncertain, as there are only 3 major

players in Europe, and no sales numbers directly available. Furthermore, most of the far

majority of the products are sold as B2B and a fair share with a service/maintenance

package, which reduces the sales price of commercial vacuum cleaners significantly.

According to the chairman of EUnited cleaning

146

an average unit sales price of 100 EUR

for commercial cleaners is consistent with the reported annual sales and total annual

turnover in the commercial vacuum cleaner industry. However, based on an online survey

the average price of commercial vacuum cleaners is assumed to be 331 euro, based on 58

different models, from different countries.

Electricity cost

The annual electricity prices from the PRIME Project

147

was used for the economic

calculations in this study. The electricity prices were reported as €/toe (ton of oil

equivalent) in fixed 2015-prices. They were then converted to €/kWh and corrected for

inflation to fixed 2016-prices as shown in Table 41. The electricity prices were given for

every fifth year and linear interpolation was used in between.

Table 41: Electricity prices with 2016 as base year will be used

148

Price in €/kWh (2016-prices)

Year

Households

Services

2005

0.159

0.127

2010

0.175

0.151

2015

0.194

0.160

2020

0.207

0.174

2025

0.213

0.179

2030

0.216

0.183

145

Note that price can be much higher, especially for robot vacuum cleaners

146

http://www.eu-nited.net/cleaning/association/technical-committee/index.html (Chairman: Charalambos Freed, Nilfisk A/S)

147

https://ec.europa.eu/eurostat/cros/content/prime_en

148

The data from primes suggests an annual increase of approximately 1% in electricity prices

115

Repair & maintenance costs

Regarding repairs, few repair shops exist for vacuum cleaners expect those who handle

warranty repairs. Also, according to various forums and websites, most vacuum cleaner

repairs (exchanging hose, suction head or other external parts) can be performed by the

end-users themselves

149

. However, some internal repairs (e.g. motor and wiring) require

professional expertise. Therefore, the cost of repair can vary greatly depending on the type

of repair.

An internet search was made for various vacuum cleaner spare part providers, to find the

retail prices of various spare parts, which are shown in Table 42.

In addition to repairs when the vacuum cleaner is broken, a general maintenance is also

needed. This consists primarily in changing bags and filters and cleaning the brushes. The

price for bags and filters included in the table are thus considered as maintenance rather

than repair costs. The bags are often sold in packs of different sizes, and according to the

data found, the most common is five bags per pack, with a new filter in approximately half

of them.

For bagless vacuum cleaners there is no need to purchase new bags, but the receptacle

should be emptied regularly, and for upright vacuum cleaners also the belts should be

checked. Furthermore, inspection of the vacuum tube or hose and power cord is

recommended. The cost of the regular maintenance is the bags and filters, which vary

depending on the vacuum cleaner type. Furthermore, the upright and robot vacuum

cleaners especially, would need new brushes and belts as part of the maintenance.

Table 42: Vacuum cleaner spare part retail prices

Spare part type

Price

Min

Max

Average

Wheels

2.3

50.9

18.8

Switch

3.7

46.9

14.6

Cable/rewind

9.5

96.7

31.1

Motor

20.0

147.7

54.8

Carbon brush

5.4

53.5

12.6

Heads

9.3

137.0

48.9

Bag frame

4.0

36.2

17.5

Hose and grips

18.1

107.4

48.2

Belts (upright)

2.3

18.9

6.7

Brush (uprights)

6.8

35.7

18.1

Batteries (robot)

17.1

120.8

59.0

149

https://www.nettoparts.dk/shop/svaerhedsgrad-stoevsuger-14550c1.html

116

Brush (robot)

13.3

45.9

27.6

Filters (Robot)

18.7

26.7

24.1

Battery charger

5.0

88.9

23.8

Bags 5-pack

8.6

If vacuum cleaners need to be repaired by a professional, the average EU average labour

cost in the category “Industry, construction and services (except public administration,

defence, compulsory social security)” is used, as shown in Table 43. The labour cost levels

are based on the latest Labour Cost Survey (currently 2012) and an extrapolation based

on the quarterly Labour Cost Index (LCI). The data covered in the LCI collection relate to

total average hourly labour costs

150

Table 43: Average total labour costs for repair services in euro per hour, in fixed 2016-prices

2005

2010

2011

2012

2013

2014

2015

2016

EU-28 countries, EUR/h

24.5

24.6

24.5

24.4

24.6

25.1

25.3

For bags and filters, it is assumed that bagged vacuum cleaners use two bags per year

151

while bagless cleaners use two filters in their lifetime, assuming that the filters can be

cleaned instead of exchanged, in most cases.

The overall lifetime expenses connected with repair and maintenance are assumed to be

20 euro per year for household mains-operated vacuum cleaners and 31 euro for

commercial cleaners. These prices do not include the price of bags and filters, which is

expected to be 25 € and 40 €, respectively. The repair and maintenance cost can be difficult

to quantify as some products are never repaired and others may be repaired more than

once.

End of life costs

Since vacuum cleaners are covered by the WEEE Directive and producers are responsible

for paying a WEEE tax or in some other way finance the EOL treatment, it is assumed that

end-users will not experience any further EOL costs. The WEEE tax paid by manufacturers

is assumed to be reflected in the sales prices of vacuum cleaners to end-users. In the end-

user life cycle cost calculations, EOL cost is therefore be set to zero.

150

http://ec.europa.eu/eurostat/cache/metadata/en/lc_lci_lev_esms.htm#unit_measure1475137997963

151

Based on inputs form stakeholders and the APPLiA consumer survey

117

9. Task 3: Users

Task 3 looks at the consumer side of the products and describes the use patterns in terms

of how and how much end-users use the different types of vacuum cleaners and what

happens to the products in the end-of-life. An important part of the consumer side for

vacuum cleaners is the discussion on consumer relevance and how well the test

methodologies reflect the user needs.

9.1 Use pattern of mains-operated household cleaners

The use pattern for the vacuum cleaners included in the scope of the current regulations

(i.e. the mains-operated vacuum cleaners) was determined in the preparatory study

152

with the following parameters:

- Average floor area covered per cleaning cycle: 87 m

- Average strokes over floor: 2 double (floor covered 4 times)

- Average cleaning cycles per year: 50

- Average duration of cleaning cycles: 1 hour

- The performance will influence the time spend cleaning

These parameters lead to the following formula for calculating annual energy consumption:

  



   





 







 

  

 



The assumed average floor space for European homes of 87 m

was originally derived from

an assumption of 100 m

average home size, but subtracting built in kitchen, furniture

etc., meaning that only 87 m2 actually needs vacuuming. These numbers are well in line

with 2012 statistics that showed and average dwelling size of 96 m

153

. Also the 2014

study on building heat load for HVACs

154

showed that the floor area was around 91 m

. It

is thus not recommended to change the constant in the formula.

The behavioural aspects such as the number of cleaning cycles per year, number of times

the nozzle passes over the floor (4 in the formula), and the assumption regarding

prolonged cleaning time with lower dust pick-up are more difficult to measure

quantitatively. However, a large survey was conducted by the industry organisation

APPLiA

155

that considered these aspects. Both industry members, consumer organisations

and policy makers had the opportunity to comment on the aspects questioned in the survey

in order to achieve robust results, and the survey itself was conducted by InSites

152

https://www.eup-network.de/fileadmin/user_upload/Produktgruppen/Arbeitsplan/eup_lot17_final_report_issue_1.pdf

153

https://ec.europa.eu/eurostat/statistics-explained/images/1/1e/CH3_PITEU17.xlsx and

https://ec.europa.eu/eurostat/statistics-explained/index.php?title=People_in_the_EU_-_statistics_on_housing_conditions

154

https://ec.europa.eu/energy/sites/ener/files/documents/2014_final_report_eu_building_heat_demand.pdf

155

https://www.applia-europe.eu/

118

Consulting. Compared to the values determined in the preparatory study, the APPLiA

survey showed the following:

- Average floor area per home: 70% between 51 m

and 150 m

- Percentage vacuuming the house at least once per week: ~85%

- Average duration of vacuum cleaning per week: 73 minutes

Even though none of the parameters are directly comparable to the ones in the formula,

the results indicate that the assumptions in the formula are within the same span. The only

directly comparable parameter is the time spent cleaning, which does not enter directly

into the regulation formula, but is an underlying assumption. The APPLiA survey shows

that the time spend vacuum cleaning is on average 73 minutes per week, compared to 1

hour assumed in the preparatory study. This is in line with the data showing increased floor

area to be vacuumed.

Some stakeholders have argued that consumers often over-estimate the time spend

cleaning compared to how long the vacuum cleaner is actually on. In one survey from the

US where 80 families answered that they cleaned around 50 minutes per week, meters on

the vacuum cleaners showed a cleaning time of around 15 minutes per week. However,

the APPLiA survey applied not only a quantities survey (online), but also a qualitative

survey, where consumers kept a ‘diary’ of their vacuum cleaning. This qualitative survey

asked more in-depth questions, which confirmed that the results reflected the actual

cleaning time. This is considered an important difference together with the much larger

number of participants and the geographic coverage. However, by not changing the

formula, which is based on average area, the additional 13 minutes will not be considered

in the energy calculations.

Based on the above findings the use pattern shown in Table 44 is assumed for mains-

operated household vacuum cleaners in the calculations in this study.

Table 44: Use pattern for mains-operated household vacuum cleaners

Parameter

Value

Average floor area covered per cleaning cycle

87 m

Average strokes over floor

4 (2 double)

Average cleaning cycles per year

Average duration of cleaning cycles

73 minutes

Influence of performance on the time spent

cleaning



  

 



Formula for calculating annual energy consumption for mains-operated cleaners

While the use pattern and cleaning habits are included as constants in the annual energy

calculation, the measured factors are what differentiates the vacuum cleaners. For example

119

the measurement and calculation of the ASE (Average Specific Energy Consumption) and

the dpu (dust pick-up).

For general purpose vacuum cleaners, the annual energy consumption is calculated as an

average of the measured carpet and hard floor energy consumptions and performances.

Manufacturers can also choose to specify their products as only for hard floor or only for

carpets, thus calculating the AE based on only the one relevant measurement.

It has been argued that the results are skewed towards the hard floor test, as it is possible

to achieve above 100% dust pick-up in the hard floor test, but not in the carpet test. This

could, however be alleviated with more consumer relevant testing by introducing also a

“debris” test as described in section 9.7

Another, opposing, argument has been made that more emphasis should be on the hard

floor performance and energy consumption, and that the current 50/50 split between hard

floor and carpet should be aligned with the average floor area of each type in Europe,

which is only around 24% carpet in 2017 according to data shared by CEN TC134

156

floor coverings. This would reflect real life better, at least for consumers with the average

share of carpets in their home. However, the inherent risk of weighing one performance

factor over the other is that vacuum cleaners with low carpet performance would be able

to obtain high AE ratings, especially since carpet performance is the parameter in which it

is most difficult to achieve high rankings

157

, this could be a risk. This could be the case in

particular for consumers with carpets in some parts of their home, who expect to buy a

good general purpose vacuum cleaner (high AE label ranking) to be used on both carpets

and hard floors, which then turns out to have poor performance on carpets.

Whether or not to change the weighting therefore depends on what is considered most

important: to have an AE value that is as close to the European average situation as

possible, or to have an AE value where general purpose vacuum cleaners should have equal

emphasis on hard floor and carpet. Based on the principle that consumers purchasing a

general purpose vacuum cleaner to clean both floor types in their home should not have

to compromise with carpet performance (even if only 24% of their floor is covered with

carpets), it is not recommended to change the weighting of the dpu

and dpu

in the AE

formula.

156

According to industry stakeholder that is a member of CEN TC 134

https://standards.cen.eu/dyn/www/f?p=204:7:0::::FSP_ORG_ID:6116&cs=16C6F66C6284BD5B1BF6C202B2189F140

157

At least with the current test standards, see for example the market averages in task2 of this report.

120

Dust pick-up in the formula

Since the formula includes the performance of the vacuum cleaner, i.e. the dust pick-up,

it indicates the efficiency, assuming that the end-user will spend longer time cleaning if

the dust pick-up is lower, hence consuming more energy

158

. This makes the formula more

complex, but it also gives a more realistic calculation of energy consumption than if not

taking performance into account. According to the organisation Topten

159

, however, it is

unclear if end-users really do adjust their cleaning habits to the performance of the vacuum

cleaners

160

, or if they will continue to vacuum as per their current habits. They further

argue that for other products there are precedence for not mixing the energy consumption

and performance into a single parameter. Consumer organisations have argued that the

dpu performance could be tackled through Ecodesign requirements alone, instead of being

a part of the AE formula.

The difference between vacuum cleaners and other products, such as dishwashers or

washing machines, is that there is no pre-installed programme with a specific duration that

can be referenced, but the cleaning time is fully dependent on the user. It therefore seems

reasonable to assume that the end-user will clean until the surface is perceived as “clean”,

which would take longer the lower the dust pick-up performance. Since the purpose of

vacuum cleaners is to remove dust, and products should be compared equally, not having

the dust pickup in the formula would make it possible for vacuum cleaners with very small

motors to achieve the best energy class, without actually achieving the purpose.

The opposite argument, that there is too little emphasis on the dust pick-up in the annulled

energy label formula, has been made by industry members, who argue that picking up

dust is the main purpose of a vacuum cleaner. The dust pick-up is therefore considered to

be an important part of energy efficiency, which is defined as the ratio between

performance and energy consumption. It is argued that improving dust pick-up enough to

improve the energy class of a product is so much more expensive than putting in a smaller

motor, that the design strategy will almost always be the latter. Therefore there is a risk

that end-users purchasing an energy label A

161

product might not get the performance they

expect, especially considering the issues with the dust pick-up tests. The same argument

is used the other way around, stating that vacuum cleaners with high performance in the

dpu tests might have issues with high motion resistance, which also affect the user

experience negatively.

Based on the above considerations it is not recommended to change the formula, but rather

focus on improving the reliability of the measurements, which are used in the formula

158

http://www.topten.eu/uploads/File/Deliverables%20ACT/D2_1_Criteria_Paper_Vacuum_cleaners.pdf

159

http://topten.eu/

160

http://www.topten.eu/uploads/File/Deliverables%20ACT/D2_1_Criteria_Paper_Vacuum_cleaners.pdf

161

According to the previous, annulled label

121

(such as the dust pickup) and to potentially add tests and Ecodesign requirements that

improve consumer relevance as discussed in section 9.7.

9.2 Use patterns for commercial vacuum cleaners

In the preparatory study, the commercial vacuum cleaners were assumed to have a

different use pattern than the household cleaners. In total it was estimated that a

commercial cleaner is 1500 hours per year throughout a lifetime of 8 years. This yields an

average of 187.5 hours per year. However, according to commercial vacuum cleaner

manufacturers this is too few hours per year, since an average year has around 260

working days, and professional vacuum cleaners are not used less than one hour per day.

However, the total of 1500 hours over a lifetime might be realistic. Hence, the difference

is the lifetime, which is assumed to be 5 instead of 8 years in this study, as also mentioned

in chapter 2.3. This gives 300 annual use hours, which corresponds to around 1.15 hours

(~70 minutes) per workday.

Table 45: Use pattern for commercial vacuum cleaners

Parameter

Value

Average floor area covered per cleaning cycle

162

87 m

Average strokes over floor

4 (2 double)

Average cleaning cycles per year

260

Average duration of cleaning cycles

70 minutes

Influence of performance on the time spend

cleaning



  

 



Formula for calculating annual energy consumption for commercial cleaners

Despite the recognition of this difference in use pattern between commercial and household

vacuum cleaners, the same formula is used for both types of products in the current

regulation. This was necessary for the sake of the previous, annulled energy label, in order

for consumers to be able to compare products between these two categories. However, in

the modelling of energy consumption and saving potentials, the commercial use pattern

will be used for commercial vacuum cleaners, assuming 300 hours of cleaning per year.

It should be noted though, that the commercial vacuum cleaner manufacturers have

suggested a completely different measurement method for commercial vacuum cleaners,

which is not based on Annual Energy but on an Energy Index (EI) (see section 9.7.4). Such

an index was suggested for the commercial vacuum cleaners to better fit the needs of the

end-users. For these users, who primarily clean by visual perception of the area being

clean, the debris is especially important, and this is included in the index. The same is the

nozzle width, since this influences the time spend cleaning, which is often done as fast as

162

This is based on the assumption that the nozzle is moved with the same speed over the floor as for domestic cleaners

122

possible in commercial settings. Therefore, the final unit of the EI is m²/min, but also

taking into consideration the quality of the cleaning (i.e. the dpu and debris pick-up).

Furthermore, the sound power is included in the EI, since this is important for the work

environment of the operators of the vacuum cleaners.

The EI consists of a number of equations, all based on existing test methods, except for

the newly approved debris pick-up test for commercial vacuum cleaners:









 







 



  



 





Where:















 















Where W

nozzle

is the nozzle width, v

stroke

is the velocity of the nozzle over the floor, and n

is the number of double strokes used in the tests when measuring cleaning performance.









 















 



 







 













 





















Power here meaning input power measured as watt during the cleaning cycles.









 





























Where L

is the sound power level. In the current standard for commercial vacuum

cleaners, the noise is measured with the nozzle lifted from the floor, hence there is no

difference on hard floor and carpet. In the case that standards are aligned with those for

domestic vacuum cleaners (where noise is measured when the nozzle is on the floor),

however, the equation is prepared for handling separate carpet and hard floor noise.

As seen from the equations, the measured values of dust pick-up, debris pick-up, input

power, and noise are all compared to a base case value. These base case values are based

on a best not yet available technology (BNAT) commercial vacuum cleaner. In other words,

the base values are theoretical best case values. The suggested base values are seen in

Table 46.

123

Table 46: base case values for the suggested equations for the EI for commercial vacuum

cleaners

Parameter

Unit

Carpet value

Hard floor value

Nozzle width, w

300

Input power

200

250

Sound power level

dB(A)

Dust pick-up, dpu

95,0%

115,0%

Debris pick-up, deb

100,0%

Furthermore, the equations contain constants, denoted “K”. These factors are based on

numerous measurements and calculations performed by the commercial vacuum cleaner

manufacturers, in order to reach a realistic result and sensitivity of the EI to each of the

parameters. Based on these analyses the suggested factors in Table 47 were derived.

Where the constants equal 1 they can in principle be removed.

Table 47:

Parameter

Factor, Ki

Nozzle width, w

Input power,

Sound power level

0,5

Dust pick-up, dpu

0,3

Debris pick-up, deb

K1, K2 and K5 in are set to 1, and could thus in principle be removed form the equations.

K3 (noise) is set to 0.5 because otherwise the influence of noise on the EI index would be

too large. The noise is included because this specific product category is used in commercial

settings, where noise is an important part of performance due to working environment.

For domestic products it is more a question of discomfort than actual health due to the

much lower use hours per person. K4 (pick-up incl. debris and fine dust) is set to 0.3

because these values are tested until now with 5 double strokes which means 10 strokes

and according to commercial manufacturers the actual cleaning patterns is closer to 3

strokes over the same area, and therefore 3/10=0,3.

The results of the analyses with different EI equations and factors, can be seen in Annex

G. The vast majority of the commercial vacuum cleaner manufacturers agree with the

above equations and have been involved in the development and measurement work

conducted.

124

9.3 Use pattern of cordless vacuum cleaners

Since cordless vacuum cleaners are often lighter in weight and designed for ease of use

for the consumer, it is reasonable to assume that they are often used for lighter cleaning

tasks, which implies shorter run times, but with an increase in the number of cleaning

cycles. Furthermore, cordless vacuums often do not have sufficient run time to run for as

long as mains-operated (50-73 minutes), as most cordless vacuum cleaners have a battery

life of 15-40 minutes while only a few can run for up to 60 minutes per time, and not at

the highest power

163

Less research exist about how cordless vacuum cleaners are used than for mains-operated,

however a few sources are available, primarily from manufacturers. One survey shows that

cordless cleaners are used for around 20 minutes per cleaning cycle, but several times a

week. Both this and other sources agree that the average use frequency of a cordless

cleaner is 4 times per week

164

. Assuming that they are used for 20 minutes each time,

gives an average of 80 minutes per week, which is close to the 73 minutes reported for

mains operated cleaners. In order to ensure comparison in calculations, 73 minutes and

87 m

cleaned per week will be assumed for cordless vacuum cleaners as well, even though

it is spread out over more cleaning cycles.

As mentioned above the battery lifetime will also influence the cleaning time per cycle.

Most cordless vacuum cleaners can operate in different power levels (for example,

minimum, medium and max), however most cordless cannot operate for 20 minutes in

max mode. Also having the cleaner in max mode all the time might not be necessary for

the end-user. However, this is the mode in which tests are conducted, and annual energy

consumption might therefore be calculated to a higher value than if the cleaner is used in

other than the max mode. This is, however, the same for all cordless cleaners, making

results comparable.

While the cordless vacuum cleaners are used more frequently it is assumed that they are

used to clean the same area as mains-operated per week, since this is based on average

home sizes in Europe. Also the assumptions of 2 double strokes and the influence of

performance on the cleaning time are assumed to be similar to that of mains-operated

vacuum cleaners. The use pattern for cordless vacuum cleaners used for calculations in

this study, based on the available information, is shown in Table 48.

163

http://www.which.co.uk/reviews/cordless-vacuum-cleaners/article/corded-vs-cordless-vacuum-cleaners

164

Based on stakeholder inputs in the study

125

Table 48: use pattern for cordless vacuum cleaners

Parameter

Value

Average floor area covered per cleaning cycle

165

87/4 = 21.75 m

Average strokes over floor

4 (2 double)

Average cleaning cycles per year

200

Average duration of cleaning cycles

20 minutes

Influence of performance on the time spend

cleaning



  

 



Formula for calculating annual energy consumption for cordless vacuum

cleaners

In addition to the above factors, the fact that cordless are battery powered means that

also other parameters are important for their use. One is the battery time, i.e. how long

the vacuum cleaner can be used before the battery needs recharging. Another is the

charging time, which influences how long it takes until the vacuum cleaner can be used

again. The remaining time it is assumed that the cordless vacuum cleaners are standing in

the charger, fully charged, only using power for maintenance charging to make up for the

battery self-discharge.

While the cleaning time is determined by use patterns, the charging time is determined by

technical characteristics. Based on inputs from stakeholders and collection of online data

166

the average weekly time spend charging and in maintenance mode (seen in Table 49) has

been determined. This is based on an average of hard floor and carpet cleaning.

Table 49: Average annual running hours in different modes for cordless vacuum cleaners.

Average time per

week

Average time per

year

Cleaning (standby of dock

without cordless)

73 minutes

63 hours

Charging

13 hours

671 hours

Charged and docked

158 hours

8026 hours

In the current draft standard for the cordless vacuum cleaner test, the energy consumption

is measured by running the fully charged cordless cleaner for five minutes on the carpet /

hard floor (while measuring the dpu) and then measuring the energy necessary for a full

re-charge. This way of measuring thus takes into account the efficiency of the power

supply, and since the test area is known, an average ASE in kWh/m

is easily derived.

165

This is based on the assumption that the nozzle is moved with the same speed over the floor as for domestic cleaners

166

Data for 28 different cordless models and their rated charging times and run times was used to calculated the average

charging time (given by manufacturer) plus 33 models tested by the Danish organization TÆNK https://taenk.dk.

126

While this could be used for a simple energy calculation, it would not be comparable with

the AE value calculated for mains-operated vacuum cleaners. However, using the AE

formula directly would not reflect the annual consumption of cordless vacuum cleaners

because of the many hours spent in maintenance mode. It is therefore suggested to add

the maintenance mode consumption to the cleaning consumption of cordless vacuum

cleaners, and to base the hours spent cleaning / charging / in maintenance mode on the

above data. The formula would thus, without changing the area, be:

  





       

  

  

 





 



Where M

is the maintenance power in “charged and docked” mode in watts. In order to

make the calculations comparable, the same area of cleaning per week is assumed, but

spread over more cleaning cycles. Despite the difference in use pattern, the total area

covered each year would be 4*87*50=17400 for mains and 4*21.75*200 = 17400 for

cordless, hence ensuring the comparison of the two types. This will thus be how the energy

consumption for cordless vacuum cleaners is calculated in this study.

Regarding the measurement data needed, the maintenance mode power consumption

measurement is not yet part of the draft standard, but it should be one of the less

complicated tests to develop. A dpu test has been drafted for cordless vacuum cleaners,

which also gives the dpu in %, equivalent to that of mains operated vacuum cleaners.

Rather than having the annual standby hours as a constant, the charging time for the

specific vacuum cleaner could be used instead to determine the annual maintenance mode

hours. This could be combined with the maintenance mode power measurement (which

starts once the charging is finished) and the run time of the appliance, and would need to

be defined in a test standard. In this study, the average constant shown in Table 49 and

the formula above will be used to calculate AE for cordless vacuum cleaners.

9.4 Use pattern of robot vacuum cleaners

Since robot vacuum cleaners need no human interaction during the cleaning cycle they can

run at times when no one is home, which typically leads to a larger number of cleaning

cycles. All robot vacuums placed on the market today have a timer setting, making it

possible to schedule cleanings during for instance the workday

167

. Many users therefore

run their robot vacuum cleaner every day, some run it 5 days a week, while a few runs it

weekly

168

169

. Different sources report different use patterns, and in this study, it is

assumed that robots are used four times per week, based on the different inputs received.

167

https://www.robotcleanerstore.com/pages/robot-vacuum-cleaners-frequently-asked-questions

168

http://www.explainthatstuff.com/how-roomba-works.html

169

https://www.reddit.com/r/roomba/comments/669dr5/how_often_do_you_use_your_roomba/

127

Most robot cleaners have a declared run time of 60-90 minutes on a fully charged battery

reported at time of sales, i.e. when the battery is new. A comparison of 52 different models

showed an average declared run time of 83 minutes

170

. However, this value does not take

into account gradual deterioration of the battery or mention the load of the motor while

measuring run time. Hence over the course of the lifetime of a robot vacuum cleaner and

considering that it might operate at various loads, it is assumed that average cycle time is

far less than declared, around 30 minutes. This also takes into account that most robots

cannot cross doorsteps and thus when started in one room, cannot cross to another after

the room has been cleaned. The mapping technology determines when the room has been

fully covered, and it is assumed that the robot will finish cleaning once this happens.

Table 50: use pattern for robot vacuum cleaners

Parameter

Value

Average floor area covered per cleaning cycle

171

87/4 = 21.75 m

Average cleaning cycles per year (50 weeks)

200

Average duration of cleaning cycles

30 minutes

Influence of performance on the time spend

cleaning



  

 



Formula for calculating annual energy consumption for robot vacuum cleaners

When the robot vacuum cleaner is not active (cleaning) or charging, it is standing fully

charged in the docking station. The same three power modes as for cordless vacuum

cleaners are therefore relevant for robot cleaners. The average charging times of robots is

based on an online search with 52 models. The assumptions for robot cleaners are

summarised in Table 51.

Table 51: Average annual running hours in different modes for robot vacuum cleaners

Average time per

week

Average time per

year

Cleaning (standby of dock

without cordless)

120 minutes

104 hours

Charging

4.4 hours

211 hours

Charged and docked

162 hours

8445 hours

Even though an average number of cleaning cycles per year and area covered per cleaning

cycle can be found for robot cleaners, it is not directly comparable to that of cordless and

mains-operated, because the robots drive around autonomously. The assumption of 2

170

Based on online surveys and results from The Danish Consumer Council THINK

171

This is based on the assumption that the nozzle is moved with the same speed over the floor as for domestic cleaners

128

double strokes, i.e. covering the area 4 times in total, can therefore not be assumed for

robot vacuum cleaners.

As opposed to manually handled vacuum cleaners, coverage of the area is an important

performance parameter for robot cleaners. How well the floor is covered depend highly on

the robot navigation system. Some of the older navigation technologies in particular can

result in the vacuum cleaner not covering all of the floor, which of course compromise the

performance in terms of cleaning. For example, if the robot does not, in an entire cleaning

cycle, drive over 4 m

out of a 20 m

room, the room coverage can be said to be 80%.

Despite the differences in how robot vacuum cleaners cover the floor in comparison to

manually handled vacuum cleaners, the energy calculations still need to be comparable.

Following the same logic as for the commercial and cordless vacuum cleaners, the area

covered each year still needs to be comparable. However, in the calculation for robots, the

room coverage should be included in a way where low room coverage leads to higher

energy consumption because it de facto decreases the average cleaning area, and having

to clean also this part would require extra energy. Furthermore, the maintenance mode

and charging times are different for robots than for cordless cleaners as seen in Table 49

and Table 51.

While the cordless cleaner draft test standards make it possible to calculate an ASE value

(in Wh/m

) like for mains-operated vacuum cleaners, this is not the case for robot vacuum

cleaners. Instead, the draft test standards specifically define a 20 m

test room (which

takes around 20-25 minutes), let’s the robot clean it, and the measures the energy

consumption for charging the battery afterwards. The energy measure is thus rather

energy per cleaning cycle instead of energy per square meter.

A suggestion for a formula for annual energy consumption for robot vacuum cleaners,

which is used in energy calculations in this study is the following:







 

  





    





 





 



Where E

measured

is the output from the test method, i.e. measured re-charging energy after

cleaning the 20 m

test room. This number is then divided by RCF

172

*20 m

and multiplied

with the average area assumed to be cleaned in an average robot cleaning cycle. This

should be consistent with the area used for the other product types. The addition of the

Room Coverage Factor in % (always between 0 and 1) in the denominator gives the actual

172

Room Coverage Factor

129

area covered of the 20 m

test room. A test method has also been developed to measure

this factor.

Since the dust pick-up test for robot cleaners is different from those for manually handled

vacuum cleaners (no counting of double strokes, measured on flat floor), as explained in

section 7.3, the values cannot be compared directly. Also, the constant 0.2 cannot be used

for robots, since it is the standard deviation between 2 and 5 double strokes, which does

not make sense for robots. The inclusion of the dpu factor in the equation is thus done

differently. While the underlying assumption for manually handled vacuum cleaners is that

end users will spend less time cleaning if the vacuum cleaner has a high dpu, the

assumption for robots is that end users will run them less frequently, if they remove dust

better, especially visible dust.

By comparing the AE calculation with the direct energy calculation (typical annual running

hours and average consumption in each), it was found that by comparing the measured

dust pick-up to the average dust pick-up (for the base case) was the best approximation

to calculating the presumed change in user behaviour caused by the effect of dpu. This is

of course based on some underlying assumptions about how much the user changes

behaviour due to the difference in dpu, i.e. the gradient of the “cleaning time vs dpu”

curve. This was found to be too steep when using the benchmark value for dpu, rather

than the average, primarily because most robot cleaners have much lower dpu than the

benchmark

173

Furthermore, in including of the dpu in the robot equation, the inputs from stakeholders

has been taken into account that the performance on carpet and hard floor should be

included separately, due to the large difference between dpu

and dpu

for robots.

As for cordless cleaners, it can also be contemplated to include the standby consumption

of the docking station standing alone, while the robot is cleaning. This would then be an

extra link in the formula and would require that a test standard is developed. This,

however, should be a relatively easy parameter to measure, and the energy consumption

is very low, so it is not critical to set a requirement. The maintenance mode power

measurement is under consideration for the standards being developed by CENELEC, but

is not included in the calculations in this study.

At this point there is too little data and information available regarding active charging

stations, which can for example empty the dust bin of the robot, to take them into account

in the calculations. However, since the maintenance mode consumption covers the entire

173

Benchmark here meaning the best observed dpu performance found in any robot at the time of the study, which is around

95% on hard floor and 36% on carpet, measured with the IEC 62885-7 (draft) Section 5.6 and Section 5.7, respectively.

130

system (robot, charging station, power supply etc.), a large part of the consumption is

considered covered.

9.5 Alternative calculations methods

During the study several alternative formulas for calculating the annual energy have been

suggested, not just how to include dpu (as discussed in section 9.1.1 above), but also the

entire structure of the formula. Specifically it is the calculation of an annual energy

consumption (AE) in kWh that has received criticism, because the actual annual energy

consumption is very much dependent on the individual end user and their behaviour,

whereas the AE value is an average based on a number of simplified assumptions about

user behaviour. While this is not in itself seen as a problem by most stakeholders, it is the

idea of calculating an actual energy consumption which might be far away from what the

consumer experiences in real life that is seen as the problem. Especially when this value

was shown as a number in kWh on the now annulled energy label.

First, it has been suggested to remove the assumptions about user behaviour from the

formula, i.e. the constants for average area (87 m

), number of double strokes (4) and

number of cleaning cycles per year (50). This would also eliminate the problem of

comparing household and commercial vacuum cleaners, which have very different use

patterns. This leaves the specific energy consumption (ASE) and dust pick-up (dpu) in the

formula (as well as standby consumption for cordless and robot vacuum cleaners), which

are also the values measured individually for each vacuum cleaner.

Without the constants in the formula, the expression would not yield and annual energy.

Instead a type of energy efficiency index has been suggested in various versions by

different stakeholders. One of the most useful methods, according to the study team, is to

compare the individual product to an average base case or benchmark, as suggested above

for including the dpu for robots. This concept could be expanded to the specific energy

(Wh/m

) on carpet and hard floor, so that both the energy consumed, and the dust

removed, is compared to a reference value. This could for example be constructed as the

following set of equations, following the same idea as with the current formula:





















































  



   



131

Alternatively, it could be put into one equation with weighting factors for carpet and hard

floor dpu:











 









  









 

However, the latter would have an intrinsic weighting of the two dpus (even without

additional weighting factors), because the range of possible values would be different on

hard floor and carpet. This could be evened out by adjusting the factors that are not 0.5

for both carpet and hard floor to more appropriate values between 0 and 1.

The inclusion of the dpu factors considers the effect on cleaning time through the dpu as

in the current formula, but in a linear manner. For example, if it takes 1 hours to clean a

home with a random vacuum cleaner with dpu=100%, the cleaning time with another

vacuum cleaner with dpu=75% would be  





 for the same room. This gives

the dpu a more “equal” weight on the overall EI score, compared to now, where an increase

from e.g. 75% to 80% has a higher influence that from 100% to 105%, thus increasing

the incentive to reach better dust pick-up. This is because the reduced number of double

strokes are not taken into account (i.e. the factor 0.2) in this calculation.

The advantages of these formulas are that they do not include a large range of constants,

but focus on the measurable performance of vacuum cleaners, making them easier to

understand. Furthermore, the concept of calculating an energy index instead of an actual

consumption does not lead to any false expectations for end-users (as compared to the

current formula which includes multiple assumptions of the use pattern, which might not

fit how the individual user cleans), and it is more in line with other household products,

which also uses EEI (energy efficiency indexes) in many cases.

Another big advantage for policy makers is that it would be very simple to update the

regulation based on technical progress in the market, simply by defining a new base case

value in the equations. The base case could either be common for all vacuum cleaners, or

be different for each type of vacuum cleaner, e.g. household/commercial, corded/cordless.

If the base cases are different, however, it is not possible to compare between the different

vacuum cleaner types. This could be relevant if different, incomparable test methods are

used to derive the measurement results (e.g. for robots and manually handled vacuum

cleaners). If an energy label is then introduced, the design of the label should differ

significantly to not give the false impression that these products are comparable.

Since test methods are still in the process of being adjusted to an extent where an adequate

base case is difficult to define (especially for the cordless and robot products), it is

132

recommended to look into the possibility of introducing an EI formula in the next revision

rather than in the current review.

9.6 Consumer relevance – consumer survey results

This section focuses on what is consumer relevant by highlighting some of the results from

the 2018 APPLiA consumer survey and by discussing the specific test aspects that have

been mentioned by stakeholders both for existing test standards (mains operated vacuum

cleaners) and for the cordless and robot tests being developed.

There are several initiatives aiming at improving standards with regard to a more consumer

relevant testing. Recently, a new WG 22 Ad-hoc Group Consumer relevant testing was

established at CENELEC TC 59X. The WG (Working Group) have prepared a draft document

titled “Consumer Relevant Product Testing” which is intended to support standard makers

in assessing standards to reflect ‘real-life conditions’ while also being suitable for producing

measurement protocols with the required repeatability and reproducibility necessary to

support Ecodesign and Energy Labelling legislation. Vacuum cleaners are among the

examples mentioned in this draft document.

At association level, APPLIA organised four workshops since 2015 with the aim of analysing

and discussing how current product testing methods could be improved to better reflect

real life use of appliances by consumers. The workshops brought together the major

stakeholders (policy makers, NGOs, consumer organisations, Member States

representatives, market surveillance authorities, laboratory experts, consultants and

industry) to discuss the topic and see practical demonstrations of what product testing is

about. Vacuum cleaners were the topic of two of these workshops. Some of the issues

discussed in the following sections were findings from these workshops. Standard makers

were encouraged - and they agreed - to take the findings of the workshops on board for

their future work.

When discussing consumer relevance, it is a trade-off between mimicking the real life use

situation as closely as possible and not increasing test complexity to such an extent that

tests become too time consuming and uncertainty increases to a level where they cannot

be used for regulatory purposes. This trade-off can also be described with accuracy and

precision: Accuracy is a measure of how close to /far from the consumers’ reality the test

results, while precision is how alike the results are each time you test, i.e. how reproducible

and repeatable the tests are.

While the most relevant measure to consumers is exactly how much energy is consumed

and how much dust is removed from his home with the specific cleaning behaviour, every

user is different and have different conditions for cleaning. Therefore tests and calculations

133

are based on averages that makes products comparable on the parameters that are

considered most important to the users.

Hence, the consumer relevance includes many aspects, such as which performance

parameters are important to end users, how and how much people use their vacuum

cleaners in real life, what are the cleaning conditions (floor types, pets, type of dirt etc.).

For example, for most vacuum cleaners, multiple tools (different brushes) and modes are

available to the end user. In addition the products are getting more and more “intelligent”

in terms of detecting which tools are applied, which type of floor is cleaned and how much

dust is in the receptacle, and then adjusting their settings to those specific conditions.

In order to maintain a high precision of the measurements, it is not practically or

economically possible to take into account all the different modes and tools available for

each different vacuum cleaners, in order to obtain high accuracy. However, consumer

surveys and consumer organisations as well as accumulating experiences from test

laboratories and marketing departments gives good indications of what is important to the

end user and how tests can be improved.

Ranking of important parameters

For consumer relevant legislation, it is important to consider what users value when

choosing products and how they use them and in which conditions. The industry

organisation APPLiA made a large consumer survey for vacuum cleaners in 2018, which

gave some information about use preferences.

One of the results from the APPLiA consumer survey ranked importance of different

parameters for purchasing a new vacuum cleaner, which is shown in Table 52.

Table 52: Percentage of consumers rating parameters important/very important in a purchase

situation

Parameter

Percentage answering “very

important” or “important”

I expect it to last a long time

91%

Its performance

90%

The ease of use

89%

The price

87%

The ease of maintenance

86%

The type /stick, robot, canister etc.)

80%

A good filtration of the dust (allergies)

79%

The time spent cleaning

77%

The noise level

67%

The energy efficiency

67%

Having/not having a bag

66%

134

Parameter

Percentage answering “very

important” or “important”

How technologically advanced it is (new features etc.)

64%

The availability of accessories

64%

Its look and feel

56%

The brand

45%

As seen from the table, consumers expect vacuum cleaners to be long lasting, easy to use

and easy to maintain. Brand and design (look and feel) are less important than good

performance, showing that users are unlikely to change them due to design or fashion, but

rather change them when they break, or performance deteriorates. This is also reflected

by 70% of the respondents in the APPLiA survey who bought a new vacuum cleaner either

because the old was broken or no longer “up to the job”.

Floor types

The APPLiA consumer survey also investigated in detail the use conditions and habits of

users. One result that is important for the regulations is the distribution of different floor

types. In the following only results from the rooms that more than 50% of the respondents

had in their homes, seen in Figure 33, are included. i.e. rooms such as garages, present in

less than 50% of homes, are not included.

Figure 33: Types of rooms that more than 50% of the respondents in the APPLiA survey have

in their homes

Figure 34 shows the distribution of flooring for each of the five room types. As seen from

the graph, the room where people most commonly have a carpet is the bedroom (29%)

followed by the living room (23%). The rest of the listed floor type are considered hard

floor types. It should be noted, however, that even when there is hard floor in a room,

many people have a rug that covers part of the floor and also needs to be vacuumed. In

the bedroom and living room, 59% and 67% of respondents had a rug. In the entrance

hall and bathroom 44% and 41% had a rug, while the fewest (24%) had a rug in the

kitchen. (See also Annex A, 6. Regarding market representative floor types).

135

Figure 34: Flooring types in the five most commonly occurring room types

Regarding the types of dirt that is cleaned with a vacuum cleaner, by far the largest

majority is identified as “general dust that has accumulated”, which is experienced in all

room types by more than 70% of respondents. This is, for most of the rooms, followed by

human hair and pet hair, except for in the kitchen, where more than 70% encounter food

wastage, and the entrance hall where 89% encounter debris and mud from outside.

Figure 35: typical dirt types in the five most commonly occurring room types

Vacuum cleaner settings

Another interesting finding from the APPLiA consumer survey is how people use the

different functions of the vacuum cleaners while they are cleaning. For example, the survey

136

found that around one third (36%) of the respondents changed the nozzle based on the

floor type, while 32% merely changed the nozzle setting.

Regarding the different types of nozzles used, 68% of the respondents use the “universal”

two-step nozzle that can be switched between carpet and hard floor setting. Around one

fourth use the specialised carpet (27%) and hard floor (25%) nozzles. Furthermore 36%

of respondents use the special nozzles for furniture, cars, skirting boards etc.

Many vacuum cleaners today let the user adjust the power setting according to the surface

being cleaned. Figure 36 shows how respondents of the APPLiA consumer survey use the

power adjustment option. This shows that 35% use the manual options, while 8% has a

vacuum cleaner that adjust power automatically. Another third (30%) always keep their

vacuum at full power, which is also the power setting they are tested with in the energy

consumption test. 15% has a vacuum cleaner without power setting option. The remainder

of respondents keep their vacuum cleaner at medium (9%) or low (2%) power settings.

Figure 36: User behaviour regarding power settings, according to APPLiA consumer survey.

9.7 Consumer relevance – testing

Carpet test

For the carpet cleaning tests, three cleaning cycles are performed and the measured carpet

dust pick-up (dpu

) is corrected by the dust pick-up of a reference vacuum cleaner when

137

the carpet was new (dpu

cal

) divided by the reference cleaner dust pick-up at the present

state (dpu

ref









 











The general reproducibility of the carpet test has been put into question by manufacturers

and test labs. The low reproducibility and repeatability are caused by a number of

parameters, such as the embedding of the dust to assess the in-depth dust removal, the

wear of the carpet and the microclimate in the carpet, which can vary significantly.

Therefore 16 labs collaborated on a RR (Round Robin) test, where the same four vacuum

cleaners were tested on the labs’ own carpet, as well as a piece of carpet that was circulated

between the labs

174

. The goal was to derive the expanded uncertainty to be able to quantify

the variations that has been observed for the test method. The results are used to assess

the verification tolerances in the regulation.

Carpet type

The carpet used in the performance testing is a wool Wilton cut pile carpet

175

produced

specifically for the vacuum cleaner test in order to ensure reproducible results. However,

a survey made by carpet manufacturers showed that the most sold carpet types are cut

pile or looped nylon carpets. Therefore, a comparable testing is ongoing to investigate the

difference of performance on wool vs. nylon carpet. In the preparatory stakeholder

meeting, it was noted that in the international ASTM test standard, the vacuum cleaner

performance is tested on four different types of carpets, which makes it difficult for

manufacturers to design product specifically to achieve high performance in the test

176

Some stakeholders therefore claim that the Wilton wool carpet is not consumer relevant

and recommend using for example the ASTM carpets, which are proven to work for testing

purposes (e.g. used in North America).

In order to ensure that the test and measured performance is as relevant as possible for

consumers, it is recommended to change to testing on a more representative carpet type,

as long as it does not add further complications and it can be ensured that the carpet

chosen does not vary considerably in quality from batch to batch. However, the choice of

carpet and investigation of different carpets’ suitability for testing vacuum cleaners

requires a lot of test work and therefore would need to be decided within standardisation

174

See Table 61

175

http://www.brintons.com.au/construction-types/

176

Final stakeholder meeting preparatory study, Jan. 2009: Annex C in the Impact Assessment working document. Page 51.

COMMISSION STAFF WORKING DOCUMENT IMPACT ASSESSMENT (2013) with regard to Ecodesign requirements for vacuum

cleaners and the Energy Labelling of vacuum cleaners. http://ec.europa.eu/smart-

regulation/impact/ia_carried_out/docs/ia_2013/swd_2013_0240_en.pdf

138

group. It should be noted that it is not yet known whether another carpet type alone will

result in better reproducibility and repeatability of the dpu

test results.

It has also been discussed that there might be a difference in the carpets used in household

and commercial settings. The commercial vacuum cleaner manufacturers see the need for

a new test carpet with improved test attributes and which is closer to the real-life condition

of commercial end-users. They have therefore suggested a specific test for commercial

vacuum cleaners, with a carpet type other than the Wilton. The standardisation group

working with household vacuum cleaners are also working on developing a test with a more

market representative carpet, however, such a test is far from being introduced, because

one of the carpets suggested has a low durability and thus changes characteristics after

just a few test runs, and another suggested carpet type needs to be investigated further.

Motion resistance

Another major issue that has been raised by several stakeholders is motion resistance.

Motion resistance arises because the vacuum created in the nozzle makes it stick so tightly

to the carpet that it is difficult or impossible to move. This very high motion resistance

arises because nozzles specially designed for increased dust pick up on carpets are used

for the test. However, it is not realistic that the end-user will vacuum with such high motion

resistance, because it is simply too much of a physical effort to push the nozzle over the

floor. The test rigs used for the performance testing moves the vacuum cleaner and have

a push/pull force of up to 100 Newtons, so this is often not a problem during testing. Using

the specialised carpet nozzles in real life is thus inconvenient at best, and for some models

it might not be possible to move, at least without turning down the suction power.

In any case, the performance measured with specialised carpet nozzles featuring high

vacuum is unlikely to reflect the real-life performance if the user decreases suction power

or chooses to use the universal nozzle instead, which the APPLiA survey showed that most

do as described in section 9.6. However, according to stakeholders it is not enough to

require the test to be performed with the universal nozzle to ensure that the nozzle is

designed for a variety of cleaning tasks and not only optimised for cleaning the specific

type of carpet in used in the test. This is for a number of reasons. For one, the “universal

nozzle” would need to be defined in the regulation, which would likely be based on specific

design/technology (such as a manual switch between hard floor and carpet), forcing all

manufacturers to have such a nozzle for their products. Some vacuum cleaner types, such

as uprights and handsticks, do not have these types of nozzles. Furthermore, many

universal nozzles exist today that have high motion resistance, so this alone would not

ensure that the test is more consumer relevant.

139

Another method to make the test results more consumer relevant could be to set a limit

value for maximum motion resistance during the carpet test. The German product testing

organisation, Stiftung Warentest

177

, considers push/pull forces over 30 N to be

unacceptable for users. For commercial vacuum cleaners, acceptable pull/push forces are

not a performance criterion, but a safety requirement mandatory to be fulfilled to comply

with the Machinery Directive

178

. For the commercial products the maximum allowable force

is 27-30 N, but not measured on the Wilton carpet, which has around double the motion

resistance.

Commercial manufacturers have tested products compliant with the Machinery Directive

to have motion resistance of up to more than 40 N on the Wilton Carpet, being around

double of what is measured on low pile carpets. Examples of the measurements for three

nozzles are given in Table 53.

Table 53: measurement of motion resistance on Wilton carpet vs a low pile carpet, performed

by commercial vacuum cleaner manufacturers.

Cleaner

Resistance Wilton

Resistance low pile

“office” carpet

800W nozzle A

30N

15N

800W nozzle B

34N

17N

800W nozzle C

41N

20N

Others have also made user panel tests to find the maximum acceptable pushing force

(forward motion resistance) for commercial use cleaning more than 4 hours/day. This limit

was found to be around 20 N on a short pile carpet (common office carpet) and then

translated (through testing) to correspond to around 40 N on the Wilton carpet. Based on

these findings the 30N used as a guideline by Stiftung Warentest is a bit too low, while 40

N is more adequate as market entry limit.

Carpet debris test

Regarding the end-user relevance of the carpet performance test, it has been

recommended to add a debris pick-up test, to simulate the removal of larger pieces of dirt

from the carpet. It has been noted by multiple stakeholders that end-users often clean

based on what they can see, hence until the floor is visibly clean, rather than based on

removing the embedded dust in carpets. While the in-depth cleaning is still important, the

debris pick-up test could be added to the carpet cleaning performance in order to nuance

the performance criteria and make the test more consumer relevant. A Debris pick-up test

177

https://www.test.de/Staubsauger-im-Test-1838262-1838266/ (Google translate: Since 2011, the testers have also

measured the dust absorption at a sliding force of 30 Newtons. This is roughly equivalent to the strength that an adult finds

acceptable in dust suction. For this test vacuum cleaners with empty dust bag or container. The testers regulate the suction

power of the vacuum cleaners so far that the nozzles can be pushed with 30 Newtons.)

178

European standard EN 1005-4, harmonized under the MD, gives an evaluation procedure for maximum acceptable forces

140

on carpet is under development both for commercial and household vacuum cleaners that

could be used for measurements.

Another parameter that is often used by consumer organisations when testing vacuum

cleaner performance is a fibre pick-up test on carpet. As described in section 9.6.2, hair is

indeed an often encountered type of dirt, and removing it from carpets is one of the trickier

cleaning tasks. However, since a fibre pick-up test is not being developed at the moment,

there are no reliable data or test results regarding performance and repeatability and

reproducibility. This parameter is therefore not suitable for this revision, but could be

considered for future revisions. However, it should be considered whether fibre pick-up

remains broadly relevant to consumers, if the debris pick-up is included in this revision.

Hard floor test

The hard floor performance test is based on removing a special type of standardised dust

from a 3-mm wide crevice in an otherwise flat, hard floor. As with the carpet test, the hard

floor test is often performed using a nozzle designed to optimise dust pick up from the

crevice. This often means a nozzle with high downwards vacuum and closed around the

sides with little or no openings. This in turn leads to dust pick-up above 100% as the dust

in the crevice outside the nozzle itself is also picked up. In real-life situations, however,

flat parts of the floor need to be cleaned, and not only crevices or grooves in the floor. The

nozzles optimised for the crevice test often push debris over the floor rather than cleaning

as a result of the closed sides, not allowing the dirt to be sucked in. This so-called crevice

test is not very consumer relevant, and has been found to result in test-optimised nozzles

that are not optimal for the types of floors and dirt encountered in real life situations.

A suggestion to make the test more user relevant is to not only test hard floor dust pick-

up with the crevice test, but to add a standardised debris test, where larger types of debris

are removed from a flat hard floor surface. Different materials have been discussed as

representative of the debris found in real life situations: from organic grains such as rice

or lentils, to small Lego bricks and brass nuts. In order to ensure repeatability and

reproducibility of the flat floor debris test, the use of organic types of debris has been

discarded, since it is difficult ensure homogeneity because the grain size, density, shape

etc. Legos and metal nuts, on the other hand, are standardised in terms of size, density

and shape and small M3 brass nuts have been found by commercial vacuum cleaner

manufacturers to provide the best type of debris, while for household vacuum cleaners,

aluminium has been discussed.

As noted by some stakeholders, it should be kept in mind that no matter which type of

debris is chosen, there is a risk to repeat the current problems, that products are

specialised and optimised to do well in the test, i.e. to pick up the specific type of debris

141

chosen. However, this is again a question of the trade-off between high accuracy that

comes close the real life and keeping the test to a simplified situation, to avoid testing

several types of dust/debris/dirt on several types of floor. The task will be to find a

representative type of debris concerning size and destiny. By testing larger pieces of debris

as well as dust, all the sizes of dust/debris in between would indirectly be taken into

account.

An important aspect of the additional debris test on hard floor is that it should be conducted

with the same nozzle and nozzle settings as the crevice test to avoid sub-optimisation for

each for the two parts of the tests and ensure the end-users a nozzle that is useful for the

full range of hard floor types they might encounter.

Specialised nozzles

The current test standards for the carpet and the hard floor dpu tests both result in

specialised nozzles optimised for the specific test conditions in order to obtain good

performance ratings on both parameters. However, the special designs compromise the

practical usability of the nozzles as explained above: the carpet nozzles obtain too high

motion resistance and the hard floor nozzle is shielded to a degree that it pushed debris

around instead of removing it.

The test-optimised design of the nozzles means that they are not useful for the end-users

in real-life situations, because they will often differ significantly from the test set-up.

Hence, the user will not get the performance they think they buy, based on the label

ratings. This is a problem for both the end-users and for the credibility of the previous,

annulled energy label and the manufacturers.

While adding the debris test as a parameter for hard floor cleaning performance and the

fibre pick-up test for carpet cleaning performance will most likely result in nozzles designed

for more varying situations, still 68% of users use the universal nozzle, while only one

fourth use the specialised carpet (27%) and hard floor (25%) nozzles

179

. A suggestion to

require all tests to be performed with the universal nozzle to ensure that the 68% of end-

users using the universal nozzle will actually experience the performance shown in the

label, was criticised by stakeholders. The main arguments against such a requirement was

that not all vacuum cleaners are equipped with what is broadly called a universal nozzle,

which would also need to be defined in the regulation, and there would be a risk to be too

design-specific, removing the manufacturers’ freedom to provide specialised tools for

specific tasks.

179

2018 APPLiA consumer survey

142

Such a test requirement would, however, not prevent manufacturers from also developing

nozzles specialised for specific floor types, but it would prevent putting them in the box

solely to justify a performance rating. At the very least it is crucial that tests performed on

the same floor type (e.g. dust and debris pick-up on hard floor) are both performed with

the same nozzle and nozzle settings, as is also stated in the draft standard.

Commercial vacuum cleaner test

Commercial vacuum cleaners are currently tested using the same test standards as

household vacuum cleaners, however, commercial vacuum cleaner manufacturers argue

that the actual use conditions are different and that the tests should be adjusted in order

to reflect these differences. Commercial vacuum cleaner manufacturers have therefore

suggested a specific commercial vacuum cleaner performance test for debris pick-up on

hard floor. The test is based on picking up M3 brass nuts and washers, laid out in a specific

pattern to avoid strategic design of the nozzle to fit the test. Brass is used to simulate a

“worst case” scenario with heavy debris, since the density is high, thus brass nuts and

washers are more difficult to pick up than any lighter materials. The test is to be performed

with the same nozzle and settings as the crevice hard floor test.

Besides the difference in test methods, it is suggested to introduce a different additional

performance parameter, namely the productivity in terms of area cleaned per time interval

(often m

per hour). According to commercial vacuum cleaner manufacturers such a

productivity parameter better reflects the demands of commercial end-users and is often

requested by them, since the salary for professional cleaning personnel is an important

cost. The equations seen in section 9.2.1 are therefore suggested to replace the annual

energy calculations for commercial vacuum cleaners, specifically. In this way the use

pattern of 50 cleaning cycles of 87 m

vacuum per year is removed from the commercial

calculations, which makes it more relevant for the commercial end-users.

Specific suggestions for commercial vacuum cleaner test

In the proposed standard, the hard floor crevice test is suggested to be backed by a debris

test on flat floor. The debris suggested is M3 nuts and washers

180

, because they are ISO

standardized and readily available for purchase anywhere. The idea with this double test

is to avoid nozzles specialized for the crevice test specifically, but to have one nozzle that

is designed to handle both dust and debris on flat floor and floors with crevices. Therefore,

a crucial condition for the suggested double hard floor test is that each part should be

performed with the same nozzle and nozzle settings in order to better mimic real life.

180

M3 nuts and washers were chosen after almost 1500 tests with seven different debris combinations including paper clips,

rice and lentils, 1x1 round Lego bricks, paper and cotton threads.

143

For the commercial carpet test the most important change suggested is the type of carpet.

The commercial vacuum cleaner manufacturers see the need for a new test carpet with

improved test attributes and which is closer to the real-life condition of commercial end-

users. The type of carpet is suggested to be chosen based on the prevailing type sold in

Europe and tests of several carpets have been and are still being conducted. However, a

better carpet with better attributes has not yet been found. Unless a more suitable carpet

is found, the commercial vacuum cleaner standard will be harmonized in this regard with

the household vacuum cleaner standard.

The tests are performed to ensure repeatability, reproducibility, user relevance as well as

testing efficiency and distinction between vacuum cleaners on the different carpets. This

would bring down the test costs significantly as the current carpet type is quite expensive

(in the range of 350 euros per meter test length) and would also be more representative

of the actual environment in which the vacuum cleaners are used. For the carpet test,

commercial vacuum cleaner manufacturers suggest setting a maximum limit for push/pull

forces, since this is an important factor especially for commercial end-users, who vacuum

many hours per day.

Definition of rated power input

As discussed in paragraph 7.3.2 there are some possible flaws in the use of EN IEC 60335-

2-2 as the harmonised standard for ‘rated power input’. There are several options for

improvement. The first option is to request CENELEC to complete the standard and in

Annex ZZ only refer to the main text –without the note on exceptions on booster setting-

- of the clause 3.1.4 of the standard as a reference for ‘rated power input’. Furthermore,

to fight possible ambiguity as regards the verification tolerances, it is recommended to

include explicitly the verification tolerances for ‘rated power input’ in a reviewed regulation

and no longer leave the definition of that regulated parameter to the standard. Given that

the booster setting option no longer applies and that ‘the average effective power intake’

during the performance test –according to EN 60312:2017—is never higher that the ‘safe’

‘rated power input’ there should be no ambiguity. It stands to reason that the verification

tolerances for the rated power input are lower than those for the energy consumption

(±10%).

The second option is to stop using EN IEC 60335-2-2 as a harmonised standard for

presumption of conformity and instead use the value of ‘the average effective power intake’

during the heaviest performance test

181

according to EN 60312:2017 as the parameter to

be regulated under Ecodesign.

181

Currently this is the carpet cleaning test, but this may change in a future regulation. Furthermore, for ‘hard-floor only’

vacuum cleaners there is no carpet cleaning test and thus the power intake during the ‘hard floor’ cleaning test is the yardstick.

144

The third option is to change the content of the standard EN IEC 60335-2-2 to make it less

ambiguous, but given the time this would take (up to 5 years), this is not a practical

solution.

Cordless and robot vacuum cleaner tests

For cordless and robot vacuum cleaners, other parameters are relevant to the consumers

besides those tested for mains-operated cleaners, Factors related to the battery are

particularly important, e.g. battery run time, charging time, maintenance consumption and

battery life. This is in addition to the performance parameters discussed for mains-operated

vacuum cleaners, e.g. debris pick-up on hard floor and fibre pick-up on carpets.

The standard for cordless vacuum cleaners includes specific measurement methods

relevant for cordless vacuum cleaners including run time while maintaining a reasonable

suction power. Such a test is intended to ensure that the declared run time and suction

power are measured simultaneously and are thus not mutually exclusive in practice. E.g.

the longest possible run time obtainable with a cordless cleaner might be while suction

power is at the lowest setting, while the highest setting suction power will result in lower

run times. In order for the consumer not to be misled, the declared run time should thus

be measured on the same suction power setting as the performance is measured with, in

order to give the consumers a coherent picture of the cordless vacuum cleaners’

capabilities.

For robot vacuum cleaners, the battery performance is also important, but in addition

factors related to autonomous operation are important such as floor coverage (i.e.

navigation system) and obstacle overcome capacity. These factors are handled by setting

up a test room in standard IEC (EN) 62929:2014.

Another important factor for both cordless and robot vacuum cleaners is the energy

consumption in the docking station in terms of maintenance power as discussed previously

in this chapter.

9.8 Testing with part load

The empty vs. part load testing is one of the key debates regarding the performance test

of vacuum cleaners and is highly linked to consumer relevance. In the existing standard,

the vacuum cleaners are tested as new (i.e. out of the box), without adding dust or dirt to

the receptacle prior to the test. This means that the receptacle (bag or otherwise) as well

as filters and crevices and nooks inside the vacuum cleaner are completely clean when

initiating the test.

The main argument against this methodology is that testing vacuum cleaners while empty

does not reflect real-life use conditions very well, as vacuum cleaners are almost never

145

empty in real-life

182

and never completely clean from dust except when they are new.

Some organisations and manufacturers therefore argue that the annual energy

consumption stated on the label is not an accurate representation of real-life

consumption

183

. A measurement method with partly filled receptacle has therefore been

suggested to better reflect real-life usage. However, as noted in the Special Review Study

on durability, half-load testing will increase the uncertainty of the test compared to empty

receptacle testing, thus creating further problems with test reproducibility. In order to

achieve high repeatability and reproducibility, highly trained personnel and special

equipment would be needed, increasing the test cost

184

, which would not only imply

increased cost for manufacturers (and eventually consumers), but it would also make MSAs

less likely to perform tests.

As described in the special review study

185

, the motor durability test

186

that entered into

force with tier II on 1 September 2017, is performed with half full bag/receptacle according

to the regulation. Some industry experts have argued that the half load might actually be

an advantage for universal motors in terms of lifetime, as the extra resistance created by

a loaded receptacle decreases the airflow through the motor and thus increase the number

of revolutions per minute, making it ‘easier’ for the motor to run, because less air has to

be pushed through the system. This will in turn cause the carbon brushes on the motor to

wear more slowly, decreasing the wear of the motor

187

. At the same time, however, less

air will mean less cooling of the motor, which will cause the motor to wear faster. However,

there is no general way to predict how different motors will be affected by the receptacle

load, and testing with half load can either increase or decrease the lifetime.

Dyson vs European Commission

The importance of the discussion of testing with part load was underlined in the Court case

of Dyson vs the European Commission

188

, which was ongoing before and during the review.

NOTE this is just an example why good test standards in general, and part load specifically,

are important, it does not reflect the official opinion of the European Commission or the

study team.

182

TOPten criteria paper

183

Topten criteria paper

184

Special review study on durability tests According to Article 7(2) of Commission Regulation (EU) No 666/2013 with regard to

Ecodesign requirements for vacuum cleaners FINAL REPORT Prepared by VHK for the European Commission 2016. page 16.

http://www.ia-vc-art7.eu/downloads/FINAL%20REPORT%20VC%20Durability%20Test%2020160623.pdf ,

185

Special review study on durability tests According to Article 7(2) of Commission Regulation (EU) No 666/2013 with regard to

Ecodesign requirements for vacuum cleaners FINAL REPORT Prepared by VHK for the European Commission 23 June 2016.

http://www.ia-vc-art7.eu/downloads/FINAL%20REPORT%20VC%20Durability%20Test%2020160623.pdf

186

Harmonised standard: Durability of the hose and operational motor lifetime, EN 60312-1:2013

187

Special review, Annex IV, p 31

188

https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:62013TJ0544(01)

146

In 2013 Dyson sued the European Commission with the claim that the tests used to

establish the energy consumption of vacuum cleaners were flawed as the energy

performance is measured only through tests conducted with an empty receptacle. The

vacuuming performance of a vacuum cleaner with a dust-loaded receptacle and, therefore,

the resulting energy efficiency, may be reduced due to dust accumulation.

On 8 November 2018 the General Court annulled the regulation on the energy labelling of

vacuum cleaners

189

on the grounds that the Commission had exceeded the limits of its

empowerment by basing the energy performance on a test with an empty bag, which was

not close enough to actual use conditions as required by the enabling act. The General

Court found it impossible to annul only the calculation method based on an empty

receptacle, and therefore annulled the whole regulation.

The Commission did not appeal against this judgment, and the annulment took effect as

of 18 January 2019.

Definition of part load

The major problem related to the motor test, and also to the suggested part load energy

performance test, is that the part load has yet to be defined. The lack of a definition means

that the tests are currently run with empty receptacles, which according to TopTen is not

in accordance with the standard

190

, however it has been allowed to test the motor lifetime

with empty receptacle but for an increased number of hours, 550 instead of 500.

The Regulation indicates that the durability test for motors should be run with half-loaded

receptacle. The major problem with “half-load” or other definitions depending on a

percentage load, is the difficulty of defining full load. If the full load of the receptacle is not

known, neither is the 50% or another percentage hereof. The same problem arises when

seeking to define partly loaded as a specific amount of standardised dust per Litre of usable

volume, since the “usable volume” would have to be defined first, and this might not be

the same as “full”.

An obvious choice would be to define full load based on the “bag-full” indicator present on

most vacuum cleaners, typically as a red bar that moves under a transparent plastic cover

as the bag fills, as seen in Figure 37. Bagless cleaners often have a clear bin receptacle

and the indication is typically ‘max’ mark on the side, indicating that when the dirt inside

reach the max mark, it should be emptied.

189

https://curia.europa.eu/jcms/upload/docs/application/pdf/2018-11/cp180168en.pdf

190

TopTen Vacuum cleaners: Recommendations for policy design, October 2017

147

Figure 37: Typical bag-full indicator on bagged vacuum cleaner (left) and bagless vacuum

cleaner (right)

However, using the “full” indicator, poses a number of problems:

• What if the vacuum cleaner has no indicator?

o Not all vacuum cleaners have a bag-full indicator or max filling level

indicator. In that case, another definition needs to be applied.

• Which angle should bagless cleaners be held when the dust-fill level is determined?

o Many vacuum cleaners, especially cylinder, can be in at least two positions.

Switching from one to the other changes how the dust is placed in the bin,

and how much dust is needed to reach the Max mark.

• Should the vacuum cleaner be turned on or off when the dust-fill level is

determined?

o Bagged vacuum full-bag indicators are often only activated when it is turned

o When bagless cleaners are turned on the dust is swirled around, distributing

it in the entire bin and making it impossible to determine whether the max

mark is reached. When it is turned off, the dust might not settle evenly.

In general, using the bag-full indicator for determining full load of a vacuum cleaner is very

uncertain. Even if the above questions were answered, it is not unambiguously clear when

the bag-full indicator is activated, or exactly when the dust reaches the max mark. The

judgement will in any case be up to manufacturers, hence adding a high amount of

uncertainty to the test, potentially decreasing reproducibility of the test results, depending

on the influence of bag filling level on the performance. Furthermore, if this definition was

used in the standard, products designed to optimise test results could be a risk, i.e.

designing the indicator to show “bag full” before it actually is and thereby potentially get a

better performance rating.

Another way to determine when the receptacle is full, is to base it on manufacturers’

declaration. However, this procedure requires that all manufacturers have the same

understanding and use the same definition of full. Is it for instance when the bag has to

be changed or the bagless receptacle needs to be emptied? And is it supposed to be

emptied when physically full, or only at a partly full state? For instance, the max mark on

bagless cleaners is not at the top of the receptacle (see Figure 37), and the bag-full

148

indicator might not activate when the bag is completely full, but a while before. Hence this

approach largely brings the same uncertainties and questions as the bag-full indicator or

max mark definition.

An outdated criterion for measuring when the receptacle is full, is when the vacuum (i.e.

pressure difference) has dropped to 40% of the vacuum measured when the receptacle is

empty. This criterion is based on paper-bags that were previously used in most vacuum

cleaners, and quite fast deteriorated the cleaning performance due to clogging. However,

the far majority of bagged vacuum cleaners today use fleece bags, which are more effective

and can take considerably more fine dust before losing performance. Furthermore, this

criterion does not work for some types of bagless vacuum cleaners that are marketed as

not losing any suction power as it fills

191

, whereas other bagless does

192

Instead of basing the part load definition on a share of the full load, an approach with a

fixed amount of dust can be followed, such as the one used by the German consumer

organisation Stiftung Warentest

193

. In their vacuum cleaner performance tests, they test

the vacuum cleaner performance with empty receptacle, with 200 g and with 400 g

standardised DMT8 dust. If the vacuum cleaner cannot hold all the dust, the loading is

stopped, and the test performed with the amount of dust that can fit into the receptacle.

The main advantage of this approach is that it eliminates the need for defining what is the

full load of each receptacle. It has to be resolved, however, how to handle vacuum cleaners

that cannot hold the specified amount of dust. This might especially be a problem for

battery operated and robot vacuum cleaners, and not so much for mains-operated. It will

also have to be decided whether the amounts shall be 200 g and 400 g, or other values.

The approach will increase test costs, since three tests (with different loads) have to be

performed instead of one. Alternatively, just one of the filling points could be chosen.

Current part load definition

Currently, the difficulty of defining full and part load is handled by using three different

criteria for when the receptacle is full:

1. The bag-full indicator, whether it is mechanic or electronic

2. When air pressure has dropped to 40% of air pressure at empty receptacle

3. Adding 100 g of DMT8 dust for each L of receptacle capacity

Whichever of the criteria is reached first is used as the definition of “full” receptacle for the

specific vacuum cleaner being tested (i.e. if the bag full indicator comes on before the

receptacle has been filled with 100 g dust/L, this will be the “full” criteria used for that

191

https://www.dyson.dk/stovsugere.aspx

192

https://learn.allergyandair.com/bagged-vs-bagless-vacuum-cleaners/

193

Füllungen jeweils 200 Gramm, danach 400 Gramm DMT8-Staub. https://www.test.de/Staubsauger-im-Test-1838262-

1838266/

149

specific cleaner). The problem with this way of defining full, is that it is easy for

manufacturers to misuse the criteria to get better performance ratings. For example, the

bag full indicator could be designed to be triggered when the receptacle is only “almost”

full, to ensure tests are made with less loading than actually intended. The same is the

case for adding dust based on receptacle capacity, since this capacity is declared by the

manufacturer, who could again declare a lower volume than the actual volume of the

receptacle

194

. The problem with the air pressure measurement is a bit different, but still

easily circumvented, for example by designing a product that adjust the motor power to

keep a constant air pressure when the receptacle is loaded.

Hence, all the above criteria entail some loopholes that could easily be utilised by

manufacturers wanting their products to look better than they are. However, this is the

best option that there is for defining part load at the moment.

Part load of bagged vs bagless vacuum cleaners

For most bagged vacuum cleaners it is generally anticipated that a loaded receptacle test

will decrease the performance because the bag itself functions as both the dust receptacle

and the primary filter, and as it fills the flow is restricted and the pressure drops. In practice

the user thus switches from lowest to highest air performance every time the bag is

replaced. This effect can be simulated over a single filling of the dust receptacle in the test

lab. Hence, somewhere in between the empty and the fully loaded receptacle is the average

performance that users experience. While this average differs depending on user behaviour

(how often they change bags), testing with some load would be closer to ‘real life’ than

testing with empty receptacle for bagged vacuum cleaners. However, the influence on the

results for dust pick-up and energy consumption seem to be small.

Another aspect of receptacle loading that has been mentioned by consumer organisations,

and which is especially crucial for bagless products is the effect of repeated loading that is

experienced in real life, which result in dust accumulating in filters. Most bagless vacuum

cleaners today use the “cyclone” technology to remove the majority of the dust from the

airflow inside the vacuum cleaners. The dust ends up in the receptacle and does not lead

to a restriction of the flow and hence no drop in pressure will appear. The share of the dust

that is not removed from the airflow by the cyclone is instead captured by the secondary

filters. The accumulation of dust in these filters over time restricts the airflow and reduces

the performances of the vacuum cleaner. In practice the user thus switches from lowest to

highest air performance every time the filter is cleaned/replaced. In order to simulate this

in a laboratory test, the receptacle would need to be filled repeatedly to simulate use

corresponding to half of the time before users change filters, i.e. halfway to the “filter

194

Results of a Round Robin Test show that measuring the maximum volume entails large uncertainties between labs, i.e. low

reproducibility. See section 9.8.5 on available data for part load testing.

150

change needed” mark or half a years’ use (see Figure 40) in order to measure at the

average point experienced by consumers. Furthermore such a test approach would require

defining when the filter needs to be changed in order to define the halfway point. However,

just as for the bagged products, the average condition actually experienced by end-users

depends largely on the maintenance behaviour.

Hence, a more consumer relevant test could be achieved relatively simple for bagged

vacuum cleaners by testing with partly loaded receptacles, but in order to capture the same

consumer relevance for bagless products, they would need to be tested with partially dust

loaded filters. This would in turn require multiple dust loadings of the receptacle, making

the test substantially more time consuming and thus more expensive. This would lead to

different test methods for the two technologies, which would then not be entirely

comparable.

In other words the performance of a bagged product oscillates from minimum to maximum

every time the bag is replaced, while the performance of a bagless product oscillates from

minimum to maximum each time the filter is changed/cleaned. Hence the overall

performance experienced by the user (over years of use) might be the same on average,

but the frequency of the cycle from maximum to minimum is different and most likely

higher for bagged vacuum cleaners (i.e. bag changed more frequently than filter). The

fairest would thus be to test all vacuum cleaners at their “average” performance state,

whether this point is determined by the loading of receptacle or filters. However,

determining this point and adding dust to simulate this point complicates the test procedure

significantly and increases uncertainty of the results to an extent that it is not practically

possible to determining this “average” point.

Another important factor to consider when choosing how to test vacuum cleaner

performance is that the dust receptacle volume of bagged vacuum cleaners is usually larger

than that of bagless vacuum cleaners. Generally speaking, bagless cleaners have dust

receptacles of around 1/2 to 1/3 the volume of bagged vacuum cleaners of similar size and

weight. Hence loading with a fixed amount of dust will not represent the same level of “full”

for the two types of vacuum cleaners, and could be especially problematic for cordless and

robot vacuum cleaners, due to the even smaller receptacles that these vacuum cleaner

types typically have. On the other hand, loading with a specific share (50%) of “full” would

lead to bagged products generally being loaded with a larger amount (in absolute value)

due to the larger receptacles. Hence, a manufacturer could choose to make the receptacles

smaller to obtain better results, at the cost of the consumer, who would then need to

empty the receptacle more often.

151

Available data for part load testing

In order to determine the consequences of part load testing, it is necessary to determine

the difference in the obtained test results from testing with empty load and the different

part load options, and whether testing with part load changes the results significantly.

According to some stakeholders, the empty receptacle performance test is enough to

compare different models fairly and that that part load testing will not make a difference

in relative ranking of products. Others argue that empty receptacle tests favours bagged

products, while loaded receptacle tests (single load) favours bagless products. It still

remains unclear which effects the different options will have on test results, and whether

it will change which vacuum cleaners can comply with the Ecodesign requirements and if

it will change how they are ranked on the energy label. However, any test approach that

systematically favours one product type (e.g. bagged or bagless) over the other should be

avoided, whether it is the empty receptacle option or any of the part load options.

There is no comprehensive data on how testing with partly loaded receptacle affects the

measurement results for vacuum cleaners, however, fragmented data from different

sources have been found.

Ongoing Round Robin Test

In order to obtain more comprehensive data, a Round Robin Test (RRT)

195

is being carried

out in order to establish the measurement uncertainty, repeatability and reproducibility of

testing with a “partly loaded receptacle”. The first part of the RTT has been finished. The

focus of this part was on volume, namely Maximum Usable Volume (MUV) and conditions

for a loaded receptacle as well as the uncertainty of air data for empty receptacle, partly

loaded receptacle and with a 200g loaded dust receptacle. The second part aiming to

determine reproducibility and expanded uncertainty for performance tests with a partly

loaded receptacle has not yet been finalised. The results of this part of the RRT have to be

taken into account when defining intervals for label classes and tolerances for market

surveillance.

The measurement of MUV is an important parameter for part load testing, since the

maximum volume of the vacuum cleaner needs to be determined in order to fill the

receptacle with DMT8 dust in the range 100 g/L (criteria 3 for full load). The determination

of MUV was made for 3 vacuum cleaners in 6 different labs by filling the vacuum cleaners

with moulding granules. The results showed large variance in when the different labs

perceived the receptacles to be full, i.e. have reached the MUV point. The results are seen

in Table 54. As seen in the table, the vacuum cleaner with the largest variance had an

average measured MUV of 1.7 L with an expanded uncertainty of +/- 0.64 L, i.e. around

195

Seven test labs are involved

152

38%. The uncertainty of this measurement alone gives a good idea of the difficulties of

measuring with part load.

Table 54: Uncertainty of measuring MUV, results from RRT by CENELC TC59X WG6

Calculated parameters

Vac 1

Vac 2

Vac 3

Average MUV, L

1.0

5.2

1.7

Repeatability, standard deviation

0.04

0.22

0.04

Reproducibility, standard deviation

0.10

0.78

0.32

Expanded Uncertainty, L

0.19

1.57

0.64

The other parameter measured in the RRT was the amount of DMT8 dust loaded in the

receptacles of the three vacuum cleaners, according to each of the three criteria mentioned

in section 9.8.3. Here the uncertainties were not calculated, but Table 55 shows the

average amount of dust (in grams) the 6 laboratories added as well as the range of filling

and the range in %. The range is the difference between the largest and the lowest amount

of dust added in thee labs. For example for the bag full indicator (criteria 1), the difference

between the lab that added most and least dust was 150 g, out of an average added 284

g. This indicates are very large uncertainty in measuring dust loading, that is observed for

all three criteria.

Table 55: Results on variation in DMT8 filling according to each of the three “bag full” criteria.

Range indicating largest minus lowest measured value

Conditions

Vac 1

Vac 2

Vac 3

Condition 1, grams of DMT8 dust

(bag full indicator)

Average

284

N.A.

731

Range

150

N.A.

506

Range, %

53%

N.A.

69%

Condition 2, grams of DMT8 dust

(suction power 40%)

Average

569

1.831

213

Range

230

401

374

Range, %

40%

22%

176%

Condition 3, grams of DMT8 dust

(filling 100 g/L)

Average

518

167

Range

225

Range, %

21%

43%

47%

The third parameter measured was the air data uncertainty with empty, half load and 200g

load. The tables below show the average suction power (in watts) for three tested vacuum

cleaners tested at 6 different labs, along with the standard deviation (repeatability and

reproducibility) and expanded uncertainty.

Table 56 to Table 58 below show the suction power data for the three tested vacuum

cleaners at peak air power. Other parameters were measured as well, but not shown here,

e.g. vacuum in box and air flow.

153

The results in the tables show that there is also quite a large uncertainty in the air data

measurements, which might not so much be due to uncertainty of the test method itself,

but rather carried over from the uncertainties of the loading and MUV procedures. There

is, however, no final conclusion of this yet.

Table 56: suction power uncertainty for vacuum cleaner no. 1 (bagless, upright vacuum

cleaner)

Vacuum cleaner 1

Empty

½ load

200 g load

Average watts

125.0

124.1

120.5

Repeatability

1.73

1.39

2.27

Reproducibility

7.11

10.5

9.77

Expanded Uncertainty (+/-)

14.22

21.00

19.54

Table 57: suction power uncertainty for vacuum cleaner no. 2 (bagged, cylinder/barrel with

large bag)

Vacuum cleaner 2

Empty

½ load

200 g load

Average watts

208.1

192.9

190.7

Repeatability

0.82

5.81

4.12

Reproducibility

8.85

10.50

16.07

Expanded Uncertainty (+/-)

17.69

21.00

32.15

Table 58: suction power uncertainty for vacuum cleaner no. 3 (bagged, cylinder with small

bag)

Vacuum cleaner 3

Empty

½ load

200 g load

Average watts

212.1

126.7

120.5

Repeatability

3.47

9.41

4.23

Reproducibility

12.71

15.00

17.31

Expanded Uncertainty (+/-)

25.41

30.00

34.62

The suction power data also shows the performance losses with empty, half load and 200g

load. The results in the tables above show that for two of the vacuum cleaners there is

only small changes in the loss of suction power, but for one vacuum cleaner (cylinder

vacuum cleaner no. 3 with small bag) the loss in suction power was 43%. This illustrates

the difference between individual models, but it is not possible to define whether this

uncertainty comes from the MUV measurement or is due to the test method itself, or why

this specific vacuum cleaners is affected by the loading.

Data from Stiftung Warentest on carpet

Test results from the German consumer test organisation Stiftung Warentest (StiWa) were

provided for the standardisation group CENELC TC59X WG6. Please note that data provided

154

here is based on a draft report from the working group, and some members might still

have comments before the final version of the report is published.

The data was based on tests of 27 corded bagged vacuum cleaners and 21 corded bagless

vacuum cleaners as well as 18 cordless bagless vacuum cleaners. The data shows the

difference in dust pick-up and input power for the vacuum cleaners at empty receptacle

and at a load of 200 g and 400 g of DMT8 dust (25 g and 50 g for cordless).

It is important to note that the dust loading might also affect other parameters, such as

dpu on hard floor, dust re-emission and noise, and therefore it does not provide a complete

picture of the effect of dust loading for all parameters. Table 59 and Table 60 show the

effect the loading has on the vacuum cleaners’ performance in terms of dpu

and input

power.

Table 59: Effect on dust pick-up (carpet) at part load (200g/25g) and full load (400g/50g)

compared to empty

Effect on DPU

Partly loaded

Fully loaded

Average

Max

Average

Max

Bagged

-1.5 %-points

-5.5 %-points

-2.5 %-points

-7.5 %-points

Bagless

-1.5 %-points

-8 %-points

-2.0 %-points

-9 %-points

Cordless

-7 %-points

- 25 %-points

Table 60: Effect on input power at part load (200g/25g) and full load (400g/50g) compared

to empty

Effect on

Partly loaded

Fully loaded

Average

Max

Average

Max

Bagged

-6 W

-40 W

-14 W

-50 W

Bagless

-4 W

-33 W

-6W

-39 W

For cordless cleaners there is not data for full load, since too few of the devices could be

filled with 50 g dust to give a result. This is due to the small dust receptacle volume of

these devices, which was on average 0.7L compared to 2 L for the corded bagless devices.

The input power cannot be measured for cordless cleaners because this is measured from

the power socket, and the energy for these cleaners comes from the battery.

Overall it can be observed that the average effect on performance of dust loading for

bagged and bagless is quite similar. However, the data set also reveals that it is not

possible to draw any general conclusions that bagged cleaners respond worse to clogging

that bagless cleaners. Cordless cleaners, on the other hand, do not respond well to dust

loading and has very large decreases in dust pick-up.

155

Hence, there are examples of both bagged and bagless cleaners that show considerable

drop in both dpu

and power input because the machines are clogged thus restricting the

air flow. At the same time there are also both bagged and bagless cleaners that show no

drop in dpu

or input power.

Overall, it was concluded that the two effects ‘lower dpu

’ and ‘lower power input’ almost

cancel each other out in the AE calculation, and the variation in annual energy consumption

on carpet due to dust loading is therefore considerable low, and for the majority of the

products tested, the change was smaller than the interval of the energy label class

corresponding to the now annulled energy label regulation.

It should be noted that the test data from StiWa is solely based on carpet testing and might

deviate from the harmonised standard on some points. The impact of dust loading on

carpet performance is expected to be lower than on hard floor, because the air flow is

already restricted to a certain degree by the carpet itself, when the nozzle ‘stick’ itself to

the carpet due to the under pressure.

A test of a single vacuum cleaner on carpet and hard floor respectively, showed that at an

air flow restriction corresponding to 200 g DMT8 dust loading, the airflow on carpet was

reduced about 1.5%, while on hard floor it was reduced about 4%

196

. Hence, the dust

loading indeed seems to have a larger effect on hard floor performance, but more

comprehensive data is needed to say anything more certain.

Other sources

A German television programme from October 2017

197

addresses the issue of performance

vs. receptacle load and whether the label value would be the same with both test

procedures. The testing was performed by the VDE Testing and Certification Institute in

Offenbach

198

, who loaded the receptacles by 70% (according to own procedure) and

repeated the dust pick-up measurements. The test included only 5 vacuum cleaners, but

indicated that the loaded receptacle had only small influence on the declared values, as

four of the five vacuum cleaners achieved the same performance class, and the last one

just barely missed the declared value. These test results indicted, even though the sample

was limited, that part load testing would not give additional information to the consumer.

Possible options for considering part load testing

It is clear that the consequences of testing with partly loaded receptacle is not easy to

predict and does not affect all vacuum cleaners (even of the same type) equally. While

196

This was based on simulations and by measuring air flow with a clamp attached to the hose of the vacuum cleaner

restricting the airflow to an extent corresponding to 200 g dust loading.

197

https://www1.wdr.de/mediathek/video/sendungen/der-haushaltscheck/video-sauber-ohne-aufwand--wie-gut-sind-smarte-

helfer-im-haushalt-100.html (link to television programme, in German)

198

https://www.vde.com/tic-en

156

there might be a difference between bagged and bagless vacuum cleaners’ performance

with partly loaded receptacles, the differences between individual products is much larger.

However, based on the data shown above and due to the ruling by the General Court on 8

November 2018

199

, testing with part load needs to be considered for vacuum cleaners.

To summarize, the following four options have been identified for how to proceed regarding

part load:

1. Status quo: Continue to test all products as new with empty receptacle.

2. Part load test option: Perform measurements with loaded receptacle: use the three

loading criteria from the standard EN 60312:2013 (bag full indicator/40% decrease

in suction power/100g/L) and measure with whichever is reached first.

3. Simulated part load testing: measure the drop in air flow for the specific vacuum

cleaner with the decided “filling principle” (from part load option 1-3), then using a

clamp on the hose to simulate the air flow restriction during performance testing.

4. Simulated part load calculation: calculate a factor for air flow restrictions by

measuring with empty and with loaded receptacle, then using this factor in the

calculation of dust pick-up to correct for the dust loading effect.

5. A combination of the above: perform some of the tests with part load, other with

simulated calculations.

As stated above, keeping the status quo would entail continued testing with empty

receptacle, but since this has been ruled unsuitable for an energy label, this would be

applicable only to an Ecodesign Regulation. Hence, in case of introduction of a new energy

label regulation, one of the other options must be followed.

The second option entails part load testing, meaning that the receptacle for the vacuum

cleaner is filled with DMT8 dust and then all performance requirements are tested as now,

but with the partly loaded receptacle. While this might seem simple to do, a procedure like

this is expected to increase test uncertainty greatly, to the extent where it would no longer

be possible to differentiate the products into different classes. The reason that the

reproducibility and repeatability is reduced drastically is that the way the dust settles inside

the receptacle can have a big influence on how the vacuum cleaner performs and this can

be changed by simple movements such as shaking or putting down the vacuum cleaner on

the floor. The rate at which the dust is loaded (i.e. vacuumed) into the vacuum cleaner

also affects how it settles. As shown by the preliminary air data from the ongoing RRT

there are very large uncertainties related to just loading the vacuum cleaner similarly

199

https://curia.europa.eu/jcms/upload/docs/application/pdf/2018-11/cp180168en.pdf

157

across laboratories. This method is therefore not foreseen to be possible within at least a

few years.

The simulated part load testing is a way of testing the vacuum cleaner with a simulated

part load, without the uncertainties of how the dust settles. It entails measuring the air

flow (and air pressure, suction power etc.) of the vacuum cleaner empty and loaded and

then adding a clamp around the vacuum cleaner hose to simulate the drop in air flow

caused by the loading and then measuring all the performance parameter with the

restricted airflow (i.e. with the clamp on). This would eliminate the uncertainties of the

dust settling inside the receptacle, but not of the degree of loading (i.e. determination of

MUV), which in itself causes great uncertainty. Furthermore, it requires that the vacuum

cleaner has a soft hose that can be closed partly by mounting a clamp (or similar) on it.

The fourth option is to simulate the part load through calculations. This entails deriving a

part load factor for each individual product from the airflow with empty loaded receptacle.

The air flow should be measured in the BEP (Best Efficiency Point)

200

for each machine with

empty receptacle (Q

empty

) and with partly loaded receptacle (Q

part load

) and the following

Part Load Factor (PLF) could then be derived:

 









The limit value ≤1 should be assigned to this factor (values above 1 could theoretically be

derived due to the uncertainties in the test methods). The PFL should then be multiplied

with the dust pick-up performance for both carpet and hard floor, before the AE value is

calculated

201

. This would result in lower (or equal) dust pick-up values than measured with

empty receptacle. The air flow is, however, only an approximation to the effect of part load

on dpu and in reality, the effect might be different for dpu

and dpu

While this option works for dust pick-up and for energy consumption, which can be

correlated to the air flow, it does not give the results with part load for dust re-emission

and noise.

The final option is thus to use a combination of the above methods to derive the most

accurate and precise results. The above PLF could be used for correcting the dust pick-up

(measured with empty receptacle) for the effect of loading. The energy consumption could

be measured with the clamp (according to option 3 above), while dust re-emission and

200

The existing test standard measures the airflow (Q) vs the vacuum/pressure (P) from fully open to fully closed during 10-15

measurements, to derive the Q vs P curve. The BEP can then be derived from this curve. This test should be repeated with

partly loaded receptacle, and the part load BEP should likewise be derived. The PLF should be calculated from the air flows in

BEP (which would also be the point where the difference between the two curves is the largest)

201

The air flow factor should be multiplied only to the dpu values, not the AE value itself

158

noise could be measured with the actual, part loaded receptacle in accordance with option

2 above.

Conclusion

Ultimately, when choosing a method for part load testing, a careful balance must be found

between the simulation of real life conditions on the one hand and cost/complexity and

uncertainty on the other. Seen from a technical point of view, either the uncertainties need

to be decreased drastically (if even possible) for the test with part load (as in option 2), or

other approximations (as in option 3-5) must be made, in order to have a test that can be

used for regulatory purposes. At the moment, not test data is available on noise, dpu

hf,

dust re-emission with part load, and it is therefore not possible to point to the best method

for approximation for each of these parameters. At the same time the actual test with part

load is not possible with the current uncertainties. The conclusion is therefore that more

(or just some) tests need to be made for these parameters before deciding upon the

methodology.

9.9 Verification tolerances

The verification tolerances stated in the regulations are to be used by market surveillance

authorities when testing products to account for uncertainties in the tests and variations

in production. The verification tolerances are closely related to the tests and the

uncertainties of them, and the standardisation group for household vacuum cleaners (CLC

TC59X WG6) has performed Round Robin tests to determine the uncertainty of the test

methods. These are shown in Table 61 for each parameter together with the label class

width and verification tolerance set out in the regulations. The expanded uncertainties

describe the uncertainties of the measuring methods alone, without the variance of the

products and are expressed as ± values. The measurements were conducted in accordance

with the current harmonised standard EN 60312-1:2017, i.e. without debris pick-up and

with the Wilton carpet and crevice test.

Please note that these results and the analysis in regard to label classes is based on the

existing test standards, in order to give some context to what the sizes of the uncertainties

are.

159

Table 61: Verification tolerances set out in the regulations and preliminary indication of

expanded uncertainties

202

Test parameter

Verification

tolerance

Label class width

Expanded uncertainty

(preliminary)

Annual energy

consumption, kWh/year

10% of declared

value

6 kWh/year

Up to ± 3.5%*

Dust pick-up on carpet,

dpu

0.03 (3 percentage

points)

0.04 (4 percentage

points)

Up to ± 0.057

(5.7 percentage points)

Dust pick-up on hard floor,

dpu

0.03 (3 percentage

points)

0.03 (3 percentage

points)

Up to ± 0.023

(2.3 percentage points)

Dust re-emission, %

15% of declared

value

Variable intervals of

0.06% to 0.40%

Up to ± 0.0012

(0.12 percentage points)

Sound power level, dB

No classes

No measurements

Operational motor life

time, Hours

No classes

No measurements

*Expanded uncertainty measured for average power, which is equivalent to AE

The measured expanded uncertainty shows, as indicated in the sections above, that

especially the dust pick-up on carpet is subject to large uncertainties, and the 0.03

tolerances as well as the 0.04 label class width is according to multiple stakeholders not

appropriate for the current test standard as it is. According to some test laboratories a

difference of up to 3 carpet dust pick-up classes has been found for the same vacuum

cleaner in the same laboratory, which is also shown by the expanded uncertainty.

The standardisations groups are currently looking into other carpet types to increase

representativeness of the tests, but it is not guaranteed that lower uncertainties can be

achieved by changing to another (lower pile) type of carpet. Furthermore, finding a carpet

that is both representative and durable enough to not change properties of the course of

many test runs requires a lot of test work, and according to the standardisation group a

new carpet type is far from being introduced.

In general, it is recommended that actual uncertainties of the test methods are taken into

account when setting the verification tolerances. For the carpet test, this means that the

current tolerance and label class width is not appropriate with the current test standard,

as the uncertainty (+/-) is higher than the label class width. And this is without taking into

consideration the variance between products.

One stakeholder recommends removing the carpet cleaning performance entirely from

both regulations, however seeing that performance is a relevant parameter for

202

Source for uncertainty data: standardisation group CLC TC59X WG6 measurements in RRT including 10 laboratories.

160

consumers

203

, less drastic action could be taken to still give consumers an indication of

carpet performance. For example, the number of classes could be reduced to 4 instead of

7 (as is possible with the new Energy Labelling Framework Regulation

204

) to increase the

class width

205

Such a solution could also be relevant for the other performance parameters (hard floor

dpu and dust re-emission). Even though the measurement method has better repeatability,

it is questionable whether the label class width may be smaller than the range of expanded

uncertainty, which is a problem. Also, the dust re-emission needs to be addressed, since

the smallest intervals are smaller than the expanded uncertainty. The standardisation

group proposes changing the dust re-emission scale entirely to a logarithmic scale rather

than a linear one, similar to the logarithmic scale for HEPA filter declarations.

Only the method for average power (measuring of ASE, i.e. equivalent to annual energy

consumption) has an expanded uncertainty well within the tolerance and the label classes

and a decrease in the tolerance could even be argued.

New test procedures and uncertainties

In relation to the above it should be noted that it is not yet clear what the uncertainties of

the potential new test parameters are (debris pick-up tests and part load testing) and

inclusion of any further test parameters and measurement methods would require further

testing to determine the uncertainty as well as the repeatability and reproducibility and

setting the verification tolerances. The same is the case for introducing more market

representative floors in the test standards, for example a new carpet type.

The number of classes and suggestions above are thus related to the now annulled energy

label, and not relevant for any possible new energy label, since the test methods must

change (at least regarding part load) if such a method is to be introduced. The RRT to

obtain the uncertainty with part load is still ongoing, and no results are available yet. The

results on air data and MUV (see section 9.8.5) however, indicate that the loading itself

entails a large degree of uncertainty. Any new or updated test methods would have to be

assessed against the thresholds suggested, since the uncertainties will change when the

test method is changed.

203

91% of respondents considered performance (as a whole) to be important/very important in the 2018 APPLiA consumer

survey.

204

Regulation (EU) 2017/1369 setting a Framework for Energy Labelling

205

According to Article 11, point 11 this is possible under certain circumstances

161

9.10 Local infra-structure

Electricity

The power sector is in a transition state moving from fossil fuels to renewable energy. The

origin of the electricity is very important factor to consider both regarding the

environmental impact of using vacuum cleaners and how it may affect the consumer

behaviour. Within the EU there are a number of renewable energy targets for 2020 set out

in the EU's renewable energy directive

206

. The overall target within the EU is 20% final

energy consumption from renewable sources. To achieve this goal the different EU

countries has committed to set their own individual goal ranging from 10% in Malta to 49%

in Sweden. In 2015 the share of renewable energy was almost 17%

207

The electricity consumption is a major part of the final energy consumption and the

electricity mix is highly relevant for vacuum cleaners. The electricity mix in EU in 2015 is

shown in Figure 38. Almost half of the electricity consumption still originated from

combustible fuels and renewable energy sources only constituted about 25 % of the

electricity generation in 2015.

Figure 38: Net electricity generation, EU-28, 2015 (% of total, based on GWh)

The reliability of the electricity grid could be in some degree affected by the transition to a

renewable energy system. With more renewable energy in the system new challenges

occur e.g. with excess production of wind energy and the bi-directional transfer of energy.

Due to technological development, the reliability in many EU countries is ensured by the

expansion of the electricity grid (transmissions lines across Europe) to distribute renewable

energy. The quality of the electricity grid in Europe is considered to be high and among the

best in the world. Every year the World Economic Forum release a Global Energy

206

https://ec.europa.eu/energy/en/topics/renewable-energy

207

http://ec.europa.eu/eurostat/documents/2995521/7905983/8-14032017-BP-EN.pdf/af8b4671-fb2a-477b-b7cf-

d9a28cb8beea

162

Architecture Performance Index report. The report is ranking the different countries on

their ability to deliver secure, affordable, sustainable energy. In recent years European

countries have dominated the top spots

208

. The 10 highest scoring countries are presented

in Table 62.

Table 62: Global Energy Architecture Performance Index report – best performing countries

Country

2017

score

Economic growth and

development

Environmental

sustainability

Energy access and

security

Switzerland

0.8

0.74

0.77

0.88

Norway

0.79

0.67

0.75

0.95

Sweden

0.78

0.63

0.8

0.9

Denmark

0.77

0.69

0.71

0.91

France

0.77

0.62

0.81

0.88

Austria

0.76

0.67

0.74

0.88

Spain

0.75

0.65

0.73

0.87

Colombia

0.75

0.73

0.68

0.83

New Zealand

0.75

0.59

0.75

0.9

Uruguay

0.74

0.69

0.71

0.82

Consumer behaviour regarding vacuum cleaners is only assumed to have a limited effect

on the electricity system as people use their vacuum cleaners at the same rate throughout

the year at different times. Robotic vacuum cleaners and vacuum cleaners with batteries

can in theory add some flexibility to the electricity system as they can be charged whenever

there is an excess of renewable energy in the system or the energy consumption is low.

The hourly load values for a random Wednesday in March 2015 for selected countries are

presented in Figure 39.

208

https://www.weforum.org/reports/global-energy-architecture-performance-index-report-2017

163

Figure 39: Hourly load values a random day in March

209

All the four countries represented in the graph have similar hourly load values with two

peaks, one in the morning and one in the evening, even though it is barely visible for

Denmark due to the scale of the graph. There are small differences in the timing of the

peaks, but the first peak fits well with the start of the workday and the second peak fits

with the end of the workday. Between the two peaks there is a falling trend in the energy

consumption. The lowest electricity consumption across the different countries are at 5

AM. For most countries, this hourly load curve fits this description the majority of the days.

For months and days with a higher or lower consumption, the profile is the same but shifted

up or down.

Products that can respond to an external stimulus (smart appliances) can provide balance

and flexibility to the energy system, but the impact of vacuum clearness is currently

assumed being low. In the future, vacuum cleaners with batteries, and especially robotic

vacuum cleaners, which can have flexible cleaning times, can be charged during the night

when the energy consumption is low. The potential depends on the future stock and energy

consumption of battery driven vacuum cleaners.

9.11 Use of auxiliary products

During the use phase many vacuum cleaners use auxiliary products in the form of bags

(only in bagged vacuums) and filters (all types). Changing the bag and filters regularly is

important for continued optimal operation of the vacuum cleaner, since excess amounts of

dust and particles can otherwise clog the vacuum cleaner, blocking the air flow.

It has not been possible to find any cordless or robot vacuum cleaners using bags, and it

is thus assumed that bags are only used in bagged mains-operated vacuum cleaners.

209

Data provided by ENTSO-E

164

Previously vacuum cleaner bags were often made of paper, but today the far majority is

so-called fleece bags made of poly-propylene material.

Most bagged vacuum cleaners are equipped with an indicator, showing the user when the

bag should be changed, however the frequency is highly dependent on the type and

amount of dirt that is collected as well as user preferences. In the preparatory study, an

amount of 10 bags per year was proposed based on the number of bags offered by

manufacturers in free bag schemes

210

, however also 6 and 5 bags per year have been

suggested

211

, and in this study 6 bags per year on average is therefore used.

According to the APPLiA consumer survey results (shown in Figure 40), 46% of respondents

with a bagged vacuum cleaner empties the bag only when it is completely full, while 24%

change it when the bag full indicator shows it is necessary, and 13% changes it after each

time either before or after vacuuming. 16% change the bag only when they can fell that

the vacuum cleaner loses suction power, which can, however, also have to do with the

need for changing the filter.

Figure 40: Consumer habits regarding changing bags and filter of their main vacuum cleaner,

according to the APPLiA consumer survey

210

Preparatory Studies for Eco-Design Requirements of EuPs (II), Lot 17 Vacuum cleaners, TREN/D3/390-2006, Final Report

February 2009, carried out by AEA Energy & Environment, Intertek, and Consumer Research Associates between November

2007 and January 2009. https://www.eceee.org/static/media/uploads/site-2/ecodesign/products/vacuum-cleaners/vacuum-

cleaners-ecodesign-study-final-report-eup-lot-17-final-report.pdf page 32, according to Electrolux website accessed 1 May 2008

http://www.electrolux.co.uk/Files/United_Kingdom_English/Files/Electrolux07_SpecBrochure_8pp.pdf

Miele UK website accessed 1 May 2008

http://www.miele.co.uk/Resources/CustomerSupport/GuaranteesWarranties/Vacuum_Guarantee.pdf

211

Abele et al. (2005) and Kemna et al. (2005) According to JRC report,

165

Filters in vacuum cleaners are used to prevent dust and particles reaching the motor and

returning to the room through the exhaust air. Some vacuum cleaners might have both a

primary and a secondary filter, but today fleece bags often function as filter as well,

rendering the secondary filter redundant. Filters can be made from different materials such

as cloth, foam, pleated paper, and fleece or other synthetic materials. Some vacuum

cleaners are fitted with HEPA filters (High Efficiency Particulate Air), which let only through

5 (HEPA 14) to 50 (HEPA 13) particles per litre of air

212

, for particles sizes down to 0.3

microns. HEPA filters are especially relevant for people suffering from asthma or allergies,

as they remove the allergens and particulates that triggers these conditions. Of course,

filters are only efficient if the vacuum cleaner is air-tight, not letting air out from the

appliance before the airflow reaches the filter. As in the preparatory study, it is assumed

that filters are replaced once a year. However, some models come with washable filters,

which are assumed not to be replaced unless they are damaged. In that case it would count

as a repair, and not as maintenance.

9.12 Repair practice

Repair is an important factor for increasing the product lifetime and depending on the type

of repair, it can be done by either end-users or professionals. Repairs such as exchanging

a hose or suction head can be done by the end-users, while problems with e.g. the motor

or electrical components is done by professionals for safety reasons

213

. If the repair is done

by professionals, the repair cost is dependent on the labour cost, which varies greatly

across Europe as seen in Figure 41. Based on labour costs alone, the amount of repair by

professionals is expected to be low in northern countries and higher in southern and

eastern countries. Another important factor for whether the end-users chooses to repair

the vacuum cleaner is its age. In the end of the lifetime, it might be perceived as too

expensive to repair compared to the cost and ease of buying a new model.

212

https://consumer.nilfisk.dk/da/cases/About%20Vacuum%20Cleaners/Pages/Nilfisk-stovsuger-filtrering.aspx and

https://www.whiteaway.com/hverdagen/post/derfor-skal-du-overveje-en-stovsuger-med-hepa-filter/

213

http://publications.jrc.ec.europa.eu/repository/bitstream/JRC96942/lb-na-27512-en-n_.pdf

166

Figure 41: Hourly labour cost in €, 2016 for European countries

The most common failures of both upright vacuum cleaners and cylinder vacuum cleaners

are related to suction and blocked filters as shown in Table 63

214

. These problems can be

interconnected and are also related to the lack of maintenance (such as changing bags and

filters), and might in some cases be possible to solve by repairing or exchanging faulty

parts. At some point the motor is also likely to fail, since universal motors are used in many

vacuum cleaners

215

. However, most motors are likely to function for at least 600 hours

regardless of the purchase price of the vacuum cleaner

216

, and at least 500 hours is

required by the current Ecodesign Regulation.

Table 63: Faults experienced with upright vacuum cleaners and cylinder vacuum cleaners

217

Upright vacuum cleaners, Faults

experienced

Cylinder vacuum cleaners,

Faults experienced

Suction deteriorated

24.3%

Suction deteriorated

19.5%

Blocked filters

21.7%

Blocked filters

17.8%

Belt broken (drive-belt rotating brush)

16.9%

Other

15.7%

Split hose

13.7%

Broken accessories

12.2%

Motor broken

13.4%

Brush not working properly

10.8%

Brush not working properly

12.0%

Casing cracked/chipped/broken

10.1%

No suction

10.0%

Overheating

8.7%

Brush not working at all

9.4%

Split hose

7.7%

Casing cracked/chipped/broken

8.9%

Motor broken

6.6%

Other

8.6%

Power cutting out

5.2%

Broken accessories

8.3%

Power cable faulty

5.2%

Overheating

6.3%

No suction

5.2%

Power cable faulty

5.1%

Brush not working at all

4.9%

214

http://publications.jrc.ec.europa.eu/repository/bitstream/JRC96942/lb-na-27512-en-n_.pdf

215

Special review study on durability tests According to Article 7(2) of Commission Regulation (EU) No 666/2013 with regard to

Ecodesign requirements for vacuum cleaners FINAL REPORT Prepared by VHK for the European Commission 23 June 2016.

http://www.ia-vc-art7.eu/downloads/FINAL%20REPORT%20VC%20Durability%20Test%2020160623.pdf

216

http://publications.jrc.ec.europa.eu/repository/bitstream/JRC96942/lb-na-27512-en-n_.pdf

217

https://www.vhk.nl/downloads/Reports/2016/VHK%20546%20FINAL%20REPORT%20VC%20Durability%20Test%20201606

23.pdf

167

Wheels/castors broken

4.9%

Handle broken

3.8%

Handle broken

4.6%

Power not working at all

3.8%

Power not working at all

3.7%

Controls broken

2.4%

Power cutting out

3.1%

Wheels/castors broken

2.4%

Handle loose

2.3%

Belt broken (drive-belt rotating

brush)

2.1%

Controls broken

0.60%

Handle loose

1.7%

For robot vacuum cleaners, less data is available as the technology is both new and in a

transition state with frequent improvements. Based on troubleshooting guides available on

the internet possible problems with robotic vacuum cleaners are related to:

• The belts and drive systems can break or be worn so the performance of the vacuum

cleaner is reduced. These parts can often be replaced

• The battery performance can be reduced

• The electronics and advanced controls can be faulty after a period of time as data

interrupting the function can be stored on the memory board. Sometimes a reset

can fix this problem

• The motor can be faulty or damaged and has to be replaced.

To avoid break downs, it is important to have proper maintenance of the vacuum cleaner

and simple maintenance instructions are often provided in the user manual. In some cases,

the user guide is also available online with additional drawings and exploded views

218

Spare parts are widely available on the internet from third party dealers

219

and the

manufacturers

220

. However, a stakeholder has mentioned that even though spare parts

may seem available on the internet, it may not always be the case for independent repair

centres, or it is not always possible for the consumer to receive the actual spare parts

within a reasonable time and cost.

A manufacturer

221

has stated that critical spare parts (parts important for the vacuum

cleaner to function) are available as long as 10 years after the last product is purchased

and minimum 10 years after the production of the last product. This is not considered to

be the standard availability of spare parts from manufacturers, as other manufacturers

have different spare part policies.

It is not known how often repair actions are carried out or which types of repair are

conducted. Consequently, it is not possible to estimate additional material for repair. The

standard value in the EcoReport tool of 1 % of the materials are used for the amount of

218

https://www.dysonspares.ie/index.php?route=information/information&information_id=70

219

https://www.partswarehouse.com/default.asp

220

https://consumer.nilfisk.com/en/products/Pages/product.aspx?fid=16718

221

BSH Hausgeräte GmbH

168

spare parts. 1 % of the materials corresponds to 70 grams for a vacuum cleaner of 7 kilos

which seems reasonable as not all consumers are expected to buy spare parts.

9.13 End of life behaviour

The material consumption and resource impact from products is closely related to the end

of life processing. Vacuum cleaners are collected at end of life and send to a facility for

reprocessing. Illegal trade and sales of scrap challenge the collection rate for some product

categories. The statistics from Eurostat shows products put on the market and waste

collected for small household appliances. This statistic does not refine the actual number

of vacuum cleaners collected so the actual collection rate can be difficult to quantify.

From 2019 onwards, the minimum collection rate to be achieved annually shall be 65% of

the average weight of EEE (Electric and Electronic Equipment) placed on the market in the

three preceding years in the Member State concerned, or alternatively 85% of WEEE

generated on the territory of that Member State

222

. In Annex D the collection rate is

calculated for small household appliances based on the average weight of EEE placed on

the market in the three preceding years in the Member State concerned

223

. The calculated

average collection rate for the EU was below 40% in 2014. The collection rate does also

cover other appliances, but it is assumed that the rates are representative for vacuum

cleaners. The collection rate should be improved to 65% in 2019. The low collection rate

of vacuum cleaners cannot be addressed in the Ecodesign Regulation but should be

addressed by each EU country which should decide how to fulfil their obligation regarding

the WEEE Directive.

Estimated second-hand use

The estimated second-hand market is based on a survey on Ebay and other similar

homepages. Overall vacuum cleaners are available on the second-hand market as used

consumer products and as refurbished products

224

. Refurbished products are described in

the medical device regulation as

225

: ‘fully refurbishing’, for the purposes of the definition

of manufacturer, means the complete rebuilding of a device already placed on the market

or put into service, or the making of a new device from used devices, to bring it into

conformity with this Regulation, combined with the assignment of a new lifetime to the

refurbished device.

Vacuum cleaners are not expected to be fully refurbished as described in the medical device

regulation, but only partly, so the vacuum cleaners are repaired, reconditioned and tested

222

http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=env_waselee&lang=en

223

http://appsso.eurostat.ec.europa.eu/nui/show.do?dataset=env_waselee&lang=en

224

http://www.ebay.co.uk/bhp/manufacturer-refurbished-vacuum

225

http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=OJ:L:2017:117:FULL&from=EN

169

before they are sold again by the manufacturers or specialised repair shops. The market

of refurbished consumer vacuum cleaners is limited and have no impact on the later tasks.

The regular second-hand market, where consumers sell their old appliances to other

consumers, is considerably larger and consists of a large variety of products from almost

new products to products that have been in operation for many years, and premium

products to low budget products.

On the internet buying guides are also available

226

, pinpointing pros and cons of second-

hand vacuum cleaners. Though the market exists, the impact of the second-hand market

is expected to be limited as the functional operation of vacuum cleaners are expected to

be unchanged. Therefore, the second-hand market is not included in later tasks.

Recyclability of vacuum cleaners

After collection the electronic scrap is treated at specialised facilities which mechanically

process the appliances. The expected waste process flow for vacuum cleaners are

visualised in Figure 42. Note that vacuum cleaners are mixed and shredded with other

types of products at end of life, and the following only relates to the handling of vacuum

cleaners.

Figure 42: Expected reprocessing of vacuum cleaners at End of life

The pre-processing is the first step in the recycling process of vacuum cleaners. This first

step often consists of manual removing of targeted components and/or materials for

further treatment. The pre-processing is very important in connection with an effective

recycling process by reducing the risk of contamination, quickly recover selected valuable

materials for further reprocessing and allow compliances with current directive on

hazardous substances

227

and waste

228

and prevent damage to the facility in the following

steps. According to the WEEE Directive components such as electronic components (e.g.

226

https://learn.allergyandair.com/buying-a-used-vacuum-cleaner/

227

http://ec.europa.eu/environment/waste/rohs_eee/index_en.htm

228

http://ec.europa.eu/environment/waste/weee/index_en.htm

170

printed circuit boards, capacitors, switches, thermostats, liquid crystal displays) and

batteries are additionally dismantled when present (see section below).

Next is a series of shredders, which reduces the vacuum cleaners to smaller pieces, so the

different materials can be sorted. The dust is removed and captured by cyclones. When

the equipment is shredded into smaller pieces (approximately 1 cm to 10 cm) different

technologies handles the sorting. These technologies are often

229

• Magnetic separation removing ferrous metals

• Eddy current separators removing non-ferrous metals such as copper, aluminium,

and zinc

• Density separators: Different types of plastic.

The effectiveness or recycling rate of the shredder (the share of recovered, recycled, and

reused materials) in this study is based on the EcoReport tool

230

but updated regarding

plastic

231

. The values used in the current study is presented in Table 64.

Table 64: Re-use, recycling, heat recovery, incineration and landfill rates assumed for the End

of life handling of vacuum cleaners

Fraction to

re-use, (%)

Fraction to

(materials)

recycling, (%)

Fraction to

(heat)

recovery (%)

Fraction to non-

recoverable

incineration,(%)

Fraction to landfill/

missing/ fugitive

(%)

Bulk Plastics,

TecPlastics*

29%

40%

31%

Ferro, Non-ferro,

Coating

94%

Electronics

50%

30%

19%

Misc.

64%

29%

Auxiliaries (Bags)*

50%

25%

*Adjusted values compared to the EcoReport tool

232

With these numbers the total recycling rate (including incineration) will be above 70 % for

products that are shredded. The numbers also show high recycling rates for metals and

lower rates for plastic. Traditionally it is also easier for recycling facilities to recover metals

than plastic. Plastic are often mixed with other types of plastics which challenge the quality

of the recycled plastic. Often recycled plastic is downgraded if it is not properly separated.

229

http://www.sciencedirect.com/science/article/pii/B9780128033630000031

230

http://ec.europa.eu/growth/industry/sustainability/ecodesign_da

231

Plastic Europe, Available at: http://www.plasticseurope.org/documents/document/20161014113313-

plastics_the_facts_2016_final_version.pdf

232

Plastic Europe, Available at: http://www.plasticseurope.org/documents/document/20161014113313-

plastics_the_facts_2016_final_version.pdf.

171

10. Task 4: Technical analysis

Task 4 contains the technical description of key components in vacuum cleaners as well as

descriptions of the different product types (working towards base case definitions)

including average performance and energy consumption levels. Furthermore, it contains a

section about material and resource consumption in different types of vacuum cleaners

including Bills-of-Materials (BOMs) and End of life (EoL).

Combined with the outcomes of task 1-3, task 4 forms the basis for further analyses in the

following tasks, including environmental and economic impacts (task 5) as well as

improvement options (task 6).

10.1 Components

In Task 1 the various vacuum cleaner categories - cylinder, upright, cordless and robot -

were introduced. This section will start with a description of the most popular type in

Europe, the mains-operated cylinder vacuum cleaner, and will then add further information

for the other types. Figure 43 shows the main components in a mains-operated vacuum

cleaner: motor, fan, receptacle, filter, hose and nozzle, which will be discussed hereafter.

Figure 43: Key components in a mains-operated vacuum cleaner

The overall energy flows related to these components are given in the Sankey-diagram in

Figure 44. It relates to a well-designed 750 W cylinder vacuum cleaner as described in the

2007 preparatory study. It uses an agitator (active nozzle) because, according to the

preparatory study

233

, it is the most effective and efficient way to clean carpets. For hard

floor cleaning it is not a necessary feature and a passive nozzle is sufficient

234

233

AEA Energy & Environment, Final Report EuP (II) Lot 17 Vacuum Cleaners, Final Report, February 2009

234

To complete the energy effort, also the manual operation of the product and/or the nozzle should be included. At test-

conditions this means a manual power of 20 N to move the nozzle at a speed of 0.5 m/s. This comes down to human power of

10W to be added.

hose

nozzle

bag

fan

motor

filter

switch

cord & plug

air + dust in

air out

dust

172

Figure 44: Sankey-diagram of energy flows in a mains-operated cylinder vacuum cleaner

(source: VHK 2017 graph on the basis of AEA Ricardo 2009 data)

The suction power of 242 W relates to an empty bag and filter and might drop to 227 W

(6.4%) when the bag is full. The minimum pressure drop should be in the range of 18-25

kPa and flow should be at least 8.5 L/s when the bag is full and probably 15-20 L/s in best

conditions. The 50% efficiency of fan plus motor is very ambitious. Still, even in this

ambitious setting motor and fan losses constitute by far the highest losses (338 W),

corresponding to almost three-quarters of losses. After that, the corrugated primary hose

of a cylinder type (as opposed to the straight tube of other types) cause considerable

aerodynamic friction losses (40 W) as well as the bag and filter (35 W). The motor losses

of the agitator are also significant (25 W).

Motor

In only a few years, the Ecodesign Regulation and the annulled Energy Labelling Regulation

have revolutionised the vacuum cleaner market. European vacuum cleaner suppliers have

switched in their top-models from a motor-type with arguably the worst efficiency (30%)

to a motor with the best efficiency around (80% or more). In these models the so-called

Universal AC/DC motors with carbon brushes was replaced by brushless electronically

commutating (EC) motors, with or without permanent magnets

235

. Motors in the range of

2000 W or more are now replaced with motors in the range of 600-800 W (electric input

power), without any loss in cleaning performance. The technical product life of these

motors, which are also quieter, is at least 5 times better than what was the average before

the regulations.

235

PM stands for Permanent Magnet motor, which also the most common form of Brushless DC Motor (BLDC). SRM stands for

Switched Reluctance Motor, a motor that does not require permanent magnets and thus also not contain Neodymium.

Neodymium is currently on the EU’s Critical Raw Material (CRM) list.

173

EC-motors, like Brushless DC (BLDC) or Switched Reluctance Motors (SRM), are the most

efficient electric motors on the market, comparable to IE4 or IE5 efficiency grades as

defined in the ecodesign electric motor Regulation

236

. In laboratory circumstances,

efficiencies as high as 96% can be reached. In practice, efficiency also depends on the load

and probably the very best BLDCs for vacuum cleaners may achieve 85% over the (variable

torque) operating range.

The technical motor life, mainly determined by the length of the carbon brushes for

universal motors

237

and for universal motors in the order of 600 hours, will for BLDC motors

be 3000-4000 hours or more at empty receptacle

238

. At 50 hours usage per year, which is

currently taken as average vacuum cleaner usage in the regulation, this implies a technical

product life of 60 to 80 years. This is probably at least twice as long as the economic life

of a standard product, i.e. the time where 99% of consumers would discard the product

for another reason (breaking of other vacuum cleaner parts, consumers attracted by new

features, etc.). The increased product-life also changes the perspective on the need for

reparability of the motor. Of course, if robot vacuum-cleaners come into scope that could

possibly vacuum your house e.g. 4 hours per week (100 hours per year), then a longer

motor lifetime would be required for them. Note that robot suction motors are smaller than

the regular VC suction, comparable to blower motors in e.g. large computer fans. They will

be of the BLDC-type.

Last but not least, a positive effect of a more efficient motor, especially a PM motor, is that

it also produces less noise (sound power, expressed in dB(A)) than the universal motor

with its mechanical commutators (carbon brushes).

As was assessed in the 2016 Special Review Study, this comes at a price: A universal

AC/DC vacuum cleaner motor can be found for as little as 4 € per unit. In January 2016, A

BLDC motor with inverter for vacuum cleaner-applications cost around 33 €. Currently,

over 2 years later (Sept. 2017), BLDC prices appear to have been decreased by 20%. Still,

in consumer prices and with the factor 3.77 mark-up

239

, this means that top-range vacuum

cleaners may cost at least 100 € more than with the universal motor

240

Fan

The typical household vacuum cleaner uses a centrifugal fan to create ‘suction power’ (a

negative pressure difference). In principle, as mentioned in the 2007 preparatory vacuum

cleaner study, there are other possibilities to create suction power, including reciprocating

236

OJ L 191, 23.7.2009, p. 26–34, https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=CELEX:32009R0640

237

Other aspects such as overheating, or just poor build quality can also influence the technical motor life

238

Note that 3000h is not a proposal for a minimum lifetime requirement of a standard product, because testing costs and a

fast reaction time for market surveillance authorities also play an important role.

239

Based on difference in manufacturer selling price and consumer purchase price from PRODCOM and GfK

240

Based on costs from Belgium

174

solutions with pistons, scroll-geometry, screws, etc. and including turbo-compressor type

solutions. In a laboratory and using clean air, some of these solutions can even be more

efficient than the current vacuum cleaner centrifugal fan

241

. However, to reach and

maintain these efficiencies in a ‘dusty’ vacuum cleaner environment requires precision

geometry and very narrow tolerances, typically achieved with machined steel parts and

thus at prohibitive prices for a mass-produced consumer product.

Hence, the vacuum cleaner uses a backward curved centrifugal fan, i.e. where the air

enters at the centre in the front and is then spun sideways using centrifugal force. A

centrifugal fan is defined as ‘backward-curved’ if centrifugal blade angle β ≤−1°, ‘radial’ if

−1°< β <1° and ‘forward-curved’ if β ≥1° (see Figure 45).

Of all the fan-types (axial, mixed flow, centrifugal) and sub-types (forward curved, radial,

backward curved, backward inclined), the backward curved (BC) centrifugal fan is the most

efficient for this and many other applications. In the latest draft Ecodesign proposal for

industrial fans, intended to replace Commission Regulation (EU) 327/2011 (‘fan

regulation’) in one or two years, the proposed total efficiency limit for fans with electric

power input Pe<10 kW is η

min

= 0.0456 LN(Pe) – 0.105 + N, where N=0.67 for BC centrifugal

fans in category B and D.

This means e.g. a minimum total efficiency of 55% for a fan with electric power input

Pe=0.7 kW. This efficiency goes up for bigger fans up to 10 kW, which has 67% efficiency.

However, the typical vacuum cleaner fan is no ordinary fan: it operates at a flow rate qv

in the range of 8-40 dm³/s (3-15 m³/h) and a pressure difference dP as high as 10-20 kPa

(10,000-20,000 Pa). For comparison, the flow is 10 times lower and the pressure difference

is 30-50 times more than in a ‘normal’ fan for a residential ventilation unit.

It is referred to as a High Pressure Low Volume (HPLV) ‘fan’ or ‘blower’. The efficiency of

this types is lower than that of a normal centrifugal backwards curved fan, because the

slim design (relatively high diameter D, compared to thickness between front and back-

plate) causes high friction losses and the gas (air) is starting to operate in the compressible

range.

241

This applies to some of the reciprocating solutions. Small turbo-compressors at the operating range of vacuum cleaners are

less efficient than the VC fan.

175

Figure 45: Backwards curved centrifugal fan (left) and fan definitions using the centrifugal

The empirical Cordier diagram in Figure 46 gives a good illustration of the interrelation

between specific speed σ, diameter

δ (compared to a unitary reference fan) and efficiency

η. It indicates that fans with small σ

(<< 0.3) generally have a significantly lower efficiency

than centrifugal fans with σ= 0.3...0.6

242

Figure 46: Cordier diagram (Eurovent/EVIA 2016 citing Eck 2003)

A HPLV-fan is defined in the draft proposal for an Ecodesign Fan Regulation

243

as a fan with

a specific speed σ

bep

<0.12. The specific speed σ

bep

of centrifugal

fans with electrical input

power input P

< 10 kW is defined as:

242

Note that the Cordier diagram is based on empirical tests of fan designs in the 1950s. Although it is a good illustration of the

principle in this case, it is no longer considered 100% state-of-the-art for all aspects.

243

OJ L 90, 6.4.2011, p. 8–21, Commission Regulation (EU) No 327/2011, https://eur-lex.europa.eu/legal-

content/EN/TXT/?uri=CELEX%3A32011R0327

backward-curved

radial

forward-curved

rotation direction

axis

blade

trailing edge

176







 



 



 











where

− σ

bep

is specific speed (dimensionless);

− n is fan speed in revolutions per second (rps);

− ρ is air density 1.2 kg/m³;

− q

v,bep

is volume flow rate at best efficiency point bep, in m³/s;

− p

f,bep

is total fan pressure at bep, in Pa;

− π is the number pi (3.14…)

Figure 47 gives an overview of total fan efficiency, i.e. based on the total pressure

difference, for a centrifugal backwards curved fan as a function of the specific speed σ

bep

(‘sigma’ in the figure) for industrial fans on the market in 2016.

Interpretation of this diagram requires caution. The best efficiency values of ~82%, at

0.2<σ

bep

<0.45, apply to large fans probably in the range of 10 kW or more. As mentioned

before, and is clear also from proposed limit values, the best efficiency values for the

current vacuum cleaner fans (0.7 kW) are some 12%-points lower, i.e. at around 70%.

Likewise, for efficiencies at σ

bep

<0.12 one can assume efficiencies 10-12% lower, e.g. 52%

instead of 60% at σ

bep

=0.1.

Based on this, a 0.7 kW vacuum cleaner fan, a reference line has been drawn in the

diagram delivered by Eurovent/EVIA. It shows that the best vacuum cleaner fan efficiency

of 70% is reached at a specific speed σ

bep

between 0.22 and 0.4 (0.3 in the figure). At

lower specific speed the maximum efficiency rapidly declines and is only 52% at σ

bep

=0.1.

177

Figure 47: Fan efficiency as a function of specific speed for industrial centrifugal fans in the

range up to 10 kW (source: Eurovent, EVIA. pers. comm.)

For a traditional vacuum cleaner fan, from before 2014, with the following characteristics:

• a speed n

of 20,000 rpm (333 rps), which is fast for a fan with a traditional universal

motor,

• a total fan pressure of 20,000 Pa and

• a volume flow rate of 40 litres/s (0.04 m³)

the specific σ

bep

is close to 0.1 and thus vacuum cleaner fan efficiency is 52%.

With the new EC-motors a fan speed n of 80,000 rpm (1332 rps) was reached in 2016.

Using the given formula, this means a specific speed σ

bep

=0.38. As the graph in Figure 47

indicates, this means the best vacuum cleaner fan efficiency is 70%. In other words, the

BLDC or SRM motor with its possibility to realise extremely high rotational speed, also

improves the strict fan performance with some 30-35%. In the latest Dyson V10 model,

released in 2018, a fan speed of 125,000 rpm (2083 rps) is reported. The fan geometry of

that model is compact and closer to that of a turbo-compressor than a traditional fan. The

fan-axis is made of a ceramic material rather than steel.

Last but not least, the new motor types are necessarily equipped with an inverter, i.e.

some powerful electronics that allow not only to efficiently regulate the motor speed but

also relatively easy and at low cost can accommodate sensors and other control options.

For instance, some manufacturers have introduced a sensor to keep the suction power

constant, independently of how full the receptacle is.

52%

70%

178

Furthermore, as with the motor efficiency, it must be taken into account that the energy

efficiency of the fan/drive/motor combination depends on the load and depends on how

the designer chooses the best efficiency point.

Traditionally in engineering the best efficiency point (BEP) of a fan-motor combination is

at around two-thirds to 80% of the maximum load. But the design-engineer may also

choose a different optimum as long as he/she stays in the stable operating range (without

severe stall, surge phenomena).

In that context the so-called ‘affinity laws’ are relevant, which say that at constant fan

diameter, the flow varies linearly with speed (rpm), the pressure varies quadratic with

speed and the power varies with the cube of speed. For instance, at 80% of the nominal

speed, the pressure drops to 64% of nominal pressure but the power drops to 51% of the

original power. Possibly, depending on the total of technical parameters, this might be an

optimal control setting for a particular load situation.

Note that the above discussion of fan design phenomena is only illustrative and aims to

give a plausible explanation of certain design phenomena. The actual optimisation of fan

aerodynamics, control options, etc. is very complex and requires not only sophisticated

computer modelling but also extensive empirical testing.

Costs play an important role. For instance, only the high-end models feature ultra-high

rotational speed values that allow to reach 70% efficiency. As will be elaborated in Task 6,

the costs that are associated with these design improvements are usually far beyond the

Least Life Cycle Costs point. For more economical models, even those using the low-end

versions of BLDC and SRM, fan/drive/motor efficiency values of 50% are more

representative of the Base Case.

Receptacle

Most consumer associations and manufacturers seem now to have accepted that there is

a market for bagless vacuum cleaners and a market for vacuum cleaners that have a bag

as the receptacle. Energetically there is not much difference. The ‘cyclone’-principle that

puts a swirl in the airstream to push out dust particles by centrifugal power does not cost

less energy than the pressure drop caused by a bag. The claim that the bag-less vacuum

cleaner keeps up performance regardless of how full the receptacle is was relevant in the

days of simple universal motors for vacuum cleaners. But especially with control-features

of EC-motors such performance can be realised at relatively low costs as well. Anyway,

consumer associations that test performance (also) with full bag did not find a significant

difference in performance between products with and without bags.

179

The main consumer choice is whether you want to pay for the bags to facilitate clean

emptying of the receptacle or not. Belgium consumer association Test-Achats stresses that

also bag-less models need to have the receptacle thoroughly cleaned and that filters need

to be changed. In their 2017 test they focus on testing vacuum cleaners with bags, because

‘c’est (souvent) mieux avec un sac’ (it’s often better with a sac) and 56% of vacuum

cleaners sold by the end of 2016 were with bag

244

245

Likewise, the German Stiftung Warentest remarks that the consumer saves costs of the

bags but the bag-less models are more expensive and manual cleaning of the receptacle

isn’t easy

246

Figure 48 shows the power consumption and price of 48 models with bags and 16 bagless

models tested in June/July 2017 by consumer associations in Germany, Belgium and the

Netherlands. Only models with power ≤900W were taken into account. The average price

for the whole population was 187 €/unit at an average power consumption of 717 W. The

average bagless model cost 230 € at a power use of 749 W. The models with bag cost 172

€ (33% less) and have a power input of 709 W (5-6% less). The overall score in the

consumer-tests for models with or without bags was comparable, with bag-less models

having a slightly better score on carpet cleaning and models with bag being more silent

and re-emitting less dust. More details are given in Annex E.

Figure 48: The volume of the receptacle is between 1.3 and 3.4 litres. Average size in the most

recent tests is 2.2 litres

244

Test –Achats 609, juin 2016, ‘Avec sac, c’est (souvent) mieux’, p. 41-43.

245

Test-Achats 620, juin 2017, ‘En plain dans le Miele’, p. 51-53

246

Test 7/2017, ‘Sauber mit weniger Watt’, p.52-55

500

550

600

650

700

750

800

850

900

950

0 100 200 300 400

Power in W

Price in Euro/unit

Vacuum Cleaners tested, June 2017

with bag

bag-less

180

Filters

The vacuum cleaner filter separates the dust particles from the air-stream. This is usually

at least a two-stage process: The first filter step can be:

• a paper/non-woven bag.

• in a bagless vacuum cleaner the separation of dust though centrifugal (“cyclone”)

forces.

• in a water filter, i.e. the air flow with dirt is forced through the water before it is

exhausted out of the vacuum. These vacuum cleaners must be emptied after each

cleaning, but they can clean dry and wet surfaces and even larger liquid spills. Note

that dry vacuum cleaners with a water filter are in the scope of the current

regulation.

Figure 49 The principle of a dry vacuum cleaner with a water filter (picture source: Kärcher

2018)

The dust stays in the bag, falls in the receptacle or stays in the water for later disposal.

The air moves on to the second stage, nowadays typically a HEPA (High Efficiency Particle

Air) filter, that takes out the last 0.1% of dust particles and prevents (together with

appropriate seals) re-emission of dust into the room.

In fact, in the dispute between bagged and bagless vacuum cleaners, the former suspect

that the cyclone-concept of a bagless vacuum cleaner is less effective than the filtering of

the bag and thus a larger part of the filter-burden is taken on by the HEPA-filter. This is

fine in the first cycle when the HEPA filter is fresh, but after a number of cycles the HEPA-

filter of a bag-less machine should be cleaned while the HEPA-filter of the bagged machine

can carry on for more cycles.

In the European Union, filtration is defined by standard EN 1822:2009. This standard

defines several classes of EPA/HEPA/ULPA air filters by their ability to retain the most

penetrating particle size (MPPS) particles, as shown in Table 65. MPPS for most filters is in

the range of 0.1 to 0.3 micrometers.

Table 65: Filter classes according to EN 1822:2009

Filter Group

Integral Value*

Local Value

181

Filter

Class

Filtration

Efficiency

Penetration

Filtration

Efficiency

Penetration

EPA- Efficiency

Particulate Air filter

E10

85.0%

15.0%

E11

95.0%

5.0%

E12

99.5%

0.5%

HEPA- High Efficiency

Particulate Air filter

H13

99.95%

0.05%

99.75%

0.25%

H14

100.00%

0.01%

99.98%

0.03%

ULPA- Ultra Low

Penetration Air Filters

U15

100.00%

0.00%

100.00%

0.00%

U16

100.00%

0.00%

100.00%

0.00%

U17

100.00%

0.00%

100.00%

0.00%

* Integral value shows efficiency of the air filter as a system and that is what average user should

be focused on. EN 1822 standard doesn't define Local Values for E10-E12 filters.

The HEPA filter can be combined with active carbon, which can absorb various chemicals

on a molecular basis, but can be problematic with larger particles. Also, a combination with

scent, to give an extra feeling of freshness, is quite common. For obvious reasons ‘scent’

does not combine well with active carbon in a filter configuration.

In more exotic models, not typical for the EU market, the filtering in a vacuum cleaner can

be combined with an ioniser, to clean the air electrostatically, or with UV light, to kill germs.

Both solutions may not be without health risk as (traces of) ozone may be generated

247

The pressure-drop caused by a typical HEPA filter is around 250-300 Pa when the filter is

empty and twice as much 500-600 Pa when it is ‘full’. As a general rule of thumb

replacement every 6 months is recommended. Compared to the suction power in a cylinder

vacuum cleaner (10-20 kPa) the filter takes up some 2-4%, depending on how full the bag

is. From the ‘A’ dust re-emission Energy-Label rating on most models and the consumer

association tests on this aspect, the HEPA filter solutions seems to be doing a good job, at

least when starting out with a fresh filter.

Hose

The hose of a cylinder vacuum cleaner is typically a flexible corrugated plastic tube,

reinforced with a metal spiral wire. Inner diameter is around 30-35 mm and the outer

diameter is some 10 mm more. As was established in the special review study, the current

lifetime test in the regulation (at least 40.000 flexes) and the associated standard is

adequate for the primary hose of a cylinder vacuum cleaner. Attached to the hose is a steel

cylinder with diameter of around 30-32 mm and a length of on average 95 cm. The cylinder

is used to manipulate the attached nozzle.

247

http://www.pickvacuumcleaner.com/exhaust-filtration.html

182

The corrugated flexible hose causes a significant pressure-drop (VHK estimate 300 Pa, 2-

3% of power). One of the advantages of upright, handstick and robot vacuum cleaners is

that they don’t use a flexible hose and thus pressure loss is much lower (<50-100 Pa). The

disadvantage of the first two types is of course in the ergonomics of having to manipulate

not just the nozzle but the full weight of the fan and motor.

For the secondary hose of an upright vacuum cleaner, which is typically made for

elongation and not only flexing, the test needs to be revised.

The hose is one of the components that may need to be replaced during the lifespan of the

vacuum cleaner. In general, replacement is easy and spare parts are amply available. The

reason for the hose breaking is rarely a break in the middle (as it would from the largest

stress in a bending test) but would be a break where it is attached to a rigid part, i.e. the

attachment to the metal tube or the attachment to the vacuum cleaner casing.

Nozzle

In recent years there has been discussion on the use of special nozzles that are part of the

product package, but which, apart from when testing the cleaning performance in an

energy label test

248

, are hardly used in normal practice. For instance this is the case for

some hard floor nozzles that adhere perfectly to the floor and pick up the dust from the

crevices very well in the test but that in practice are not so useful because they do not pick

up, but rather push around, the larger debris and hairs. In fact, most consumer

associations advocate the use of the universal nozzle fit for both hard-floor and carpet

cleaning for performance testing, especially since the nozzle design has large impact on

the cleaning performance. Consumer surveys in the Netherlands show that more than half

of the consumers never use any other nozzle (See Annex E). The other half might use the

smaller nozzles for cleaning furniture, curtains or automobile-interiors.

In nozzle-design there are two philosophies: the passive nozzle, popular in continental

Europe for its simplicity and effectiveness on all sorts of floors, and the active nozzle,

popular in the UK and Ireland and praised for its superior performance on carpets. “Active”

implies that the nozzle is equipped with a ‘beat & brush’ agitator, e.g. a rotating roll with

brushes, that gives a mechanical stir to the carpet to facilitate better dust removal from

the carpet. It is traditionally found on upright vacuum cleaners and on some handstick

vacuum cleaners. It is reported to be especially effective with the removal of hair and fibres

from the carpet. Consumer associations mention that the active nozzle brings additional

material use and add a risk for product failure.

248

According to the previous, annulled Energy Labelling Regulation

183

From the point of view of energy efficiency, it is difficult to say whether the active nozzle

has a positive or a negative impact. On one hand, the agitator takes up extra motor power,

often drawing its electricity from a (rechargeable) battery. On the other hand, it appears

from consumer association tests that the main fan motor power can be reduced to e.g.

650 W instead of 750 W to get the same carpet cleaning performance. According to some

stakeholders the active nozzles are more likely to break and thus contributes to larger

material consumption.

Batteries

In the current regulation all types are mains-operated, except for a possible battery-driven

active nozzle. But the possible newcomers cordless and robot vacuum cleaners both have

batteries and thus battery chargers. In task 3 the running hours of cordless and robot

types in the various modes were assessed, while charging, while operating as a vacuum

cleaner and when fully charged and docked.

The power consumption in those modes can be estimated from consumer tests (see Annex

E) and sometimes from product specification sheets. The Stiftung Warentest assessment

of February 2018

249

specifies for instance the running time in maximum and minimum

power mode. The two best performing cordless vacuum cleaners have a maximum power

in the range of 400-450 W with a runtime of 8 to 15 minutes. This means an effective

battery capacity of 60 to 100 Wh. At minimum power the runtime becomes 27 and 82

minutes, respectively. There are 5 models with a maximum power in the range of 250-350

W with a runtime of 14 to 37 minutes, meaning that the battery capacity is in the range of

50 to 80 Wh. Prices for the replacement batteries for cordless vacuum cleaners in the test

vary between 30 € for a low capacity (ca. 30 Wh) NiMH battery and 105 € for a 100 Wh

Li-ion battery. At the moment, Li-ion batteries are the most popular, despite their higher

prices.

Vacuum cleaner batteries are typically of the Ni-MH (Nickel Metal Hydride) type or Li-ion

(Lithium-ion) type

250

. The former features a product-life of ~400 charges and has a

memory effect that may reduce long-term capacity

251

; the latter will recharge 1000 times

(or more) and has no memory effect.

At for instance 5 recharges per week, a Ni-MH battery will last less than 2 years and the

Li-ion battery lasts 4 years. Furthermore, the self-discharge of Li-ion batteries is in the

249

Viel Lärm um nichts, Test 2/2018, p. 52-56

250

Note that NiCd (Nickel Cadmium) batteries are now banned in the EU

251

Memory effect relates to a diminished battery capacity in time, as a result of supoptimal (incomplete, or too soon) charging.

In some cases the effect is reversible e.g. by applying a full discharge/charge cycle.

184

order of <5% per month. For NiMH it is in the order of 5% in the first week and 50% in

the first month

252

253

This means, for instance, that a large 100 Wh Li-ion battery loses less than 5 Wh/month.

On a continuous basis (1 month is 720 hours) this is 0.007 W. For a NiMH battery of the

same capacity the power loss is 10 times more.

It is important not to overcharge the Li-ion batteries, this is one of the reasons why the Li-

ion cells are not ‘trickle charged’. Trickle charging is charging at a very low current, just

enough to compensate for self-discharge, to spare battery-life. It is typical a strategy for

lead-acid and the now forbidden NiCd batteries.

254

Unfortunately it is also applied to Ni-MH

batteries in some vacuum cleaners and can then lead to maintenance’ (charged and

docked) losses of 4.5 W, whereas in fact the self-discharge is only 0.07 W in a worst case.

Furthermore, it might spare battery life when done correctly, but also for Ni-MH

overcharging is sub-optimal for battery life.

Charging conventional Ni-MH batteries is slow, at 10-12 hours per charge, whereas the Li-

ion batteries can be recharged in 1 to 3 hours. Li-ion cells have a higher voltage than Ni-

MH cells: 3.6 V versus 1.2 V. Vacuum cleaner batteries will thus show a voltage that is a

multiple of 3.6 V, usually between 18 and 36 V.

255

An important energy-related feature of batteries is the charging efficiency, i.e. the ratio of

electric power output and the electric power input for charging. For Li-ion batteries this

amounts to 85%. For Ni-MH batteries a typical value is 69%.

Last but not least, the efficiency of the battery charger plays a role. A battery charger is

basically an external power supply (EPS) and a regulator. The Impact Assessment report

on External Power Supplies mentions EU proposals no-load power use of 0.3 W for a

multiple voltage EPS with PO (power output) < 250W source and an active efficiency of

86%

256

. Assuming the 2012 US DoE standards for EPS to be representative of the average

power supply cost, the EU proposal for e.g. a 120 W PO would cost 1.99 € more (consumer

price incl. VAT). When saving 30 kWh/year at 0.20 €/kWh the EU consumer would pay

back this 1.99 € in 4 months.

The European consumer association ANEC/BEUC notes that the maintenance mode of

battery load in portable appliances (trickle charge) is considerable

257

. The consumer test-

252

https://turbofuture.com/misc/Which-is-better-Nickel-Metal-Hydride-NiMH-or-Lithium-Ion-Li-ion-batteries

253

There are low-self discharge (LSD) rate types available. They are more reliable than the standard NiMH but they have lower

capacities, usually around 2000mAh.

254

https://batteryuniversity.com/learn/article/charging_lithium_ion_batteries

255

The capacity of a 3.6V Li-ion cell is around 1.5Ah (→ 4.4Wh), so often the capacity can be calculated in that way. E.g. a 36V

Li-ion battery will have 10 cells and thus a capacity of 44Wh. A Ni-MH cell, at 1.2V, will have a typical capacity of 2.2Ah.

256

Viegand Maagøe A/S, internal draft. 2018.

257

Comment by ANEC/BEUC on the draft interim report, January 2018.

185

institute ICRT

258

tests of cordless vacuum cleaners and robot vacuum cleaners show that

the average load over 24 hours in the ‘charged and docked’ condition of models over the

past years varied between <0.5 and 8 W. This means a yearly ‘maintenance mode’ energy

use of 60 kWh, which is higher than the total yearly energy use of regulated canister

vacuum cleaners. ANEC/BEUC suggests that values of 0.5 W or maybe 1 W max for this

condition are perfectly possible, as some of the models currently on the market already

would comply. Setting a requirement on a 24-hours average would still allow docking

stations to use more energy for a short time to perform relevant tasks such as updates.

This is close to the limits set in the standby regulation

259

effective from January 2019,

which differentiates between three standby modes. A similar solution could be considered

for battery operated vacuum cleaners.

While on most energy and environment aspects the Li-ion batteries score best, there is the

problem of the cobalt content. Cobalt makes up 10-20% of the battery weight. The Li-ion

battery’s specific capacity is around 100 Wh/kg and the average household cordless stick

vacuum cleaner battery weighs around 0.4-0.7 kg (say 0.5 kg on average). So each of

these contains 0.05-0.1 kg of cobalt. Note that there are several Li-ion types and not all

use cobalt. Dyson vacuum cleaners, for instance, uses Aluminium Nickel instead of cobalt.

This changes the battery properties. No cost information could be found to compare the

different Li-ion battery types, but the recent article by Charles Amoabeng Nuamah

260

gives

an overview of relevant selection criteria and typical characteristics of 6 Li-ion types.

The criteria are

• Specific energy: this defines the battery capacity in weight (Wh/kg). The capacity

relates to the runtime. Products requiring long runtimes at moderate load are

optimized for high specific energy.

• Specific power: this is the ability to deliver high current and indicates loading

capability. Batteries for power tools are made for high specific power and come with

reduced specific energy.

• Performance: how well the battery works over a wide range of temperature. Most

batteries are sensitive to heat and cold and require climate control. Heat reduces

the battery lifetime, and cold lowers the battery performance temporarily.

• Lifespan: this reflects cycle life and longevity and is related to factors such as

temperature, depth of discharge and load. Hot climates accelerate capacity loss.

Cobalt blended lithium ion batteries also usually have a graphite anode that limits

the cycle life.

• Safety: this relates to factors such as the thermal stability of the materials used in

the batteries. The materials should have the ability to sustain high temperatures

before becoming unstable. Instability can lead to thermal runaway in which flaming

258

http://www.international-testing.org/

259

OJ L 225, 23.8.2013, p. 1–12, COMMISSION REGULATION (EU) No 801/2013 https://ec.europa.eu/info/energy-climate-

change-environment/standards-tools-and-labels/products-labelling-rules-and-requirements/energy-label-and-

ecodesign/energy-efficient-products/mode-standby-and-networked-standby_en

260

See https://owlcation.com/stem/Comparing-6-Lithium-ion-Battery-Types

186

gases are vented. Fully charging the battery and keeping it beyond the designated

age reduces safety.

• Cost: cost of lithium-ion batteries plays a major role in determining the initial

product price. Hence cost is an important factor when selecting the type of lithium-

ion battery.

Table 66: Comparison properties of Li-ion battery types (L =Low, M=Moderate, H=high)

Lithium-ion battery

Types

Specific

power

Specific

energy

Safety

Lifespan

Costs

Perfor-

mance

Lithium Cobalt Oxide

Lithium Manganese Oxide

Lithium Nickel Manganese

Cobalt Oxide (NMC)

Lithium Iron Phosphate

Lithium Nickel Cobalt

Aluminum Oxide (NCA)

Lithium Titanate

NMC batteries are the most popular type for vacuum cleaners. There are two subtypes, i.e.

NCM 1-1-1 with equal parts of Ni, Co and Mn (molar ratio) and NCM 5-3-2. Dyson (and

e.g. Tesla for cars) is using NCA.

For batteries in robots nearly all of the above applies, except that the power consumption

of the robot is lower and the battery capacity smaller. Battery capacity, in Ampere hours,

depends on the power consumption of the robot vacuum cleaner. Top models may have a

power consumption of 70-80W and feature batteries with capacities of 3.6 Ah to 5.2 Ah

batteries. They will typically use Li-ion cells. Low-end robot vacuum cleaners may feature

a power consumption of only 11-24 W and battery capacities lower than 2 Ah. They will

typically use Ni-MH types.

The endurance of batteries, in terms of how many cycles they can withstand without losing

capacity, is highly dependent on how they are used and charged. In Table 67 the estimated

number of discharge/charge cycles for Li-ion batteries before the capacity drops to 70% is

shown. The depth of discharge (DoD) constitutes a full charge followed by a discharge to

the indicated percentage. Partial charge and discharge reduce stress to the battery and

therefore prolong the battery life

261

Table 67: Cycle life of LI-ion batteries as a function of DoD.

Depth of discharge

Discharge cycles

(NMC / LiPO4)

261

https://batteryuniversity.com/learn/article/how_to_prolong_lithium_based_batteries

187

100% DoD

~300 / 600

80% DoD

~400 / 900

60% DoD

~600 / 1,500

40% DoD

~1,000 / 3,000

20% DoD

~2,000 / 9,000

10% DoD

~6,000 / 15,000

Plug and power cord

On average, according to the latest test by Stiftung Warentest, a cylinder vacuum cleaner

has a power cord of 10-11 metres. The power cord is retractable, using a mechanical

spring. The retraction mechanism of the power cord is one of the components that most

frequently needs repair and is often not easy to repair.

10.2 Materials and resource level

Resource efficiency is a growing concern within Europe and globally. More raw materials

are categorised as critical and the dependency of these materials are increasing. APPLIA

has initiated a collaboration with Digital Europe and recyclers (e.g. EEra

262

) to assess the

possibilities of how to comply with the information requirements in the WEEE directive

(article 15, specified in Annex 7). They have discussed how the information should be made

available, and came up with the idea of an online joint platform

263

, which contain necessary

information on all product categories (also taking into account different technologies).

The following section provides an overview of the material composition and distribution of

vacuum cleaners, and compare typical products to best available technology to support the

resource efficiency assessment. The inputs will be used to model the environmental

footprint in later tasks. The material composition provides also valuable inputs to the

discussion on resources.

Material consumption in vacuum cleaners

In November 2015, JRC-IES Ispra published its case study on durability of vacuum

cleaners

264

. As such, the study not only looked at the durability aspects but made a

complete analysis of all environmental impacts, based on a product analysis of a recent

cylinder vacuum cleaner. As such it constitutes the most recent Bill-of-Materials available

of a random cylinder vacuum cleaner. Below is an exploded view of a vacuum cleaner

presented in Figure 50.

262

http://www.eera-recyclers.com/about-us

263

To the knowledge of the study team, this platform has not yet been launched.

264

Silvia Bobba, Fulvio Ardente, Fabrice Mathieux, Durability assessment of vacuum cleaners, JRC-IES, Technical support for

Environmental Footprinting, material efficiency in product policy and the European Platform on LCA, November 2015.

188

Figure 50: Example of an exploded drawing and spare parts listing for the canister (left) and

the nozzle plate (right)

In Table 68 is the bill-of-materials of an average vacuum cleaner from the JRC study

presented. The bill-of-materials presented only serves as an example of the variety of

materials included in vacuum cleaners and which components the different materials are

present. In general, the material composition and weight of vacuum cleaners are expected

to be very similar to the values presented in the preparatory study. Only the material

composition of robot vacuum cleaners has been added, which is derived from a study on

end of life resource recovery from emerging electronic products

265

Table 68: Bill-of-materials, Cylinder Vacuum Cleaner (source: JRC-IES 2015)

Component

Material

Mass (kg)

Hose

ABS

0.461

0.214

0.018

Rubber

0.003

Motor

Aluminum (cast)

0.042

Brass (CuZn20)

0.025

Copper sheet

0.124

Copper windings

0.0326

Core

0.271

Mounting

0.0579

265

Parajuly, K., Habib, K., Cimpan, C., Liu, G. and Wenzel, H. (2016). End-of-life resource recovery from emerging electronic

products – A case study of robotic vacuum cleaners. Journal of Cleaner Production, 137, pp.652-666.

189

Component

Material

Mass (kg)

BMC-GF (polyester- glass-fibre

reinforced)

0.267

Graphite

0.007

0.016

0.259

Rubber sealing compound

0.133

Steel

0.614

Canister case

ABS

POM

0.042

Rubber

0.002

Steel

0.004

Cord Reel

Brass

0.004

ABS

0.142

0.021

Rubber

0.002

Steel

0.052

Plug & cord

PVC

0.194

Copper

0.089

Nozzle plate

ABS.PP

0.052

PE-HD

0.02

0.219

Steel

0.019

Filter

PE-HD

0.017

Wheels

0.209

Cables

Brass

0.002

0.015

PVC

0.011

Wires

0.005

Cables

Brass

0.001

PVC

0.002

Wires

0.002

Printed Circuit Board

(PCB)

PCB

0.012

Steel

0.014

Packaging

PE-LD

0.06

Manual

Paper

0.1

Packaging

Cardboard

1.1

TOTAL

6.957

The weight in grams and percentage distribution of various materials can be seen in Table

69 for typical representative products for each vacuum cleaner type.

190

Table 69: The assumed material composition in the current study.

Category

Materials

Household mains-operated

Commercial

Cordless

Robot

Bulk Plastics

11 -ABS

3643

5795

1624

2657

TecPlastics

12 -PA 6

638

144

286.5

337

Ferro

24 -Cast iron

863

1436

400

823

Non-ferro

31 -Cu

tube/sheet

307

766

354.5

224

Non-ferro

27 -Al

sheet/extrusion

544

1336

480.5

344

Coating

Electronics

98 -controller

board (PCB)

295

607

Misc.

various other

materials

728

1631

Total weight

6780

11110

3440

5041

Bulk Plastics

11 -ABS

54%

52%

47%

53%

TecPlastics

12 -PA 6

Ferro

24 -Cast iron

13%

12%

16%

Non-ferro

31 -Cu

tube/sheet

10%

Non-ferro

27 -Al

sheet/extrusion

12%

14%

Coating

Electronics

98 -controller

board (PCB)

12%

Misc.

various other

materials

11%

15%

Total

100%

All vacuum cleaner types are mainly made of plastics, the share ranging from 50% plastics

for cordless vacuum cleaners to 56% for household mains-operated vacuum cleaners. A

notable difference is in the amount of electronics, where robotic vacuum cleaners have the

highest share and amount. Note that the batteries are included in the non-ferro materials

and weighs approximately 500 grams in robotic vacuum cleaners.

Many vacuum cleaners use consumables in terms of bags and filters during their use phase.

Based on a JRC report, the following assumptions are made regarding the composition and

weight of bags and filters

266

http://publications.jrc.ec.europa.eu/repository/bitstream/JRC96942/lb-na-27512-en-n_.pdf

191

• Bags made of propylene, estimated weight per bag: 0.04 kg.

• Filters made of PE-HD, estimated weight per filter: 0.0017 kg.

The impact of bags will be quantified in later tasks based on these assumptions, but the

dust bags and filters can also be made of other materials e.g. dust bags made of fleece

(PET)

267

and filters of polyester

268

Critical materials and components

The awareness of critical resources is increasing, and the Commission carries out a

criticality assessment at EU level on a wide range of non-energy and non-agricultural raw

materials. In 2017 the criticality assessment was carried out for 61 candidate materials

(58 individual materials and 3 material groups: heavy rare earth elements, light rare earth

elements, platinum group metals, amounting to 78 materials in total). The updated list of

critical raw materials is presented in Table 70.

Table 70: List of critical raw materials

Critical raw materials 2017

Antimony

Fluorspar

LREEs*

Phosphorus

Baryte

Gallium

Magnesium

Scandium

Beryllium

Germanium

Natural graphite

Silicon metal

Bismuth

Hafnium

Natural rubber

Tantalum

Borate

Helium

Niobium

Tungsten

Cobalt

HREEs*

PGMs*

Vanadium

Coking coal

Indium

Phosphate rock

*HREEs=heavy rare earth elements, LREEs=light rare earth elements, PGMs=platinum group metals

Each type of vacuum cleaner may contain several raw materials categorised as critical.

Raw materials such as vanadium and phosphorous are in some designations of steel used

as alloying elements. These alloying elements are not included in this assessment as they

are very difficult to quantify, and more obvious choices are present such as:

• Printed circuit boards which may contain several critical materials such as

palladium, antimony, bismuth, tantalum etc.

269

• Motors which may contain rare earths

• Cobalt in batteries

Simple printed circuit boards are present in most mains-operated vacuum cleaners, e.g. to

hold switches, resistors, etc. Only in the (rare) case of using frequency converters the

electronics can become a little more complex. But gold-bumps to hold ICs (Integrated

Circuits) are not generally present in most vacuum cleaners. Instead, cordless vacuum

267

https://www.miele.co.uk/domestic/1779.htm?info=200046044-ZST

268

https://www.nilfiskcfm.com/filtration/filters/

269

http://www.wrap.org.uk/sites/files/wrap/Techniques%20for%20recovering%20printed%20circuit%20boards%2C%20final.pd

192

cleaners will feature battery chargers, usually a power supply with a regulator (but no ICs

with gold-bumps) and of course the battery cells (with cobalt).

Proper electronics boards, as referenced in the Ecoreport, can be found in all robot vacuum

cleaners. But the amount of critical raw materials is properly higher in robotic vacuum

cleaners as they contain the highest amount of printed circuit boards and at the highest

grade. The average composition of a printed circuit board is assumed as follows

270

• 70% - Non-metallic e.g. glass-reinforced polymer

• 16% - Copper

• 4% - Solder (containing tin)

• 3% - Iron. ferrite (from transformer cores)

• 2% - Nickel

• 0.05% - Silver

• 0.03% - Gold

• 0.01% - Palladium

• <0.01% - Other (bismuth, antimony, tantalum etc.)

This means that robot vacuum cleaners contain gold in the range of 0.03 grams which

originates from the printed circuit boards. The grade

271

of printed circuit boards in vacuum

cleaners can be discussed, but the complexity of robots is increasing which imposes higher

grades of printed circuit boards to be used. For robots the grade is assumed to be

comparable to a midrange laptop.

The printed circuit boards and wires are already targeted components according to the

WEEE-directive. The same goes for batteries and electronic displays.

Copper is also very important to remove before shredding, not only because it is identified

as a critical raw material, but to minimise the risk of copper contamination in the iron

fraction since it can directly influence the mechanical properties of the recycled

iron/steel

272

. Avoiding contaminants is one of the key points of design for recycling. In

order to avoid contamination, it is important to

273

• Reduce the use of materials, and especially the materials that will cause

contamination in the recycling process (e.g. metal screws in plastics or combination

of steel and copper). It should be considered how the materials would behave in

the sorting and processing at end of life.

• Identify materials in assemblies combined in an inappropriate way so resources are

lost during recycling. E.g. the connection between a metal screws and plastic, where

one of the materials may be lost due to incomplete separation.

270

http://www.wrap.org.uk/sites/files/wrap/Techniques%20for%20recovering%20printed%20circuit%20boards%2C%20final.pd

271

The grade of PCBs is dependent on the amount of precious metals (e.g. gold and silver), which can vary between the

category of WEEE and its age.

272

http://www.rmz-mg.com/letniki/rmz50/rmz50_0627-0641.pdf

273

Reuter, M.A. & Schaik, A.V.A.N., 2013. 10 Design for Recycling Rules , Product Centric Recycling & Urban / Landfill Mining.

, pp.1–15.

193

Manufacturing and distribution

During manufacturing primary scrap is generated, but the primary scrap production is

estimated to be negligible. It is assumed that cuttings and residues are directly reused into

new materials within the factories, making material losses very low.

Additional materials are used in the distribution of products. Usually cardboard, plastic and

expanded polystyrene are used to protect the product during transport. Packaging

materials are sorted by the end-user and recycled, burned or landfilled. Cardboard is easily

recyclable while the plastic is probably burned or recycled. Regarding the expanded

polystyrene it can be compressed and recycled into polystyrene. The problem is the density

and volume of the expanded polystyrene. It must be compressed to make it both affordable

and environmentally sound.

The distribution of products depends on the location of sales and production, but generally

large cargo ships are used for intercontinental transport, while trains and road

transportation are used for shorter distances. The impact depends on the specific product

and its geographical route, but in most life cycle assessments the transportation impact

turns out to be negligible compared to the environmental impact of the rest of the product.

Vacuum cleaners are no exception, as most are assumed to be shipped by freight ship or

by truck. Both alternatives in general have a low impact in the final assessment.

Recycled content

Hereafter a brief discussion of the recycled content of vacuum cleaner materials is given.

Note that in this report ‘recycled content’ always refers to material input (in g) that is

derived from post-consumer waste. Recycling waste from primary scrap, which is a

common and profitable activity in industry, is not included.

• The non-ferro metals in vacuum cleaners are mainly copper and aluminium.

Globally, more than half of the copper products are made from recycled copper.

However, this percentage mostly comes from recycled content of tube and sheet

alloys (bronze, brass), which are not in vacuum cleaners. The wire windings of the

motor have to be very pure for optimal electric conductivity. The same goes,

although to a lesser degree, for the copper wire in power cords. Overall for copper

in the vacuum cleaners a recycled content of 10% is assumed. The aluminium in

vacuum cleaners is usually situated around the motor, as part of the frame or

motor/fan housing. Typically, it will be aluminium diecast, which uses 85% recycled

content.

• The ferro-materials relate mostly to the core material of the motor, some small

steel parts (e.g. axes for wheels, spiral wire) and the tube holding the nozzle. For

the core material of the motor there are numerous alternatives, e.g. soft iron, steel

194

laminates, etc. There is no environmental profile in the EcoReport, but assuming a

laminated steel core, the galvanised steel sheet with a recycled content of 5%,

probably comes closest. Cast iron (85% recycled content) could be used for the

motor frame. The chromed hose could be rolled from sheet (5% recycled content).

• Technical plastics, actually usually thermosets like epoxy resin or polyester

compounds, are only a small fraction of the total plastics. They are used where

temperature-resistance is required and/or as casing/mounting plate of electric parts

(switches, etc.). Recycled content of thermosets is usually 0%.

• Bulk plastics, like PP (polypropylene) and ABS, constitute half or more than half of

product weight. Most goes into the casing. Normally, the recycled content is 0%,

but since a few years there are some manufacturers that have started to use

considerable fractions of recycled PP and ABS, up to 70%. Assuming that these

manufacturers might constitute 20% of the market it means that on average there

is a 14% recycled content for bulk-plastics. Note that PP is also a common non-

woven material for filters and bags.

• As regards electronics (including batteries) only very small fractions of recycled

content, i.e. those relating to valuable materials like gold (in robot VCs), palladium

(in condensers), cobalt (batteries), etc. can be assumed. Given that they constitute

a high environmental impact, it can be said that the recycled content represents a

negligible mass, but at least 20-30% (say 25%) of the impact.

• The packaging of vacuum cleaners is now mainly an LD-PE (low-density

polyethylene) bag, a cardboard box, possibly with cardboard or EPS (expanded

polystyrene) inserts for the corners. The cardboard is 90% made of recycled

material. The manual is made of printing paper, very often also from recycled

material (50% assumed).

Use phase

There are two main non-energy material strategies linked to the use phase:

• Reduction of the consumption of bags and filters, e.g. by re-usable/washable filter-

boxes, cyclone separation (‘bagless’), etc.

• A longer product life to slow down the material cycle of vacuum cleaners and thus

save materials in production and end of life. This can be achieved by increasing

reparability by setting minimum technical life requirements on certain components

and keeping spare parts available

Both directions will be discussed in section 12.

End of life

At end of life the waste stream can be split into re-use, recycling, heat recovery,

incineration without heat recovery (of hazardous materials in general) and landfill.

195

Furthermore, especially if the product sales were increasing or declining rapidly over a

relatively brief period in time, there is a mismatch between the mass of materials in

production and the mass being discarded. This is caused by the time displacement between

acquisition and disposal of the products, which means that in the meantime the material

is in the stock. For instance, for cordless and robot vacuum cleaners this 'in-stock' material

plays an important role, causing a delay from purchase to disposal of materials (i.e. the

vacuum cleaners bought today will not be seen in disposal until 5-6 years time). For the

more traditional vacuum cleaners where the markets are more mature, the 'stock' plays a

minor role, since the input and output of products to and from the stock is more or less

constant.

The starting point for the end of life process is the collection, which was discussed in section

9. While, according to the WEEE Directive, the collection rate should become 65% in 2019,

the collection rate for small appliances (including vacuum cleaners) was only 40% in 2014.

This means that 60% ended up in the mixed household fraction, where there is also

recycling, e.g. of the metals, batteries and possibly robot-PCBs (Printed Circuit Boards),

but where e.g. vacuum plastic plastics are usually not singled out and go to heat recovery.

As regards 're-use' there is something of a definition problem. In the German study on

obsolescence re-use could be perceived as people giving away the product either to family

and friends or to a 'green' shop. That is a route that the first users may follow in 7-8% of

the disposals. It will help to really get to the projected product life of 8 years (for mains-

operated household vacuum cleaners), but there is very little in terms of design, and thus

also in terms of Ecodesign Regulations, to do about it.

In this study re-use is assumed to be the case if there is systematic refurbishment of the

product and that is much rarer and more estimated to happen only in 1% of the disposals.

From the viewpoint of designing a new product, which is the perspective of Ecodesign

measures, recycling relates to two aspects: recyclability of the product at the end of life

and maximum use of post-consumer recycled content for the new product. If the two are

in balance, there is a true ‘circular economy’. However, after a life of intensive use, most

products and their materials degrade and thus there is inevitably some downgrading.

As regards ‘Design for Recycling’ there are different directions. The concept dates back to

the late 1970s and was initially synonymous only to ‘design for disassembly’, i.e. facilitating

mainly manually separate material fractions of discarded products. However, over the last

50 years the economic reality of recycling EEE (Electric and Electronic Equipment) did not

evolve in the direction of sophisticated manual dismantling, but instead (apart from some

worthwhile components specified in the WEEE-Directive) focused on a first very rough

196

manual split, feeding a shredder and then physical/chemical processes (magnetism,

floating, etc.). In the case of vacuum cleaners, for instance, the recyclers cut off the power

cord for its copper content to be gained from specialised processing. In cordless vacuum

cleaners the batteries are of course removed and in the case of robot vacuum cleaners,

the printed circuit boards (PCB) are also removed beforehand to follow a different

processing route, usually also involving a shredder.

After the shredder, the metal parts are separated by physical means (magnetic, eddy-

current, specific weight). In the remaining flow the bulk-plastics PE, PP, PS and ABS are

separated individually on the basis of specific weight

274

. The diversity of the remaining

plastics types is too large and their total quantity too small to make the potential gains to

be derived from their separation worth the extra costs involved in the process.

The most used plastics in vacuum cleaners are PP (polypropylene) and ABS (Acrylic

Butadiene Styrene). As mentioned, the post-shredder separation of these two plastics is

current practice and thus there is no need for detailed Design for Disassembly. However,

it is important to keep the PP and ABS as pure as possible in each moulded part, i.e. to

avoid glass-fibre reinforcements, fillers or large quantities of additives such as

(halogenated) flame retardants. Also blends and co-polymers of PP and ABS in single parts

should be avoided as much as possible. However, and this has led to misunderstandings

when using simplified matrices of ‘compatible’ plastics, there is no significant negative

effect for recycling to use different parts of ABS or PP or any plastic in one product as long

as each part is pure. Also it is not problematic to use metallic fasteners

275

. Finally, in this

case the marking of larger plastic parts, which manufacturers anyway undertake on a

voluntary basis, is useful in case of extensive disassembly. If there is no such disassembly

of every component the impact of marking will be insignificant.

With the motors (metal) becoming smaller and with increased use of plastics, the plastics

are now 60% or more of the total material input and thus 60% of the future waste stream

when the products currently put on the market reach their end of life.

The recycling rates in this study are based on the EcoReport tool

276

but updated regarding

plastic

277

. The values used in the current study are presented in Table 64 in section 9. At

the moment some 29% of the plastics is considered to be recycled, 40% goes to heat

274

http://www.ecodesignlink.be/en/basic-plastic-types?parent=176

275

Based on the stakeholder inputs care must also be taken towards vague description such as “It must be possible to separate

the connections easily” since it is imposible for the market surveailince authorities to control this.

276

http://ec.europa.eu/growth/industry/sustainability/ecodesign_da

277

Plastic Europe, Available at: http://www.plasticseurope.org/documents/document/20161014113313-

plastics_the_facts_2016_final_version.pdf

197

recovery, 31% to landfill (see section 9). The credit for recycling in Ecoreport amounts to

40% of all impacts

278

After having missed the recycling stage, it is important for possible heat recovery from the

remaining fractions, mainly plastics and electronics, that there are no hazardous materials

included. Apart from the materials mentioned in RoHS, for which no special action would

be required, this includes “Substances of Very High Concern” in REACH and plastic-

additives such as halogenated flame retardants.

If any of these hazardous materials are present, the fractions need to be incinerated

without heat recovery or there is the risk that they end up in landfill. Otherwise, at least

for those fractions with a combustion value, they will contribute to heat recovery. For these

materials there is a credit of 30% for all environmental impacts according to EcoReport.

Blue Angel requirements

The best available vacuum cleaners regarding resource efficiency are considered to be

those who are awarded the German eco-label the Blue Angel as this eco-label also sets

requirements concerning the resource efficiency besides more common requirements as

the energy consumption. The Blue Angel requirements concerning the resource efficiency

are:

Material requirements for the plastics used in housings, housing parts and accessory

parts (suction tube/hose, nozzle etc.) No substances may be added to the plastics as

constituent parts which are classified as:

• Carcinogenic of categories 1A or 1B according to Table 3.1 of Annex VI to Regulation

(EC) 1272/2008

• mutagenic of categories 1A or 1B according to Table 3.1 of Annex VI to Regulation

(EC) 1272/2008 According to DIN EN 60312-1, para. 3.4.

• Toxic to reproduction of categories 1A or 1B according to Table 3.1 of Annex VI to

Regulation (EC) 1272/2008

• Toxic to reproduction of categories 1A or 1B according to Table 3.1 of Annex VI to

Regulation (EC) 1272/2008

• Being of very high concern for other reasons according to the criteria of Annex XIII

to the REACH Regulation, provided that they have been included in the List (so-

called Candidate List) prepared in accordance with REACH, Article 59, paragraph 1

• Halogenated polymers shall not be permitted. Nor may halogenated organic

compounds be added as flame retardants. Moreover, no flame retardants may be

added which are classified pursuant to Table 3.1 or 3.2 in Annex VI to Regulation

(EC) 1272/2008 as very toxic to aquatic organisms with long-term adverse effect

and have been assigned the Hazard Statement H 410 or Risk Phrase R 50/53.

278

Another solution, not taken into account, is to improve the recycling facilities by investing in improved sorting technologies

or new technologies such as carbon capture technologies . Carbon capture technologies can in the future use CO2 (e.g. from

combustion of plastic) as a feedstock for polymers. See https://setis.ec.europa.eu/setis-reports/setis-magazine/carbon-

capture-utilisation-and-storage/co2-feedstock-polymers

198

Recyclable and easy-to-maintain design. The appliance shall be so designed as to

allow quick and easy disassembly with a view to facilitating repair and separation of

valuable components and materials. This means that:

• It must be possible to separate the connections concerned by the use of ordinary

tools and the points of connection must be easily accessible

• Plastics should consist of one polymer only and plastic parts greater than 25 g in

mass must be marked according to ISO 11469 to allow for a sorting of plastics by

type

• Disassembly instructions must be made available to end of life recyclers or

treatment facilities in order to recover as many valuable resources as possible,

• The plastics used should consist of recycled material, if possible.

Durability. The appliances shall meet the following durability requirements:

• The motor shall have a minimum service life of 600 hours

• The suction nozzle must be able to withstand the impact of at least 600 drum

rotations (or 1200 falls from as high as 80 cm).

• The suction hose must withstand at least 40,000 deformations

• A threshold and doorpost impact test of at least 500 cycles.

Spare Parts Supply. The applicant undertakes to ensure spare parts supply for appliance

repair for at least 8 years from the time that production ceases. Spare parts are those

parts which, typically, may break down within the scope of the ordinary use of a product -

whereas those parts which normally exceed the average life of the product are not to be

considered as spare parts. Also, the applicant undertakes to provide after-sales services.

The product documentation shall include information on the above requirements.

Currently there is only one vacuum cleaner awarded the Blue Angel Ecolabel

279

and this is

considered the BAT regarding resource efficiency. Note that no disassembly requirements

are included in the requirements which probably is due to the shredding at end of life.

Instead the focus is on maintainable design, durability and the spare parts. These are all

factors that can improve the lifetime of products. The impacts of an improved lifetime

should be thoroughly assessed in later task to determine the possible trade-offs between

improved lifetime and energy efficiency.

10.3 Products

Based on the sections above, the average technologies and Best Available Technologies

(BAT) for each main product type will be examined in this section. A suggestion for the

Best Not Available Technologies (BNAT) is also given for each type.

279

https://www.blauer-engel.de/en/products/home-living/staubsauger

199

Mains-operated household vacuum cleaners

Average technology

This category includes mains-operated cylinder, upright and mains-operated handstick

(also called ‘compact’) vacuum cleaners, which are all covered by the current Ecodesign

Regulation and the annulled Energy Labelling Regulations. These types have different form

factors leading to different ergonomic advantages and disadvantages:

• Lightweight vs. heavy

• Lightweight but noisy

• Lightweight as a whole product but not easy to handle

• Heavy but easy and versatile to handle due to hose plus cleaning head

• Easy to store versus taking up a considerable storage space and time to set up

• Standard equipped with a sturdy agitator in the cleaning head (‘active nozzle’) and

a secondary hose for non-floor cleaning tasks, etc.

These differences co-exist and serve different audiences with different preferences.

However, for the purpose of setting Ecodesign requirements and Energy Label class limits

these differences are not decisive for the current or, for that matter, a possibly revised

regulation.

In section 8.5 the mains-operated vacuum cleaners were the prime subject as regards

performance data. For the BAU (Business as Usual) reference year 2016 the average

performance values are:

• Energy consumption AE of 38 kWh/year

• Power input P

eff

881 W

• Hard floor cleaning dpu

is 1.08

• Carpet cleaning dpu

is 0.81

• Dust re-emission d

is 0.3%

• Average (linear) sound power 80 dB(A)

BAT

In section 10.1 several options for improvement at component level were suggested, which

ultimately lead to the Best Available Technology (BAT). As regards the consequences for

the vacuum cleaner performance, the study team did its desk research of manufacturer’s

sites, Swiss Topten

280

, consumer associations (see Annex E), met with manufacturers in -

and outside the stakeholder meetings. The conclusion is that for mains-operated household

vacuum cleaners there are models in the highest energy label classes

281

for energy

efficiency (A+++)

282

and performance classes (A)

282

, but never for the same model. For

example, the Electrolux PURED9 GREEN model has an energy class A+++

282

(9.9

280

www.topten.eu

281

According to the the previous, annulled Energy Labelling Regulation

282

According to the previous, annulled Energy Labelling Regulation

200

kWh/year, 350W) but a carpet cleaning class of ‘C’

283

282

. Of the same model there is also

a version DELUXE with carpet cleaning class ‘A’

282

, but then the energy class is ‘A++’

282

(16 kWh/year, 400 W)

284

. The price of this new model is 400 € at the moment

285

This illustrates that there is a clear inverse relationship between carpet cleaning

performance dpu

and energy efficiency. This cannot be said about the hard floor cleaning

performance. Rather, every type of vacuum cleaner, even with very low suction power,

can get a good hard floor cleaning dpu

rating with the current crevice test. In the energy

efficiency rating of the general purpose vacuum cleaner, the most popular type, both the

dpu

and dpu

play an equal

role and the dpu

thus tends to ‘soften’ the inferior carpet

cleaning performance of some products.

The cleaning performance ratings of the annulled energy label are not always in sync with

the findings of consumer associations, who generally perform also debris (rice, lentils) and

fibre pick-up tests (simulating pet hair). Especially for the latter, the ‘active’ nozzle is

reported to make a large difference, whereas in the standard carpet tests the ‘passive’

nozzle is performing just as well.

For all these reasons, energy efficiency is to be seen in conjunction with performance.

As regards recycling the ‘PURED9 GREEN’ model is best-in-class with 70% recycled content

of the plastics in the product and 100% recycled materials (cardboard and PE) for the

packaging. The product weight (bare) is 7.09 kg, which is at the level of the base case. On

sound power, the score is 67 dB(A), comparable to e.g. the Rowenta Silence Force Compact

286

, but less quiet than the Miele C3 Silence EcoLine - SGSK3 at 64 dB(A) and the Bosch

In’genius Prosilence

287

at 59 dB(A). Handstick models perform better on material

consumption than the larger mains-operated types (cylinder and uprights) due to the lower

product weight of around 3 kg

288

BNAT

The Best Not yet Available Technology (BNAT) is a vacuum cleaner in the highest label

performance class for all aspects, i.e. a model with A+++, A, A, A (according to the

annulled Energy Labelling Regulation) and a sound power of 59 dB(A) or lower. As indicated

above, there are models that are almost there, but turning the energy class A++

289

into

an A+++

289

or achieving an A in carpet cleaning performance at energy class A+++

289

283

https://www.electrolux.ch/de-ch/vacuums-home-comfort/vacuum-cleaners/vacuum-cleaners/vacuum-cleaner/pd91-green/

284

https://www.electrolux.fr/vacuums-home-comfort/vacuum-cleaners/vacuum-cleaners/vacuum-cleaner/pd91-8ssm/

285

But streetprice will usually drop after the novelty wears off.

286

https://www.rowenta.be/nl/Schoonmaken/Stofzuigers-met-zak/SILENCE-FORCE-COMPACT-4A%2B-

RO6371EA/p/2211400326?gclid=CjwKCAjworfdBRA7EiwAKX9HeC_jr_gRPP_8hP367723L78YGdKhxjGDp8zzsM8ESoQ9iPBceOhsO

BoCVGwQAvD_BwE

287

https://www.coolblue.be/nl/product/772255/bosch-in-genius-prosilence-

bgb8a32w.html?ref=410179&gclid=CjwKCAjworfdBRA7EiwAKX9HeBXPT7GQPoPfmqyO5gk4RCxFrNPW_z9toOtCArfOtJxifzXHs6K

P_hoCKGsQAvD_BwE#product_specifications

288

E.g. Kärcher VC5

289

According to the previous, annulled Energy Labelling Regulation

201

might turn out to be very difficult, especially - depending on the final decision making - if

the testing becomes more realistic and more challenging, e.g. including debris pick-up for

hard-floor and fibre pick-up for carpet cleaning.

As regards circular economy there is a matter of opinion: are lightweight, compact

solutions that use fewer virgin plastics and metals to begin with to be preferred, or is a

current weight vacuum cleaner with high recycled content considered better? In the first

case, building on the corded handstick of 3 kg, of which e.g. 2 kg of virgin plastics, is

probably the way forward. In the second case, a 7 kg cylinder type with 70% recycled

content of the 5 kg of plastics is setting apart only 1.5 kg of virgin plastics. Or are both

strategies equally valid?

Table 71. Base case 1: Household mains-operated vacuum cleaners’ energy, performance,

price

BAU

BAT

BNAT

2016

2018

2025

2030

Rated power

900

300

dpu

0.81

0.91

dpu

1.08

1.11

AE (kWh/year)

33.6

33.7

37.0

36.6

9.9

9.5

Price incl. VAT, €

123

115

113

380

430

Table 72. Base Case 1: Household mains-operated vacuum cleaners’ materials (product life 8

years, package 0.08 m³)

Life Cycle materials

Production

Use

End of life

Impacts per product

Virgin + recycled

Only recycled

Disposal

Recycle

Recover

Materials

Bulk Plastics

3,643

911

1,129

1,093

1,457

TecPlastics

638

198

192

255

Ferro

863

345

820

Non-ferro

850

340

808

Electronics

Misc.

734

661

255

479

Auxiliaries

640

Total weight

290

6,784

2,271

708

2,353

3,419

1,720

Commercial vacuum cleaners

Commercial dry vacuum cleaners are typically used for cleaning offices, shops, restaurants

and hotels. They are not of the wet & dry barrel type, excluded from the scope of the

290

Average weight of one appliance

202

regulation, that is typically used to clean workshops and industrial premises and is able to

pick up liquids when necessary.

Commercial dry vacuum cleaners are generally not very different from household vacuum

cleaners, except that they usually have a sturdier construction and larger receptacle (8-15

litres) allowing them to operate for 300 hours per year, i.e. 6 times more than household

vacuum cleaners.

Having said that, there are some exceptions, such as the Nilfisk that is a cordless 10 kg

cylinder vacuum cleaner primarily designed for commercial purposes. It comes with 2

battery-packs. Each pack, recharges in only a few hours and allows the vacuum cleaner to

operate for 40 minutes at 600 W. Thus, in practice, an operator can operate at least for 80

minutes without interruption at maximum power, take a short break and start again. There

are also commercial vacuum cleaners with a backpack, corded and cordless.

The table hereafter shows two canister type models (VP930, VP300), one backpack vacuum

cleaner (GD10 BACK corded, but ‘backvacs’ are also available as cordless) and a very

efficient cylinder model VP600 that is also available in a cordless version

291

. It shows

performance and energy values comparable (or better) than the household types.

Figure 51. Commercial, cordless, backpack vacuum cleaner (source: Hoover)

Table 73. Nilfisk commercial cylinder vacuum cleaner examples (source: Nilfisk.com, Sept.

2018)

Product NILFISK

VP930

ECO

HEPA A++

VP300

GD10 BACK

VP600 ECO

HEPA

VP600

BATTERY

price (Nilfisk-shop.nl, sept.

2018)

299 euro

189 euro

659 euro

469 euro

~1000

euro

power (W) in 2 settings battery

190/465

Rated power (W)

400

600

780

330/550

650

291

Commercial cordless VCs are not proposed to be in scope here, but shown for information

203

Airflow (l/sec.)

25.5

24/28

21.7/26.7

Weight (kg)

7.9

5.2

Vacuum at nozzle (kPa)

13.4

15/18

Dust bag capacity (l)

Main filter area (cm²)

2400

1250

2400

Suction power end of tube (W)

120

112

225

75/155

45/116

Length x width x height (mm)

440x390x3

395x340x3

380x260x570

480x300x270

480x300x2

Product NILFISK

VP930 ECO

HEPA A++

VP300

GD10 BACK

VP600 ECO

HEPA

VP600

BATTERY

Cable length (m)/ plug type

15/EU

10/EU

15/EU

Sound pressure (dB(A) BS 5415)

Sound power (dB(A) IEC 704)

65.5

Protection class / ip protection

II / IP20

IP20

II / IP20

Main filter type

HEPA 13

Energy efficiency class

A++

Dust pick up on carpet

Dust pick up on hard floor

Dust re-emission class

Sound power dB(A) IEC/EN

60335-2-69

65.5

70/74

Annual energy consumption

(kWh/year)

Cable length (m)

Number of filters

N/A

Hose length (m)

1.9

N/A

Hose diameter Ø (mm)

N/A

Product NILFISK

VP930 ECO

HEPA A++

VP300

GD10 BACK

VP600 ECO

HEPA

VP600

BATTERY

Sound pressure DB(A) IEC/EN

60335-2-69

N/A

58/62

56/61

(@1.5m

ISO 11203)

Hepa filtration

H13 Exhaust filter

Two speed

N/A

yes

*=With battery :Li-ion, 36V, 7.8 Ah (-->280 Wh), 2.8 kg, charge time <40minutes

Table 74. Base case 2: Commercial mains-operated vacuum cleaners (BC2)

BAU

BAT

BNAT

2016

2018

2025

2030

Rated power

900

300

dpu

0.81

0.91

dpu

1.08

1.11

AE (kWh/year)

43.90

30.73

35.60

34.76

12.71

11.63

Price incl. VAT, €

331

380

430

The sturdy construction also is evident from the bill-of-materials as seen in Table 75.

204

Table 75. Base Case 2: Commercial mains-operated vacuum cleaner materials (product-life 5

years, package 0.1 m³)

Life Cycle materials

PRODUCE

USE

END OF LIFE

impacts per product

Virgin + recycled

only recycled

Disposal

Recycle

Recover

Materials

Bulk Plastics

5,795

1,449

1,796

1,739

2,318

TecPlastics

144

Ferro

1,436

574

1,364

Non-ferro

2,102

841

126

1,997

Electronics

Misc.

1,631

1,468

571

1,060

Auxiliaries

1,000

Total weight

11,110

4,332

1,111

3,625

6,204

2,392

Cordless handstick vacuum cleaners

Manually-operated household cordless vacuum cleaners are assumed to be used for the

same amount of total cleaning as mains-operated household vacuum cleaners However,

most cordless vacuums often would not have sufficient run time to be used for as long as

the mains-operated household vacuum cleaners, as most cordless vacuum cleaners have

a run time of 15-40 minutes while only a few can run for up to 60 minutes at the lowest

power setting

292

The capacity of a cordless is also smaller than that of a normal vacuum cleaner, i.e. in the

range of 0.2-0.8 litres compared with around 2-3 litres for an average-sized standard

vacuum cleaner according to Which?

293

. The same source also finds that, while a carpet

dust pick-up of 79% is average for a cylinder vacuum cleaner, the cordless vacuum cleaner

only reaches 47%. In other words, the average cordless would not meet the 2017

Ecodesign requirements for carpet cleaning (minimum dpu

75%) and possibly could only

enter as a hard-floor only model (minimum dpu

98%).

Especially over the last 5 years there has been a lot of progress in performance, battery

capacity and lifespan for cordless vacuum cleaners. Belgian consumer association Test-

Achats

294

tested 10 cordless handstick vacuum cleaners in 2013 and largely confirmed the

findings of Which?: Lower suction power, overall lower performance, limited battery

autonomy (between 9 and 33 minutes), recharging times between 3 and 17 hours

depending on the model. The overall conclusion was that the average cordless vacuum

cleaner has a lower performance than equivalent corded models. However, it should be

noted that the performance of the best cordless vacuum cleaners (only a few models) is

292

http://www.which.co.uk/reviews/cordless-vacuum-cleaners/article/corded-vs-cordless-vacuum-cleaners

293

https://www.which.co.uk/

294

Test-Achats 575, Aspirateurs Balais, Mai 2013, p. 38-39.

205

close to the performance of corded models. Finally, with a price varying between 118 and

380 € (average 196 €) Test-Achats found the product to be expensive. Weight of the tested

products varied between 2.4 and 3.9 kg (2.9 kg on average).

Five years later, published in Feb. 2018, the Stiftung Warentest (StiWa) again tested 10

cordless vacuum cleaners and found 2 models, Bosch Athlete and Dyson V8, to currently

have a ‘satisfactory’ cleaning performance compared to corded alternative. Stiwa does not

specify a virtual (because not compulsory) label classes, but a carpet cleaning performance

class of at least ‘C’ for these two models is not unlikely. The other 8 of 10 models were still

judged to be disappointing. See Annex E. There is also more variation in form factors than

5 years ago. There are now models where the motor (and receptacle) is in the middle, at

floor level and at the top (hand-level), as shown in Figure 52).

Figure 52. Examples of form factors for cordless stick models

(a) Hoover FE144LG011, 14.4 V NiMH-battery (estimated capacity study team 32Wh), runtime 25

minutes (estimated motor power input 76W), charges in 12h, bagless (cyclone technology), 2 speed

sections, bin 0.6 L. Street price (BE) 119 €

295

(b) Gtech AirRam MK2,’upright’, 22V, 2Ah (44Wh--> estimate study team 60-70W motor input), 3h

loading, 3.2 kg, uses washable filter box (reusable), telescope handle, www.gtech.co.uk

output (max. setting), 3.5h loading, 2.68kg product weight, bin 0.76 L, runtime 7(at max power)-

60(at minimum power) minutes, street price 629 €

296

Recently a cordless cylinder model from Nilfisk, designed both for the household (‘Family’)

and commercial sector, has also entered the market. The aim is to achieve a long run time

at high suction power without the user having to drag the full extra weight of 2-3 kg

batteries around.

295

https://www.unigro.be/nl/elektro-en-huishouden/stofzuigen-en-reinigen/stofzuigers/snoerloze-steelstofzuiger-hoover-

fe144lg011/1003079?channable=e50079.MTAwMzA3OS0tLU5M&gclid=Cj0KCQjw6MHdBRCtARIsAEigMxF1B17W6hQyzfcHFgb66

vYLUOxdR1Wn089vS9b__n7e21E6g79jwtIaAoLjEALw_wcB

296

https://www.dyson.be/nl-BE/stofzuigers/snoerloze-stofzuigers/dyson-v10/techniek.aspx

(a) (b) (c)

206

Note that the above models are all advertised (also) for carpet cleaning, i.e. as ‘general

purpose’. But there are also typical ‘sweepers’ and ‘electric broom’ types, with a form factor

as (b) in the figure above, i.e. a rotating brush without filtration and a 10-15 W suction

power

297

that is just enough to keep the dust from falling out of the small bin next to the

brush. If their performance allow, they could be in scope of a revised regulation as ‘hard-

floor only’. More sophisticated ‘hard floor only’ products are certain types that combine the

dry vacuum cleaning with a humid mop.

As the APPLIA database did not distinguish cordless vacuum cleaners specifically, data was

collected from retailers online for 27 cordless models from 16 different brands, which are

shown in Table 76. Note that not all data points were available at all retailers.

Table 76: Average data for cordless handstick cleaners collected from online retailers for 27

models from 16 brands.

Cordless vacuum cleaners

Average data

Max run time

34 min

Charging time

248 min

Motor power

241 W

Suction power

79 W

Battery voltage

30 V

Price

221 €

Bagless share

99%

Note that there is at least one manufacturer that offers a bagged cordless stick model

298

Table 77. Base case 3: Cordless vacuum cleaners’ energy, performance, price, 2018 data

Characteristics

BAU

BAT

BNAT

Maintenance consumption, charged and

docked [W]

2.6

1.0

0.5

Standby dock, when cleaning [W]

1.7

0.5

dpu

0.63

0.75

0.80

dpu

0.45

0.98

ASEc [Wh/m2]

0.59

0.56

ASEhf [Wh/m2]

0.57

0.56

AE [kWh/year]

21.88

20.14

19.55

Consumer price incl. VAT, €

221

500

630

The materials cycle is given in Table 78.

Table 78. Base Case 3: Cordless vacuum cleaners’ materials (product-life 6 years, package

0.05 m³, dock/charger included)

Life Cycle materials

PRODUCE

USE

END OF LIFE

297

E.g. https://www.gtech.co.uk/cordless-vacuum-cleaners/sw20-premium-cordless-floor-sweeper.html, featuring 7.2V battery

and a 60 minutes runtime.

298

https://www.gtech.co.uk/gtech-pro.html

207

Impacts per product

Virgin + recycled

Only recycled

Disposal

Recycle

Recover

Materials

Bulk Plastics

1,624

406

503

487

649

TecPlastics

287

115

Ferro

400

160

380

Non-ferro

835

334

793

Electronics

295

148

150

Misc.

Total weight

3,440

974

814

1,897

764

Robot vacuum cleaners

A robot vacuum cleaner is a self-propelling, cordless floor cleaning device capable of

determining its own trajectory in cleaning and in tracking its power-charger/docking

station. Consumer prices range from less than 100 Euro for models with low-end cleaning

and battery performance to 700-1000 Euros for models with best cleaning and battery

performance.

Manufacturers include:

− US robotics specialists such as iRobot (Roomba brand) and Neato

299

(Botvac,

Connected)

− European vacuum cleaner manufacturers such as Vorwerk (DE, e.g. VR200, also

owns Neato), Dyson (UK, 360 eye), Bosch (DE, Roxxter), Miele (DE, Scout)

− Asian vacuum cleaner manufacturers Samsung (Powerbot, Navibot), LG (Hombot),

Techtronics industries TTI (Dirt Devil, VAX, Hoover brands), Chiuwi (ILIFE)

− Chinese smartphone manufacturer Xiaomi

Figure 53 illustrates a typical high-end robot vacuum cleaner

300

. The geometry is typically

cylinder or D-shaped, diameter 34-36 cm, height 9-10 cm and includes a ‘bag-less' dustbin,

0.4 – 0.7 litre, HEPA filter, battery pack and the following active components

301

Motors

• 2 large drive wheels, independently driven (2xDC motor+gearbox), also drives main

brushes, spring-hinged (vertical object detection+ switch) and controlled

(tachometer for position feedback)

• 1 castor wheel, positioned through small DC motor (belt drive)

• 2 side-brushes each with DC motor

299

Recently acquired by Vorwerk

300

Note that the illustration is not an existing model but merely an illustrative drawing by VHK

301

Note that the list only present possible component and not a complete list. Robot vacuum cleaners may have less

components.

208

• 1 centrifugal backwards-curved fan, DC-motor driven (compare: PC cooling fan for

graphics card); turbo-compressor type and cyclonic dust separation is also found.

Sensors (optional)

• IR sensors (LED + receiver), side and cliff detection

• IR receivers for tracking docking station and/or virtual wall

302

• sensor to detect magnetic tape

• mechanical bumper ('keypad') sensors for collision detection

• piezo-electric sensor for dirt-detection

• tachometer for drive wheels

• drop sensor for drive wheels

• laser distance sensor or camera

• ultrasonic sensor

• gyroscope

• electronic compass

• fan speed control, including sensor

Printed circuit board

The Printed Circuit Board (PCB) of a high-end robot vacuum cleaner is similar to that of a

low-end laptop or smartphone. The latest model from Xiaomi contains a central processing

unit (CPU) in the form of an Allwinner R16 quad-core System-on-Chip (SoC), 512 Mb RAM

(Random Access Memory), 4 Gb flash memory (eMMC, embedded MultiMediaCard)

controlled by a 32-bit microcontroller unit (STM32 MCU) and a wireless (WiFi) module. The

SoC and STM are equipped with an UART (universal asynchronous receiver-transmitter)

for communication through a serial port. Also there is an UART for the LIDAR laser

rangefinder.

Other models, e.g. of the Roomba 650, also feature a PCB with a large inductor and big

capacitors. All other components on the PCB are small SMDs (surface mounted transistors,

diodes, etc.) and connectors for wiring to and from the active components.

Communication (optional):

• Remote control (battery driven controller)

• One or two push-buttons

• Display: LED-lit segments or LED-display

• Voice control

• Smart phone control: through WiFi (in HomeLAN) and/or Bluetooth

Peripherals (optional):

302

Virtual wall: Active perimeter control through battery-driven IR signal, e.g. between two 'towers' (3 x 1.5V alkaline

batteries, 6 months life); alternative is magnetic tape (passive control)

209

• Docking station with battery charger, IR transmitter (for the VC to find the way

home) and possibly electromagnetics to facilitate docking.

• Virtual walls and/or magnetic tape

• Additional cleaning aids (e.g. mops)

More information on the construction and reparability of robot vacuum cleaners can be

found on so-called ‘teardown’ and test sites

303

304

Figure 53: Robot vacuum cleaner (illustrative only, VHK 2018)

The cleaning algorithm, i.e. the pattern in which the robot moves across the floor, varies

from brand to brand and model to model. The pattern can be random or mapped following

a zig-zag, crisscross, or spiralling pattern

305

, or it can be controlled by simultaneous

303

https://robomow.jimdo.com/xiaomi-mi-robot-vacuum-saugroboter-test/

304

https://www.fictiv.com/blog/posts/the-great-robotic-vacuum-showdown-part-2-neato-xv-21

https://www.fictiv.com/blog/posts/the-great-robotic-vacuum-showdown-part-1-roomba-650-navigation-system

https://www.fictiv.com/blog/posts/the-great-robotic-vacuum-showdown-part-1-roomba-650-mechanical-system

305

https://www.vacuumcleanerbuzz.com/articles/how-does-a-robot-vacuum-cleaner-work/

castor wheel DC motor

drive wheel

gearbox & spring

DC motor for

drive wheel &

rotating brush

fan, BLDC motor

sidebrush, DC motor

rotating brush

castor wheel

PCB

battery pack

dustbin

35 cm

9 cm

Top view- section

Side view

Bottom view

LIDAR laser distance

sensor

IR cliff sensors

Gyroscope

Drop sensor

Tachometer

Fan speed-control/sensor

Ultrasonic sensor

Dirt sensor (piezo-electric in dustbin)

IR wall sensor

Mechanical (‘keypad’)

collision sensor

Electronic compass

Display and/or

pushbutton(s)

Top view Bottom view

210

localisation and mapping (SLAM), which requires more processing power (Figure 54 to

Figure 56).

For instance, the early Roomba models followed a combination of a “wall following” pattern,

where it drives along walls and a “random bounce” pattern, where it crosses the floor in a

straight line until it hits an obstacle and then moves away in a random direction. Newer

models use the SLAM technology, which uses slightly more power due to increased

processing power, but on the other hand has a much lower coverage time

306

. The

algorithms, no matter which model, should all ensure that every part of the floor is covered,

but it cannot be guaranteed depending on the shape and size of the room, and some places

might be covered multiple times. It is therefore not comparable to the 2 double strokes

assumed for manual vacuum cleaners

307

Figure 54: Robot cleaner using a random bounce pattern to cover the surface

Figure 55: Robot cleaner using a random + spiralling pattern to cover the surface

308

306

https://infoscience.epfl.ch/record/177726/files/vacuum-taros2012-camera-ready.pdf (same as

https://link.springer.com/chapter/10.1007/978-3-642-32527-4_12 )

307

https://www.cooksillustrated.com/articles/182-testing-robot-vacuums?incode=MCSCD00L0&ref=new_search_experience_3

308

Pictures from https://www.cooksillustrated.com/articles/182-testing-robot-vacuums

211

Figure 56: Robot cleaner using SLAM technology to map the room

No sources have indicated any use of bags for collecting dust in the robot vacuum cleaners.

Instead robot vacuums have a bin that must be emptied regularly: some suggest after

every run

309

, but it depends on the amount of dirt collected. Most robot cleaners have

changeable filters and moving brushes that should be cleaned regularly and changed (often

brush sets are available) when worn. It has not been possible to find any solid evidence of

how often brushes need to be changed, but based on anecdotal evidence, once a year was

assumed.

The top-three robot models in a recent German consumer test reveal a hard floor cleaning

performance almost as good as that of an average (150-200 Euro) cylinder vacuum

cleaner

310

, while carpet cleaning performance is only half as good in comparison (Figure

57). The dust-retention of a robot cleaners is considerably worse than that of a standard

vacuum cleaner. However, it should be noted that there is a difference in the standards

used for robot and for a standard cylinder vacuum cleaner, so the performance is not

directly comparable. Table 79 gives some general characteristics from a 2017 test by

Stiftung Warentest of 6 robot models.

309

https://taenk.dk/test-og-forbrugerliv/hus-og-have/robotstoevsugere/robotstoevsugere-fordele-og-ulemper

310

Note that the performance cannot be directly compared as there is no crevice test on hard floor for robots.

212

Figure 57: Dust pick-up for an average cylinder cleaner and the three best robot cleaners on

flat floor without crevice (source: Stiftung Warentest 2017).

Table 79: characteristics of 6 robot vacuum cleaner models (source Stiftung Warentest 2017)

Feature

No. 1

No. 2

No. 3

No. 4

No. 5

No. 6

Average

Price in EUR

725

980

495

545

360

525

605

W declared

Weight (kg)

4.2

3.9

4.1

4.4

3.5

2.9

3.83

Height (cm)

Width (cm)

Time programming

Boundary-limit

Magnet

Dual Mode

Magnet

Optional

Magnet

Charging station

Y & cable

Charging time (from

empty) in min

143

146

114

144

118

Operational when

charged in min

103

Price of battery in EUR

120

160

93.5

102

Price of main brushes*

or set in EUR

25*

19.9*

Price of filter in EUR

9.9

The high-end robot vacuum cleaners advertise 20 'Airwatts' suction power (qv 5-13

dm³/min and dP 1-1.8 kPa), which is only 5-18% of that of an average cylinder vacuum

cleaner (see Task 3). The relatively limited suction power is a key factor in the relatively

low dust retention performance.

Cleaning performance not only depends on suction power. Whereas most of the cylinder

vacuum cleaners have a 'passive nozzle', robot vacuum cleaners heavily rely on the use of

rotating brushes and other 'active' devices to pick up dust and fibres. Consumer association

100

Cylinder

Robot 1 Robot 2 Robot 3

Dust pick

-up (in %)

Cleaning Cylinder vs. Robot VC

(Stiftung Warentest, 2017)

Carpet dpu

Hardfloor dpu

213

tests show that many robot cleaners have problems cleaning tight corners and that

especially low-end models skip parts of the designated floor area. In those cases,

secondary (vacuum) cleaning will be needed. In any case, many manufacturers indicate

that their robot cleaners are only suitable for hard-floor and low-pile (<1 cm) carpet

cleaning.

Privacy and security aspects are important: robot vacuum cleaners are often linked to the

Internet, either via WiFi and/or via smartphone. They store the complete lay-out of the

home. Some types are even equipped with cameras. In short, there are many ways that

privacy can be invaded if proper security measures are not taken. More information can be

found in the media

311

Energy aspects

As regards the energy consumption of robot vacuum cleaners, the winner of the 2017

German test requires around 84 Wh for one recharge. A recharge takes about 3 hours, so

average power input is 28 W. At 200 cleaning cycles per year (see Task 3) this means

13.75 kWh/year. It depends on the mode, floor type and floor geometry, but for now it is

assumed that this gives 1 hour of operation. The manufacturer gives an average power

consumption during cleaning of 70 W

312

, which suggests a recharge efficiency of 80%. This

is in line with results from Vaussard et al. for Li-ion batteries, presented in Table 80 and

Table 81.

The off mode is when the switch is turned off, and the cleaner is not connected to power

313

The idle mode is when the robot is turned on, but not moving or vacuuming. The results

cannot be used directly in the energy consumption analysis as it measures energy drawn

from the battery, however, it does provide some valuable insights of the technologies used

in each of the robot cleaners.

In order to calculate the overall energy consumption of the robot vacuum cleaners, the

measurements of electricity supplied from the grid are used. Here three power modes are

identified: (1) the consumption of the base station only, which corresponds to when the

robot is vacuuming or otherwise away from the charging station while the power is still

plugged in. (2) the station + robot idle mode, which is when the robot is placed in the

charging station, but is fully charged. (3) Recharging mode, which is when the robot

311

http://www.zeit.de/digital/datenschutz/2017-12/34c3-hack-staubsauger-iot

312

The manufacturer indicates between 60 and 90 minutes operating time. Average power is 70W (50-60W fan, 10W brush, 2.5

W standby) in normal mode; 50W in Eco-mode (30-35W fan, 7W brush, 2.5W standby). The 70W during 1h versus the 84Wh

re-charge energy suggests a recharge efficiency of 80%. The Stiftung Warentest 2017 test indicates 47 minutes operation,

presumably at normal (non-ECO) mode.

313

No switch was available for robot 7.

214

returns from a cleaning task and the battery is charging. No wattage was stated for the

recharging mode, but the total recharging energy in kWh was given.

Table 80: Measurements of robot vacuum cleaner energy consumption when in use

314

, energy

from battery

Mode:

Unit

Robot 1

Robot 2

Robot 3

Robot 4

Robot 5

Robot 6

Robot 7

Off

[W]

0.0068

0.0087

0.064

0.0075

1.47

Idle

[W]

1.09

2.4

2.93

3.9

2.99

3.85

1.97

Cleaning

concrete

[W]

15.6

20.5

13.03

19.98

12.9

23.2

29.95

Cleaning

carpet

[W]

16.6

24.5

15.25

22.9

13.7

27.8

30.19

Recharge

efficiency

0.64

0.33

0.65

0.57

0.71

0.84

0.37

Technologies

Battery

Ni-MH

Li-ion

Ni-MH

Mapping

Random

CV-

SLAM

CV-

SLAM

CV-

SLAM

Laser

SLAM

Table 81: Measurements of energy consumption from electricity grid

315

Power mode:

Unit:

Robot

Base station

only

[W]

1.2

3.51

1.23

1.94

0.94

0.66

0.4

Station + robot

[W]

6.13

5.95

4.32

8.06

3.19

3.61

4.63

Recharge

energy

[kWh]

0.06

0.07

0.06

0.03

0.05

0.07

Cleaning time

[min]

158

202

104

107

102

The robot vacuum cleaners were not previously covered by the Standby Regulation, since

many models have maintenance charging in the Station + Robot mode, which could be

considered a primary function

316

. However, from January 2019 robot vacuum cleaners will

be subject to the networked standby requirements. The Base station only-mode could be

considered as a sort of standby, but since this might include energy for communicating

with the robot, neither this state is in scope of the Standby regulation. As seen from the

measurements big

differences exist for both modes and there is thus a large room for improvement. The

lowest consuming “base station only” mode is below the Standby Regulation requirement

of 0.5 W (robot 7), whereas the highest is 3.5 W (robot 2), which is a difference of a factor

7. For the station + robot mode (i.e. maintenance charging), the lowest consumption is

314

https://infoscience.epfl.ch/record/206269/files/EPFL_TH6522.pdf

315

https://link.springer.com/chapter/10.1007/978-3-642-32527-4_12

316

FAQ for the Standby Regulation

215

Robot 5 and 6, which were both equipped with Li-ion batteries. All other investigated

models had +4 W consumption in this mode, the highest being robot 4 with a consumption

of 8 W, more than double of the Li-ion models.

Manufacturer Vorwerk

317

measured the energy consumption of 6 robot types and confirms

the often high energy consumption when the robot is charged and docked. The graph below

compares daily energy consumption (in Wh) during charging after one cleaning cycle in

the IEC navigation room (see section 7) versus energy consumption during the time the

robot is charged and docked. The least energy efficient model (RUT3) consumed 90-95 Wh

for charging after a runtime of 23 minutes (implying power use during cleaning operation

of around 25 W

318

) and 195 Wh for 24 h at the docking station (8 W). The least energy

consuming models feature only ~50Wh for a day at the docking station (2W).

The following energy consumption is defined for the average Base Case and Best Available

Technology for robot vacuum cleaner

319

. Note that the AE calculation is based the

calculation method presented in task 3, and the dpu factors are based on test according to

the draft standard for robots, thus not directly comparable to dpu of mains operated

vacuum cleaners. In addition, the current performance is based on consumer test

organisations, products for sale online and inputs from stakeholders.

Table 82. Base Case 4: Robot vacuum cleaners’ Energy and performance

BAU

BAT

BNAT

Maintenance mode consumption, charged and docked [W]

3.7

2.0

0.5

Standby consumption, dock, when cleaning [W]

0.99

0.50

dpu

first pass

0.13

0.36

0.50

dpu

first pass

0.60

0.95

1.00

Cleaning cycle energy, carpet [Wh/cycle]

42.50

26.00

33.00

Cleaning cycle energy, hard floor [Wh/cycle]

42.50

26.00

33.00

Room coverage factor

83%

95%

Average AE (Kwh/y) – Based on test room

42.43

16.94

4.27

Average AE (Kwh/y) - no carpet

42.43

17.74

5.39

Table 83. Base Case 4: Robot vacuum cleaner materials (product-life 6 years, package 0.05

m³, dock/charger included)

Life Cycle materials

PRODUCE

USE

END OF LIFE

impacts per product

Virgin +

recycled

Only

recycled

Disposal

Recycle

Recover

Materials

Bulk Plastics

2,657

664

824

797

1,063

317

Presentation on energy consumption by Maike Brede (Vorwerk) at Suzhou IEC meeting, Oct. 2017. pers. comm. Vorwerk.

318

At recharging efficiency assumned 85% (typical LI-ion, for NiMH would be lower) plus docing station/charger use during

assumed 3h charging

319

Based on inputs from stakeholders

216

TecPlastics

337

104

101

135

Ferro

823

329

781

Non-ferro

568

227

539

Electronics

607

152

304

310

Misc.

Total weight

4,991

1,372

1,315

2,529

1,198

217

11. Task 5: Environmental and economic impact

In accordance with the MEErP methodology task 5 identifies the relevant base cases and

quantifies the current baselines in terms of economic and environmental impact for each

of the base cases. The economic impact is calculated as the life cycle costs of products for

the end-user, while the environmental impact is quantified in terms of energy and resource

aspects. The inputs for the calculations consist of the data presented in the previous tasks.

The calculations are performed with the ErP EcoReport tool, which is an Excel sheet

developed specifically to aid in the impact analysis of Energy-related Products

320

. All

calculations in this task is based on the year 2016, which is the latest year with sufficient

available data. The EcoReport tool includes a range of background data for calculating

impacts of different materials, distribution, and disposal methods.

The calculations in EcoReport tool are made for each of the following four base cases

identified for the purpose of this study:

Base case 1 (BC1): Mains-operated household vacuum cleaners

Mains-operated household vacuum cleaners are in principle the household products already

covered by the regulations, including cylinder, uprights and mains-operated handstick

vacuum cleaners.

Base case 2 (BC2): Commercial vacuum cleaners

Commercial vacuum cleaners are also covered by the current regulations, and are all

assumed to be mains-operated.

Base case 3 (BC3): Cordless

Cordless vacuum cleaners, as defined in task 1, are battery driven, manually handled

vacuum cleaners intended for floor cleaning, and are all assumed to be household.

Base case 4 (BC4): Robot vacuum cleaners

Robot vacuum cleaners are also battery driven, but can clean autonomously, not needing

the interference of a human being.

5.1 Inputs for baseline calculations

The inputs needed from the previous tasks to establish a baseline scenario for each base

case, is summarised in the following.

320

https://www.eup-

network.de/fileadmin/user_upload/Produktgruppen/Methodology_prep_study/MEErP_study_by_vhk/20110819_Ecoreport_2011

_MEErP.xls

218

Sales, stock and economic base data is all found in task 2, and is summarised in Table 84.

Table 84: Base case economic and market data for EcoReport, from task 2. All data is for 2016.

Description

Unit

Househol

d mains-

operated

Commercial

Cordless

Robot

From

section

Product Life

years

8.3

Annual sales

mln. Units/

year

27.69

3.27

7.39

2.00

8.2

EU Stock

mln. Units

248.74

18.43

28.01

9.48

8.4

Product price

€ / unit

122.53

306.71

220.86

344.99

8.7.2

Electricity rate

€ / kWh

0.196

0.163

0.196

8.7.3

Repair and

maintenance

costs

€ / unit

8.7.4

Bags and

filters

321

€ / unit

169

Discount rate

(interest minus

inflation)

8.7.1

Escalation rate

(projected

annual growth of

running costs)

8.7.1

Present Worth

Factor (PWF)

(years)

7.03

4.58

5.42

Automatically

calculated in

EcoReport

The present worth factor, which are automatically calculated in EcoReport is calculated by

the following formula:

   





Where:

• N is the product life

r is the discount rate minus the growth rate of running cost components (e.g. energy and

water rates)

The energy consumption inputs are derived from the use patterns and formulas in task 3

and the technical product data from task 4. For all calculations the data purchased from

321

Based on the use of 2 bags and 0.5 filter per year over the lifetime for domestic mains operated vacuum cleaners and

commercial vacuum cleaner. Cordless and robots are assumed to use two filters over their lifetime.

219

GfK is used. The derived average energy consumption for each base case 2016 is shown

in Table 85.

Table 85: Average annual energy consumption (based on AE values) for each base case in

2016.

Description

Average AE value,

2016

Presented in

section:

Household mains-

operated

33.66 kWh/year

10.3.1

Commercial

184.33 kWh/year

10.3.2

Cordless

21.88 kWh/year

10.3.4

Robot

42.43 kWh/year

10.3.4

In addition to the energy consumption during the use phase, the materials in the product

itself contain a considerable amount of embedded energy e.g. the energy used to mine the

raw materials and produce the finished materials. Some of this energy can be recovered

at end of life when products are either reused, recycled, or burned. When products are

landfilled this energy is lost. The necessary inputs are presented in Table 86.

Table 86: Inputs to calculate the environmental impacts and where they are presented

Description

Presented in section:

The material composition and weight of the materials for

the different vacuum cleaners

10.2.1, Table 69

Description of the manufacturing process and the values

used in the EcoReport tool

10.2.3 (description) and below in

this section (value used in

EcoReport tool)

The distribution phase and values used in the EcoReport

tool (Volume of package during transportation.

Below in this section

Share and weight of materials send to re-use, recycling,

incineration and landfill at End of life

9.13.1, Table 64

The environmental impacts and commodity prices of gold,

copper and cobalt are:

Below in this section

The manufacturing process is assumed to be negligible or at least small compared to other

impacts. Furthermore, it is not possible to add or adjust values for the manufacturing

process itself. The only adjustable input in EcoReport regarding manufacturing is the

percentage of sheet metal scrap. The default value is 25%, which is kept. Changing this

value will only have a very limited impact on results, since this is not a widely used material

in vacuum cleaners.

220

The distribution phase is included in the calculations but have a very limited impact on the

overall analysis. This phase comprises the distribution of the packaged product and covers

all activities from OEM (Original Equipment Manufacturer) components to the final

customer. However, the only parameter that can be changed in EcoReport is the volume

of the final package. The volume of the packaged product from the preparatory study is

used in the current study. The volumes of the package for the different base cases are

assumed to be:

• Mains-operated household vacuum cleaners: 0.08 m

• Commercial vacuum cleaners: 0.1 m

• Cordless vacuum cleaners: 0.05 m

• Robot vacuum cleaners: 0.05 m

In addition to the impacts calculated with EcoReport, the economic value and

environmental impacts of selected raw materials are investigated. The needed inputs are:

• Gold: 250 GJ/kg, 22500 CO

-eq/kg

322

and 35150 euro/kg

323

• Copper: 50.9 MJ/kg, 2.7 CO

-eq/kg

324

and 5.9 euro/kg

325

• Cobalt: 130 MJ/kg, 100 CO

-eq/kg

326

and 5.9 euro/kg

327

11.1 Outputs from baseline calculations

For each base case the following environmental and economic impacts are calculated:

• Life cycle Impacts per product over its lifetime – one product

• Impacts of all appliances sold in 2016 over their lifetime – the sales in 2016

multiplied with the impacts of one product

• Impacts of all appliances in stock in 2016

All impacts are divided into five different life cycle phases

328

• The material phase: in this phase the weight of the materials is multiplied with the

LCA Unit Indicators

329

so the impacts of using the different materials can be

calculated.

• The manufacturing phase: the manufacturing phase describes the (OEM)

manufacturing of metals and plastics materials. The specific weights per process

are calculated automatically from the material phase.

• The distribution phase: this phase covers all distributing activities from OEM

components to the final customer.

322

http://ec.europa.eu/environment/integration/research/newsalert/pdf/302na5_en.pdf

323

Price assessed in November 2017 at: http://www.infomine.com/investment/metal-prices/gold/1-day-spot/

324

EcoReport tool

325

Price assessed in November 2017 at: http://www.infomine.com/investment/metal-prices/copper/1-year/

326

http://www.iaeng.org/publication/WCE2015/WCE2015_pp863-865.pdf

327

Price assessed in September 2018 at: http://www.infomine.com/investment/metal-prices/cobalt/1-week/

328

The lifetime and life cycle are different parameters. However, the lifetime of vacuum cleaners is included in the use phase of

the life cycle

329

see MEErP 2011 Methodology, Part 2

221

• The use phase: for the use phase, the average product life in years and the annual

energy consumption are multiplied with each other to calculate the energy

consumption during the whole lifetime.

• The disposal and recycling phase: these phases deal with the impacts of end of life.

In the recycling phase, the recycling of the different materials is credited, and a

negative value can appear (due to avoiding the production of new materials).

In addition to total energy consumption and greenhouse gas emissions, other impacts are

calculated in the EcoReport Tool. All the impacts over the product life cycle are presented

in Annex F for the different base cases. The impact categories are:

• Other Resources & Waste

o Total Energy (MJ)

▪ of which, electricity (MJ)

o Water – process (litre)

o Water – cooling (litre)

o Waste, non-hazardous/ landfill (g)

o Waste, hazardous/ incinerated (g)

• Emissions (air)

o GWP100 (kg CO

-eq.)

o Acidification (g SO

-eq.)

o Volatile Organic Compounds (VOC) (g)

o Persistent Organic Pollutants (ng i-Teq)

o Heavy Metals (mg Ni eq.)

o PAHs (mg Ni eq.)

o Particulate Matter (g)

• Emissions (Water)

o Heavy Metals (mg Hg/20)

o Eutrophication (g PO

)

All impacts are further divided in the different life phases of the product which are the

material phase, manufacturing phase, distribution phase, use phase, disposal phase and

the recycling phase.

Mains-operated household vacuum cleaners

The energy and global warming (GWP) impacts of mains-operated household vacuum

cleaners over a lifetime (8 years) are presented in Figure 58.

222

Figure 58: Total energy consumption and emission of CO

-eq of mains-operated vacuum

cleaners – the impact of one vacuum cleaner over a lifetime

The energy consumption in the use phase of mains-operated household vacuum cleaners

has decreased over the past years, but is still the greatest share of the energy consumption

in the life cycle with 71% of total energy consumption. The material phase is responsible

for 20% of the energy consumption. It should be noted, that if the lifetime of vacuum

cleaners decreases, the importance of the material phase will increase.

The energy consumption and greenhouse gas emissions are closely connected and there is

a high correlation between the parameters. For energy consumption in the use phase there

is a clear correlation between energy used and CO

emitted. However, for materials the

total energy consumption and emitted CO

differs depending on the material. For

household mains-operated vacuum cleaner, the use phase is responsible for 67% of the

global warming potential (GWP) due to emission of greenhouse gasses.

Some of the use phase impacts are caused by the use of bags. For mains-operated

household vacuum cleaners the impact of the bags over a lifetime is based on the use of 2

bags

330

and 0.5 filter per year over the lifetime of 8 years, and an average weight of each

bag of 0.04 kg and each filter of 0.0017 kg, which gives approximately:

o 11 MJ of total energy consumption, responsible for approximately 0.3% of

the energy used

o 0.6 kg CO

-eq emitted, responsible for approximately 0.4% of the emitted

-eq

Besides total energy consumption and emission of CO

eq, other impacts are calculated in

the EcoReport Tool. All the impacts over the life cycle are presented in Annex F. Here it is

visible that the use phase has the highest impact in 6 out of the 15 impact categories, and

the material phase has the highest impact in 8 of the impact categories.

330

For the average domestic mains operated vacuum cleaner

223

Commercial vacuum cleaners

The environmental impacts of commercial vacuum cleaners over a lifetime (5 years) are

presented in Figure 59.

Figure 59: Total energy consumption and emission of CO

-eq of commercial vacuum cleaners –

the impact of one vacuum cleaner over a lifetime

Commercial vaccum cleaners have a shorter lifetime than household vacuum cleaners, but

commercial vaccum cleaners are used for more hours. This means the the use phase is

connected with the highest energy consumption in the life cycle of commercial vacuum

cleaners. The use phase is responsible for 90% of the total energy consumption in the life

cycle. The material phase is responsible for 7% of the energy consumption.

The energy consumption and the emission of CO

eq are closely connected. For commercial

vacuum cleaner, the use phase is responsible for 88% of the emission of CO

-eq. The

material phase is responsible for 7% of the emission of CO

-eq.

Some of these impacts are caused by the use of bags. For commercial vacuum cleaners

the impact of the bags over a vacuum cleaner’s lifetime is based on 10 bags

331

and 0.5

filter per year over the lifetime of 5 years, and an average weight of each bag of 0.04 kg

and each filter of 0.0017kg, which gives approximately:

o 14 MJ of total energy consumption, responsible for approximately 0.1% of

the energy used

o 0.8 kg CO

-eq emitted, responsible for approximately 0.2% of the emitted

-eq

Besides total energy consumption and emission of CO

eq, other impacts are calculated in

the EcoReport Tool. All the impacts over the life cycle are presented in Annex F. Here it is

331

For the average commercial vacuum cleaner (50% bagged)

224

visible that the use phase has the highest impact in 8 out of the 15 impact categories, and

the material phase has the highest impact in 6 of the impact categories.

Cordless vacuum cleaners

The environmental impacts of commercial vacuum cleaners over a lifetime (6 years) are

presented in Figure 60.

Figure 60: Total energy consumption and emission of CO

-eq of cordless vacuum cleaners –

the impact of one vacuum cleaner over a lifetime

Cordless vaccum cleaners have the second lowest overall impacts of all vacuum cleaners,

as most cordless vacuum cleaners are lightweight (few materials) and have a lower energy

consumption in the use phase. However, cordless have a high energy consumption in

maintenance mode. The use phase of cordless vacuum cleaners is connected with the

highest consumption of energy in the life cycle. The use phase is responsible for 69% of

the total energy consumption in the life cycle. The material phase is responsible for 26%

of the energy consumption.

The energy consumption and the emission of CO

eq are closely connected. For cordless

vacuum cleaners, the use phase is responsible for 63% of the emission of CO

-eq. The

material phase is responsible for 30% of the emission of CO

-eq.

Besides total energy consumption and emission of CO

eq, other impacts are calculated in

the EcoReport Tool. All the impacts over the life cycle are presented in Annex F. Here it is

visible that the use phase has the highest impact in 5 out of the 15 impact categories, and

the material phase has the highest impact in 10 of the impact categories.

Robot vacuum cleaners

The environmental impacts of commercial vacuum cleaners over a lifetime (6 years) are

presented in Figure 61.

225

Figure 61: Total energy consumption and emission of CO2-eq of robot vacuum cleaners – the

impact of one vacuum cleaner over a lifetime

Robot vaccum cleaners have the second highest life cycle impacts of all vacuum cleaners,

as robot vacuum cleaners use a high amount of energy in the maintenance mode and also

contains a high amount of PCBs. The use phase of robot cleaners is connected with the

highest consumption of energy in the life cycle. The use phase is responsible for 53% of

the total energy consumption in the life cycle. The material phase is responsible for 40%

of the energy consumption.

For robot vacuum cleaners, the use phase is responsible for 47% of the emission of CO

eq. The material phase is responsible for 44% of the emission of CO

-eq.

Besides total energy consumption and emission of CO

eq, other impacts are calculated in

the EcoReport Tool. All the impacts over the life cycle are presented in Annex F. Here it is

visible that the use phase has the highest impact in 4 out of the 15 impact categories, and

the material phase has the highest impact in 11 of the impact categories.

EU Totals – Environmental impacts

The EU totals are the environmental impacts aggregated to EU-28 level. For the EU totals

the following is calculated:

• Environmental impacts during the entire life cycle of vacuum cleaners sold in 2016

is calculated by multiplying the annual sales with the impacts of each of the base

cases and presented in Table 87.

• Environmental impacts of vacuum cleaners (EU-28 stock) is calculated by

multiplying the current stock with the impacts of each of the base cases and

presented in Table 88.

Table 87: Environmental impacts during the entire lifetime of vacuum cleaners sold in 2016

Materials

Household mains-

operated

Commercial

Cordless

Robot

Total

226

Bulk Plastics (kt)

TecPlastics (kt)

Ferro (kt)

Non-ferro (kt)

Electronics (kt)

Misc. (kt)

Total weight (kt)

Total Energy (PJ)

162

of which, electricity (PJ)

131

Water (process) (mln.m

)

Water (cooling) (mln.m

)

Waste, non-haz./ landfill*

(kt)

130

Waste, hazardous/

incinerated* (kt)

GWP100 (mt CO

-eq.)

Acidifying agents (AP) (kt

-eq.)

Volatile Org. Compounds

(kt)

Persistent Org. Pollutants

(g i-Teq.)

Heavy Metals (ton Ni eq.)

PAHs (ton Ni eq.)

Particulate Matter (kt)

Heavy Metals (ton Hg/20)

Eutrophication (kt PO

)

‘The combined energy consumption of all vacuum cleaners sold in 2016 will amount to 162

PJ during their lifetime resulting in 7 Mt CO

eq emitted. The highest impacts are connected

with mains-operated household vacuum cleaners as they have the highest annual sales.

In Table 88 the annual impact of all vacuum cleaners (impacts by the stock for one year)

is calculated which allows for comparison with the EU totals from all energy-related

products (values for EU is a part of the EcoReport Tool).

Table 88: Annual environmental impacts of vacuum cleaners (EU-28 stock)

Materials

Household

mains-operated

Commercial

Cordless

Robot

totals

Plastics (Mt)

0.120

0.020

0.014

0.006

Ferrous metals (Mt)

0.024

0.005

0.003

0.002

206

227

Non-ferrous metals (Mt)

0.024

0.007

0.006

0.001

Other resources & waste

Total Energy (PJ)

159

75697

of which, electricity (TWh)

2800

Water (process)* (mln.m

)

247000

Waste, non-haz./ landfill* (Mt)

0.12

0.03

0.02

0.01

2947

Waste, hazardous/ incinerated*

(kton)

0.00

Emissions (Air)

GWP100 (mt CO2-eq.)

5054

Acidifying agents (AP) (kt

SO2eq.)

22432

Volatile Org. Compounds (kt)

8951

Persistent Org. Pollutants (g i-

Teq.)

2212

Heavy Metals (ton Ni eq.)

5903

PAHs (ton Ni eq.)

1369

Particulate Matter (kt)

3522

Emissions (Water)

Heavy Metals (ton Hg/20)

12853

Eutrophication (kt PO

)

900

The annual energy consumption of all vacuum cleaners (in the stock in 2016) in EU-28 is

calculated at 233 PJ which leads to 10.5 Mt CO

-eq released to the atmosphere. This means

that vacuum cleaners are responsible for 0.3% of the energy consumption (0.79% of the

electricity consumption) and 0.21% of the CO

-eq in the EU.

11.2 Consumption of critical raw materials and other materials of high

importance

The consumption of critical raw materials (cobalt) and the consumption of other materials

with high importance (gold and copper) are determined for each of the base cases. The

amount of cobalt, gold and copper are calculated and the derived impacts regarding

energy, emission of CO

-eq and market value in euros are presented in Table 89.

Table 89: The amount of cobalt, gold and copper and the derived impacts regarding energy,

emission of CO2-eq and market value in euros per product

Base case

Resource

Kg CO

-eq

Euro

BC 1

Gold

0.02

4.13

0.37

0.58

228

Copper

307.00

15.63

0.83

1.81

Cobalt

BC 2

Gold

0.001

Copper

766.00

38.99

2.07

4.52

Cobalt

BC 3

Gold

0.09

Copper

354.50

18.04

0.96

2.09

Cobalt

12.00

1.56

1.20

0.61

BC 4

Gold

0.18

45.53

4.10

6.40

Copper

224.00

11.40

0.60

1.32

Cobalt

20.00

2.60

2.00

1.02

Cobalt, copper and gold have limited impacts compared with the impacts of the use phase

of vacuum cleaners. Copper is responsible for less than 1 % of the emission of CO

-eq over

a lifetime and gold and cobalt has an even lower impact. The robotic vacuum cleaner has

the highest contest of printed circuit boards, the biggest battery and thus the highest

content of gold and cobalt. Even the “high” content, the combined value of the gold and

cobalt in the robotic vacuum cleaner, is limited.

The impacts of the mentioned raw materials are also aggregated to EU-28 level. For the

EU totals the following is calculated:

• The EU consumption of critical raw materials in vacuum cleaners is calculated by

multiplying the current stock with the amount of cobalt, gold and copper in each of

the base cases and presented in Table 90.

Table 90: The amount of cobalt, gold and copper and the derived impacts regarding energy,

emission of CO2-eq and market value in euros for the total stock of vacuum cleaners

Base case

Resource

Tonne

Million euros

BC 1

Gold

1.05

94136

147

Copper

77845

3.96

210180

459

Cobalt

0.00

BC 2

Gold

0.01

0.00

286

Copper

16237

0.83

43840

Cobalt

0.00

BC 3

Gold

0.44

39260

Copper

6989

0.36

18871

Cobalt

237

0.03

23659

BC 4

Gold

0.29

25755

Copper

1408

0.07

3802

Cobalt

126

0.02

12572

229

The impacts of the raw materials are limited

332

compared to the other impacts imposed by

vacuum cleaners in the use phase. However, the value for the amount of cobalt, gold and

copper present in the EU stock are significant. The combined impact and the value of gold

and copper in all vacuum cleaners (stock) are presented in Table 91.

Table 91: The combined impact and value of gold and copper in all air conditioners (stock)

Total Energy (PJ)

GWP100 (t CO

-eq.)

Total (mln. €)

Gold

1.8

159437

249

Copper

5.2

276694

605

Cobalt

0.0

36231

Total

7.0

472362

872

Cobalt, gold and copper are accountable for an energy consumption of 7.0 PJ and an

emission of 0.47 Mt of CO2-eq. The combined value of copper, gold in the EU stock amounts

to 0.87 billion euros.

11.3 Life cycle cost

Based on these inputs EcoReport automatically calculates the Life cycle costs (LCC) with

the following formula:

     

Where:

• LCC is Life Cycle Costs

• PP is the purchase price of the vacuum cleaner

• OE is the operating expense and are the combined costs of electricity

333

, bags,

filters and the repair and maintenance.

• PWF (Present Worth Factor) is a formula described below and is based on the

concept of time value of money

334

• EoL is End of life costs (disposal costs, recycling charge) or benefit (resale) which

are assumed to be negligible.

The life cycle cost is thus the cost experienced by the user, when purchasing a vacuum

cleaner in 2016, with discounted energy prices throughout the lifetime of the product. The

life cycle costs of the four different base cases is calculated in the EcoReport tool and

presented in Table 92.

332

Taking environmental impacts beyond energy and GWP into account, raw materials are connected to very severe

environmental and health issues (gold: use of mercury; copper: acid mine drainage, water contamination in mining etc.)

though these aspects are difficult to assess with MEErP methodology.

333

The service rate is solely used for the commercial vacuum cleaners; thus, it is assumed that all household vacuum cleaners

are used in households and all commercial vacuum cleaners are used at service premises.

334

Time value of money is the idea that an amount received today is worth more than if the same amount was received at a

future date.

230

Table 92: Life cycle costs of the three base cases (VAT included)

Household

mains-operated

Commercial

Cordless

Robot

Product price (€)

123

307

221

345

Electricity (€)

137

Repair & maintenance costs (€)

Bags and filters

Total (€)

244

568

347

501

For all base cases the highest expenses are connected with the purchase of the vacuum

cleaner. Commercial vacuum cleaners have the highest expenses in the use phase, which

is due to the extensive use of these vacuum cleaners. Cordless vacuum cleaners have the

second highest expenses in the use phase, however it is approximately on par with

household mains-operated vacuum cleaners and robots. Over a lifetime the mains-

operated household vacuum cleaner has the lowest cost followed by cordless vacuum

cleaners. The life cycle cost for each of the base cases is also aggregated to EU-28 level.

For the EU totals the following is calculated:

• Annual consumer expenditure in EU-28 is calculated based on the life cycle costs

per product multiplied by the annual sales and presented in Table 93.

The annual consumer expenditures in EU-28 of the different base cases are presented in

Table 93. The product price and installation costs per product is multiplied by annual EU

sales to arrive at the annual consumer expenditure for EU28. The lifetime electricity costs

per product multiplied by the annual EU stock and divided by the lifetime to arrive at the

annual consumer expenditures for electricity in the EU-28, the same is done for repair &

maintenance costs.

Table 93: Annual consumer expenditure in EU28

Household

mains-operated

Commercial

Cordless

Robot

Total

Product price (mln. €)

3393

1003

1632

690

6718

Electricity (mln. €)

1667

552

254

2551

Repair & maintenance

costs (mln. €)

656

113

211

117

1097

Bags and filters

2012

386

188

2662

Total (mln. €)

7728

2054

2285

961

13028

The highest costs are related to mains-operated household vacuum cleaners which have

the highest annual sales. As the table above shows, every year EU consumers are spending

almost 13 billion euros on purchase and operation of vacuum cleaners. Approximately 20

% (2.6 billion euros) are related to electricity expenses.

231

12. Task 6: Design options

According to the MEErP, Task 6 builds on the Base Case models described in Task 5 to

identify design options, assess their costs and benefits, determine the combined impact of

clusters of options that give the Least Life Cycle Costs (LLCC), the Best Available

Technology (BAT) and the Best Not yet Available Technology (BNAT). Note that there is

not enough information on commercial vacuum cleaners to make an independent cost-

analysis. Hence, we will assume similar costs as for the household models, but with a

higher mark-up for extra sturdiness and higher retail costs.

For materials efficiency we will, in line with the Waste Directive hierarchy, look at cost-

effective individual options to Reduce, Re-use, Recycle, Recover and Remove (the 5Rs),

which (under Reduce) includes durability aspects.

According to Article 15(6) of the Ecodesign Directive 2009/125/EC, and also taking into

account the boundary conditions stipulated in Article 15 (5), the Task 6 assessments are a

basis for possibly setting more stringent and/or new Ecodesign requirements. Furthermore

it will be the basefor rescaling the energy label classes in accordance with the Energy Label

Framework Regulation, Regulation (EU) 2017/1369.

In section 10 the various technologies and design options for components were presented

in section 10.1, including possibilities to improve the circular economy aspects in section

10.2. In section 10.3 the energy, performance and price characteristics were given for BAU

(Business as Usual, starting from current average), Best Available Technology (BAT) and

Best Not yet Available Technology (BNAT) relating to four Base Cases:

• Household mains-operated vacuum cleaners (BC1)

• Commercial vacuum cleaners (BC2)

• Cordless vacuum cleaners (BC3)

• Robot vacuum cleaners (BC4).

The current state of the material flow over the life cycle was given in section 10.

As such, most of the quantitative basis for the design options in Task 6 is available. This

section will be limited to identifying/describing the design options, present additional

information where information is lacking and facilitate the impact assessment for Task 7.

The design options will be presented per Base Case.

12.1 Household mains-operated vacuum cleaners (BC1)

The following design options for this category were identified and investigated:

232

Option 1: More stringent energy requirements

Investigating more stringent Ecodesign requirements on energy and more ambitious

energy class categorisation is the default first step with the review of the regulations.

However, while the energy consumption during the use phase is still the most important

impact for most environmental impact categories (global warming, acidification, etc., see

section 11), the current Ecodesign and Energy Labelling Regulations for vacuum cleaners

have been very effective in reducing the average power from around 2200 W before the

2014 measures, to 900 W or less since the second tier in 2017. The Dutch consumer

association Consumentenbond mentions that replacing an average 2013 model (at 165

kWh/year) by a new 2018 model (at just below the limit of 900W or 43 kWh/year) saves

as much as 26 €/year on the electricity bill

335

As mentioned in section 10, the average power input is now as low as 700 W. Section 8.5.1

indicates that only 7.5% of 2018 models has a rating between 800 W and 900 W. Assuming

that these models would, once eliminated, ‘return’ to the population at just below the limit,

a limit at 800 W gives EU energy savings in 2030 of lower than 0.1 TWh

336

. The extra

product price that the 7.5% of current consumers would have to pay to get this 11% saving

can be estimated from the difference between BAU 2016 (900W, 38 kWh, 122 €) and BAU

2018 (700 W, 30 kWh, 170 €). This comes down to a difference 48 € for a saving of 200W

or 8.6 kWh/year. At 100 W these figures halve, so the consumer pays 24 € to get a 4.3

kWh/year saving during a product life of 8 years. This is a saving of 34.4 kWh over life; at

an electricity rate of 0.2 €/kWh in 2015 prices, this comes down to almost 7 € saving. Net

costs of this measure for the consumer are thus 24 €−7 €=17 €. For the EU, i.e. 7.5% of

the estimated 200 million households owning a mains-operated vacuum cleaner (15

million), the extra cost for those households would be 255 € million in around 2030 if the

measure is implemented in 2021-2022. Setting the level at a more ambitious level, e.g. at

750 W leading to a cut-off rate of around 30%, will only aggravate the situation.

In task 7 these projections will be elaborated with proper discounting, but it is easy to see

that from the perspective of Life Cycle Costs there is no monetary gain in setting more

stringent Ecodesign requirements for mains-operated vacuum cleaners.

Option 2: More realistic performance, indirectly better energy efficiency

This option aims at (indirectly) achieving better energy efficiency and functionality by

prescribing more realistic and challenging performance tests. As the dust pick-up (dpu

and dpu

) in those tests are a part of the formula for the annual energy consumption AE,

335

‘Strengere regels voor stofzuigers’, Consumentengids, July/Aug. 2018, p.26-29.

336

5 kWh/annually per unit for 7,5% of ca. 30 million sales accumulating over 8 years stock life→ 5kWh x 0,075 x 30 x 8 mln.=

90 mln. kWh=0,09 TWh electricity

233

tougher performance tests will increase the ambition level of the Ecodesign energy limits,

even if they are nominally kept at the same level.

It is proposed to add a debris test to the hard floor test, in addition to the current crevice

test, to seek more differentiation. Results from Round-Robin Tests are not yet available,

but it is assumed that this will lower the current dpu

values by at least 10 percentage-

points (0.1), because the nozzles have to be opened more up. It will prohibit some

manufacturers to continue to use special test-nozzles with full enclosure of the suction

area, just to get a better crevice performance, because such a nozzle would not work for

debris pick-up.

Likewise, it is proposed to add a debris pick-up test to the dust pick-up test for carpets.

Debris pick-up test is also being developed for carpet, but no results are ready yet. It is

important that such a test and test-conditions are formulated carefully so as not to have

unwanted side-effects. It is a known phenomenon that active nozzles have a superior

performance in that test over passive nozzles. On the other hand, active nozzles cost

energy and for people not having pet hair in their home, passive nozzles are seen as quite

sufficient for good cleaning. Hence, the requirement should not lead to additional

production and use of active nozzles.

In order to compensate for the negative impact on the cleaning performance, the products

need to be at least 10% more efficient, i.e. ‘virtually’ the power has to go from 700 W to

630 W. To go from 900 to 700 W (minus 22%) made the VC around 40% more expensive.

Assuming the same proportionality, to go from 700 to 630 W (minus 10%) makes the

vacuum cleaner 18% more expensive. Instead of 170 € the new average price would

become 200 €. Having said that, it also has to be taken into account that timing plays a

role, because price are decreasing –on average at 1% per year—due to learning effect,

larger production volumes, competition, etc. So, for a new measure to be implemented in

e.g. 2022, 5 years from now, the price impact is expected to be 10 € less: new price 190

€ in 2022.

Please note that, as an outcome of the extensive RR Tests, the industry is currently

undertaking in the context of standardisation activities, it is possible that a more real-life

dimension could be added with testing of several types of carpets. However, given the

large deviations in intermediate results between the different laboratories that the study

team has witnessed thus far, it is not very likely that this could be implemented in a legal

context of accuracy, reliability and repeatability.

234

Option 3: Recycled content and/or light-weighting

As mentioned in section 10, for a balanced circular economy it is important to address both

sides of the recycling balance: increased recycled content of materials in production and

increased recycling at the product’s end of life. Given that already a few manufacturers are

reaching a high recycled content of up to 70% for their plastics parts, demonstrating that

it is economical, it is plausible that Ecodesign measures set targets for a minimum recycled

content for the plastic parts, and/or that it could be included as a parameter on the energy

label. As discussed in section 10.2, for metals and electronics it is functionally unacceptable

to have contaminations that go with post-consumer recycled materials (even when small)

or the recycled content is either already very high (e.g. 85% for most die-casts).

There are a few problems to solve: first of all, how can the legislator implement and execute

control over any requirement on recycled content? The dispute has always been that either

there is a burdensome administrative route with a disproportionate burden for all

concerned or there should be very sophisticated lab-tests to track to contaminations and

loss of properties due to recycled content. And even then, there is the problem of

estimating a percentage of recycled content. Another option is to do a visual inspection of

vacuum cleaners, as recycled plastics, that can withstand mechanical loads often are

through-and-through black

337

. However, recycled plastic can be a mix of colours and virgin

plastic can be black. Hence, it is not possibly to rely on a simple visible inspection. A

practical solution could simply be statements (proof of paper) and calculations on the

content of recycled plastic according to the standard under development prEN 45557:2019,

however, this cannot used to ascertain the content.

A second factor is that it is not economically advantageous to circumvent a requirement to

use a reasonable fraction of recycled plastics: the pellets cost around half of pellets from

virgin material. The two plastics that constitute most of the plastics in vacuum cleaners

are PP and ABS.

In this case, and probably in more cases where injection moulded parts are used nowadays,

the solution might be simple for two reasons: first, and different from a few years ago,

recycled plastics have considerably lower costs than virgin material. Second, recycled

plastic granulates for injection moulding are almost without exception black. This does not

mean that the vacuum cleaners need to be black, using techniques such as in-mould

decoration (IMD, similar to in-mould labelling IML) the colour comes from a very thin but

scratch-resistant foil that forms an outer layer. Other, less frequently used, possibilities

are thermoformed inlays in injection-moulds or 2K (2-component) injection moulding.

337

For ejection moulded plastic which often are used for vacuum cleaner.

235

The table below shows that currently (September 2018), the recycled ABS and PP pellets

cost around half of the virgin material pellets. So even with the extra costs of colouring

techniques as described above and with possibly a bit more material due to differences in

mechanical properties, the use of recycled materials is economical and does not have to

lead to a higher cost.

Table 94 . Prices of plastic injection moulding grades

Material

Recycled

Virgin

Difference

EUR/kg

ABS pellets 2018

1.46

2.60

-78%

PP pellets 2018

0.89

1.77

-99%

PP pellets 2015 plastic recyclers Europe

0.90-0.95

1.43-1.50

-73%

source 2018: www.plasticsnews.com; conversion 1 lbs=0.4535 kg, 1 US $= 0.86 EUR

prices at annual volumes of 2 to 5 million lbs.

injection moulding grade pellets, typically colour black

source 2015: Plastics Recyclers Europe, Increased EU Plastic recyling targets:

Environmental, Economic and Social Impact Assessment, prepared by BIO, 2015

The graphs Figure 62 show that this was not always the case. In the period 2012-2015 the

price difference of the raw materials was only 25%.

236

Figure 62: Pricing history of recycled injection grade PP (above) versus virgin PP (below).

Source: www.plasticsnews.com , extract 2018)

237

Recycled plastics in acceptable quality are currently available only for bulk-plastics like PE,

PP, PS and ABS. This covers 90% of the plastics used in the average vacuum cleaner, but

of course elastomers (rubbers) and thermosets (e.g. polyester in the motor housing)

cannot be covered by recycled plastic. Also, the availability of recycled plastic is currently

high, but recycled plastic is somehow a limited resource. If the demand increases the

availability might decrease, but on the contrary, the economy of recycling may be improved

so more plastic is recycled. Currently, a share of 60-70% seems technically feasible also

for various form factors.

However, setting a simple percentage might be counter-productive, as it could lead to

industry making heavier products for no other reason than to meet the percentage. In

addition, it might penalize manufacturers that choose a different strategy to avoid using

virgin material, i.e. to use much less to make the products lighter. For instance, at the

moment 4.3 kg (3.6 kg bulk- and 0.6 technical) of plastics are used, of which 0.9 kg (22%)

are assumed to come from recycled plastics. Setting a recycled content target of 70% for

plastics would increase that number to 3.0 kg. So only 1.3 kg of virgin plastics will be used.

Together with the recycled content of metals and packaging (1.9 kg) this means that two-

thirds (66%) of the 6.8 kg product is made of recycled materials. It also means that 2.3

kg of virgin material is still required. If a 2.3 kg product, e.g. a corded stick, can achieve

the same energy efficiency and performance as a 7.1 kg product than this is not just equal

in avoiding environmental performance, but actually superior because also at the end of

life, there is only 2.3 kg to worry about in terms of disposal, recycling, etc. Achieving a

weight of only 2.3 kg probably requires the best possible material properties and might be

impossible to reach with a 70% recycled content target.

Rather than setting the requirement for a share of recycled content, it is technically better

to set a maximum target for the assumed virgin material, i.e. product weight minus

recycled content, and give a credit for the fact that there is no material loss for the avoided

kilos at end of life.

A monetary LCC calculation for this option is not necessary, because manufacturers are

assumed not to encounter extra costs when setting the target for plastics only. As a positive

impact it is assumed that per vacuum cleaner 2 kg extra of bulk plastics (assumed PP) will

come from recycled content. The recycled PP will also have its impact for collection and

recycling, as is calculated in the EcoReport, but much less.

Another option is to inform the consumers of the content of recycled plastic on the label.

However, recycled plastic can both be pre- and post-consumer recycled. It is assumed that

most plastic used in vacuum cleaners is pre-consumer recycled plastic as this fraction is

available in a higher quality. Post-consumer recycled plastic is more difficult to use in

238

consumer products as the quality is lower, and the risk of unwanted contaminants is higher.

This means that the general application of post-consumer plastic is limited and therefore,

it should be valued if manufacturers are able to include post-consumer recycled plastic in

their products. Based on the prEN 45557:2019 the following calculation method could be

used for calculating the content of pre- and postconsumer recycled plastic:















  



  















 



  

To value the greater challenge by using post-consumer recycled plastic and the

environmental benefits by avoiding downgrading or incineration of plastic it is suggested

to calculate a combined indication of the amount of content of recycled plastic with the

following weighing and formula:





















 











This means that a product made of 100% pre-consumer recycled plastic obtains a mark of

50%, and a product made of 100% postconsumer plastic obtains a mark of 100%.

However, although the annulled label contained a lot of information already, it is assumed

that most consumers are familiar with the sign for recycling. A conceptual drawing of the

mark is presented in Figure 63 below.

Figure 63: Conceptual drawing of a recycling sign

Option 4: Increase product life

Increasing product life is an option in the circular economy concept that aims to slow down

the materials cycle of products. For instance, if the lifetime of mains household vacuum

cleaners goes from 8 to 10 years, it is assumed that 25% less material resources will be

needed. That is to say, if there are no negative repercussions. They can become less

efficient due to wear and tear. Slowing down the introduction of new products in the market

will also slow down possible energy efficiency improvements and possible savings on

auxiliary materials. If the increase in product life requires the use of extra materials and/or

239

materials that represent an extra environmental burden, that has to be taken into account

in calculating the benefits.

At the moment, product life of household mains-operated vacuum cleaners is in the order

of 8 years. For commercial vacuum cleaners the expected life is 5 years and for cordless

and robots a life of 6 years is assumed.

There are a number of options to prolong the average product life:

• Increase the technical product-life of critical components such as the motor and

hose;

• Facilitate reparability by ensuring that spare parts are available for a sufficient time

after the production of a model stops (Blue Angel suggests 8 years) and that critical

parts are easy to replace;

• If possible through the design, promote re-use in the sense of refurbishing

338

, to

give the products a second life.

The options are described in more detail hereafter, but the estimated impact of the

measures would indeed increase the product life from 8 to 10 years. This means that

consumers holding on to their product for 2 years more will miss out on two years of energy

savings, compared to buying a more efficient product earlier.

But they will also postpone an investment decision of around 170 € for 2 years. At a going

rate of 5% for consumer loans and at current inflation of 1%, the interest on such a loan

would be 7 € per year. So the consumer is saving 13.20 € net by increasing product life by

two years. The monetary situation may change if the product life increase involves the

costs of a repair. However, in the 2016 JRC study discussed hereafter calculates that even

a repair of 22 € to prolong the product life can be economical, also taking into account the

technical impact.

As regards promoting re-use of the whole product, the possibilities of the regulator are

limited. What can be addressed is the re-usability of filters, as will be elaborated hereafter.

Increasing technical life

As discussed in the 2016 special review study there seems to be an opportunity for

increasing the durability of the motor to 550 hours without much extra costs, i.e. merely

a few euros to increase the size of the carbon brushes, while optimising also thermal and

mechanical design of the universal motor. The extra cost, in consumer prices is estimated

at around 2 €.

338

As mentioned in section 4, the fraction of VCs given away, or even sold for a small amount, to family, friends and charity is

not included in the definition of ‘re-use’. Refurbishing means checking/repairing/replacing all components and ensure a full

second life fort he product.

240

For the primary hose of a cylinder vacuum cleaner the 40,000 flexes in the current test

seems adequate, as only 7.7% of consumers experience technical issues with the hose

339

The hose is both an important and exposed part of the vacuum cleaner, subject to a lot of

wear and tear directly imposed by consumers and consumers have a large influence on the

durability of the hose. This means that if the requirements increase the consumers might

still experience faults on the hose due to mishandling. However, the hose should be

available as a spare part.

For the secondary hose of an upright vacuum cleaner it was agreed that a new test

procedure would be elaborated in the standard, but no concrete proposals are on the table

right now.

For cordless and robots, the battery lifetime has a great influence on the overall lifetime of

these appliances. However, no official measure for battery lifetime exists, but the computer

Ecodesign Regulation

340

has an information requirement of battery lifetime based on the

number of charging cycles it can last. The battery capacity falls over time with the number

of charging cycles, and the share of power drawn from the battery out of its total rated

capacity (also called Depth of Discharge, DoD) is crucial for the lifetime in terms of the

capacity left after a number of cycles. It is therefore recommended to set the requirement

according to a definition including DoD and threshold for remaining capacity, for example

‘after 600 charging cycles with 90% discharge in each cycle, 75% of the battery capacity

should remain’

341

. This means that cordless and robots will need 2 batteries on average in

their lifetime of 6 years, since they are used 200 times a year. Setting stricter requirements

for the batteries may lead to oversized batteries which challenge both the comfort level for

the consumer and the resource efficiency of the battery. Instead batteries should be

available as a spare part.

Better reparability

The 2015 JRC-IES study on durability of vacuum cleaners has calculated the life cycle costs

for a better reparability of vacuum cleaners. They follow the MEErP and the Ecoreport for

important parts of the study. As regards the LCC, the study assumes a base purchase price

of 150 €, a repair cost of 20% of the discounted original purchase price (22 €), electricity

consumption of 25 kWh/year an electricity rate of 0.205 €, an improvement multiplier

(δ=85%) for the energy use of the new product that replaces the broken-down product,

339

https://www.vhk.nl/downloads/Reports/2016/VHK%20546%20FINAL%20REPORT%20VC%20Durability%20Test%2020160623.

pdf

340

Commission Regulation (EU) No 617/2013 of 26 June 2013 implementing Directive 2009/125/EC of the European Parliament

and of the Council with regard to ecodesign requirements for computers and computer servers. OJ L 175, 27.6.2013, p. 13–33

341

EN 61960:2011 could be used for measuring battery endurance in cycles (part 7.6.2 or 7.6.3 in the standard)

241

etc. For various scenarios of lifetime extension the results are shown in the figure below.

It shows that for this product in all cases the Life Cycle Costs with the repair are lower.

Figure 64: LCC of the base-case (first column) and the durable scenario (second column)

(source: JRC-IES 2015)

One of the conditions for these repair-scenarios is of course that the repair is feasible

because the spare part is still available. For other Ecodesign products the availability of

spare parts, after the model has seized to be produced, is to be guaranteed for up to 7

years. Also in this case, a period of 7 or 8 years, 8 years being a condition in the Blue

Angel environmental mark, seem reasonable.

Another condition is that the labour cost will be limited or at least a part of the repairs can

be done by the consumer/user of the product. In that sense, it seems reasonable to

demand that the most repair-prone products, like hoses, must be replaceable without

special tools.

Re-use

Studies on the re-use of small household appliances are scarce. Prakash et al. (2016)

342

discusses the product life and grounds for discarding hand mixers (first-hand service life

10-11 years) and electric kettles (first-hand service life 5-7 years). 63% of discarded

mixers and 71% of discarded kettles were fully functional, 22% of mixers and 11% of

kettles showed defects but still worked and 11-13% didn't work. As method of discarding

these small appliances 7.8% of respondents mentioned re-use (sold or given away).

342

Prakash, S., Stamminger, R. et al., Einfluss der Nutzungsdauer von Produkten auf ihre Umweltwirkung: Schaffung einer

Informationsgrundlage und Entwicklung von Strategien gegen „Obsoleszenz“, Umweltbundesamt Texte, 11/2016.

242

In the 2016 preparatory Ecodesign study on washing machines

343

the authors found two

references for what was called ‘re-use’ (sold or given away) of large household appliances:

Magalani et al. 2012

344

found that in Italy 8% of discarded products were re-used. A WRAP

2011 study for the UK estimates that 3% of large household appliances are re-used, after

discounting for the fact that 25% of products offered for re-use are unrepairable

345

Based on this (scarce) information, it is estimated that for 7 to 8% of smaller appliances

like vacuum cleaners there is a second owner. Like with large household appliances there

are no studies that regard how long the average second-hand life of a vacuum cleaner in

the EU actually is. The study team assumes that the second owner should be able to use

the vacuum cleaner at least 2-3 years

346

. As mentioned before, in the context of this

preparatory study for Ecodesign and Energy Labelling measures for new products, giving

away old vacuum cleaners is not relevant, because it cannot be changed through

regulation. What could be relevant is where re-use of the whole product requires a full

refurbishing, preparing the product for a true second life. Such an activity could not be

identified, but because we cannot exclude it, the EcoReport assumes that full refurbishment

applies to 1%.

What is relevant is the re-usability of filters, e.g. at least the HEPA and motor filter. The

current cost of a HEPA filter is in the range of 10-35 €, depending on brand and type. The

motor filter is normally washable but a new one costs 5-15 €. Manufacturers would

recommend replacing at least the HEPA filter annually or bi-annually, so making them re-

usable saves on monetary costs, but also on materials and energy as clogged filters have

a higher air resistance (up to 2% energy savings is expected). The latter energy saving is

especially true for the many users that never change their HEPA filter

347

. As regards the

bag versus bagless discussion, the past years have taught that there is a market for both

and that the regulator should not interfere. What can be done is giving the information to

the consumer, e.g. via the energy label, whether a product is bagged or bagless, however,

this information is often very easily accessible for the consumer already.

343

Boyano Larriba, A., Cordella, M., Espinosa Martinez, M., Villanueva Krzyzaniak, A., Graulich, K., Rüdinauer, I., Alborzi, F.,

Hook, I. and Stamminger, R., Ecodesign and Energy Label for household washing machines and washer dryers, EUR 28809 EN,

Publications Office of the European Union, Luxembourg, 2017, ISBN 97892-79-74183-8, doi:10.2760/029939, JRC109033

344

Magalani, F.; Huisman, J. & Wang, F. (2012). Household WEEE generated in Italy: Analysis on volumes & consumer disposal

behaviour for waste electrical and electronic equipment.

Available at http://www.weeeforum.org/system/files/2012_ecodom_weee_arising_in_italy_en.pdf.

345

WRAP (ed.). Benefits of Reuse: Case Study: Electrical Items, 2011. Available at

http://www.wrap.org.uk/sites/files/wrap/Electricals%20reuse_final.pdf.

346

This implies the 8 years product life for the cylinder VC is composed of 93% at a service life of 7.85 years plus 7 % at a

service life (first- and second-hand) of 10 years.

347

Stakeholder feedback in first stakeholder meeting 2018.

243

Option 5: Recycling

The technical recycling options have been discussed extensively in section 10.2. Currently

48% of the product and use phase waste of BC1 is recycled (4.1 of 8.7 kg), of which one

third of the plastics and auxiliaries (filters and filter bags), most of the metals and

cardboard packaging, very little of the electronics (just the precious and critical materials

if present).

The main challenge for BC1 is increasing the recycled fraction of the 4.3 kg plastics in the

product and 1.5 kg (also mainly plastic) auxiliaries. Of this total 5.8 kg plastic fraction in

the disposed product, 1.7 kg (30%) goes to recycling, 1.8 kg (32%) to heat recovery, 2.1

kg (38%) goes to landfill or incineration without heat recovery. The policy objective in the

WEEE will be to achieve 65% recycling of small appliances like the vacuum cleaner. The

waste policy objective is to abolish landfill in the EU (although there might always be a

small fraction of illegal dumping). Putting this all together means that 3.64 kg (65%),

almost 2 kg extra should go to recycling and the rest to heat recovery

348

In the Ecoreport tool the credit for recycling is 40% of all impacts. In this case it is proposed

to take PP as the reference plastic

349

. With over 30 million sales and 8 years lifetime (in

2022) the saved impacts from 2022 until 2030 would be 40% of the impacts of 240 million

kg PP, i.e. amongst others 7 PJ primary energy and 192 million kg CO

equivalent.

Recover means that the product is incinerated with waste heat recovery. Most vacuum

cleaners, without batteries, do not contain toxic materials and can safely follow that route,

even if the vacuum cleaners are not collected separately as WEEE this will happen. For

discarded PP the combustion value is still some 75-80% of the combustion value of the oil

feedstock that was needed for its production. For ABS this is around 50-60%. The metals,

if not removed beforehand, can still be found in the remains and ashes and thus will always

be used for recycling.

At the moment, 40% of plastics is incinerated with heat recovery. To ensure that heat

recovery can take place and products don’t go to landfill or non-recovery (hazardous)

incineration it is important that no hazardous materials such as those mentioned in RoHS

and REACH are included (see Blue Angel requirements). Also halogenated flame retardants

should be avoided. Once the Member States have met the waste target of abolishing

landfill, the share of incineration with heat recovery versus the landfill will not decrease.

348

Of course, in combination with the previous design option of re-using the filters it might not be 2 kg but less, but for a first

calculation 2 kg is taken.

349

Main impacts Ecoreport per kg PP: 73 MJ primary energy of which 7 MJ electric and 53 MJ feedstock, 4 g hazardous waste,

28 g non-hazardous waste, 2kg CO2 eq. (GWP), 6 g SO2 eq. (Acidification), 1 g particulates (PM), 165 g phosphate eq.

(eutrophication).

244

12.2 Commercial mains-operated vacuum cleaners (BC2)

For commercial mains-operated vacuum cleaners, Base Case 2, the same options and

considerations apply as for Base Case 1. But the absolute numbers of the impacts will be

different, in accordance with the data supplied in section 10.3.2.

12.3 Cordless vacuum cleaners (BC3)

The most important design option is whether or not to include cordless vacuum cleaners

in the scope. This adds to the energy and material consumption in the scope and thus the

energy and material saving potentials. At the moment they are the fastest growing

segment in the vacuum cleaner market. Most models have a performance that would

qualify them as ‘hard floor only’, because they do not meet the performance requirements

for carpet cleaning, but there are now a few models that would qualify as ‘general purpose’

and at least one leading manufacturer, Dyson, who claims to no longer invest in new corded

machines but only develop cordless vacuum cleaners. These are all important reasons to

take the cordless vacuum cleaners on board.

Option 1: Power limits for maintenance mode

The most important energy saving option for cordless vacuum cleaners is the maintenance

mode power consumption, i.e. when the battery is fully charged and in a docking station.

According to section 10, the power consumption in this mode varies between less than 1

W and 8 W, depending on the model. The estimated market average is 2.6W or rather 50%

of the total energy consumption of the vacuum cleaner (21 kWh on 41 kWh/year total). As

discussed in section 10, there is no technical reason for this. For Li-ion batteries there is

no ‘trickle charge’ and even for NiHM its merits from a long-life point of view are

questionable. This option entails to bring down the maintenance mode (charged and

docked) power to 0.5-1 W. Thus saving 8 kWh/year (13 € over product life in energy) at

no cost at all.

Options 2 and 3

Options and considerations under option 2 to 3 of BC1 apply, albeit with different figures.

Option 4

As regards option 4, the increase of product life, the battery is a highly relevant additional

issue on top of what is mentioned with BC1. The battery life of cordless and robot models

is relevant for the lifetime and thus the resource efficiency. Given a motor life of 550 hours

for cordless vacuum cleaners, 550 hours battery life under normal usage conditions also

seems reasonable. At the moment this is only feasible with a Li-ion battery (500-1000

cycles, no memory effect

350

). A battery-life to match a 1200 hour robot motor life is not

technically feasible at the moment; 600 hours is probably all that can be achieved at a

350

Memory effect relates to a diminished battery capacity in time, as a result of supoptimal (incomplete, or too soon) charging

245

reasonable size of the battery of around 0.5 kg. Potential buyers will have to be made

aware that the battery will have to be replaced every 2-3 years, probably at a cost of

around 80-100 € for Li-ion. NiMH may have to be replaced twice as much for the same

lifetime, but it also costs half.

Option 5

Option 5, recycling, could be extended with considerations regarding batteries:

In addition to what was said with BC1, the batteries are a separate issue for recycling. Li-

ion is now the most common type. The challenge is to recover the cobalt, typically 10-

20%, from the battery. This is technically difficult

351

, but cobalt is rare and much in

demand. This rising demand made its price triple in recent years to currently 30 $/lbs (57

€/kg). With a typical vacuum cleaner battery weighing 0.5 kg and thus cobalt 0.05-0.1 kg

this means that there is 2.85 € to 5.70 € to gain there

352

. Another possibility is to use Li-

ion batteries with lower share of cobalt, i.e. NCA (2-4% Co) instead of NMC (4--8% Co).

The second most common battery type is NiMH (Nickel Metal Hydride), which contains

typically 35% of nickel and possibly up to 15% of cobalt. Also here the recycling process

is challenging but worthwhile

353

. It is possible to set a limit of e.g. 20% cobalt on the

battery. This would merely be pre-emptive, i.e. to avoid that certain Li-ion batteries such

as LCO (Lithium Cobalt Oxide) types would ever be used. At the moment there would be

no impact from such a measure.

According to the Battery Directive 2006/66/EC the EU Member States should, from 2016,

collect 45% of the batteries in the waste stream and of this 50% should be recycled. Note

that from 1.1.2017 it is forbidden to use Ni-Cd batteries or other types containing lead,

mercury and cadmium for vacuum cleaners. Apart from helping Member States to meet

Battery Directive targets the Ecodesign measures could also aid the WEEE targets by

prescribing easy dismounting of batteries and easy disassembly of the PCBs of robot

vacuum cleaners.

12.4 Household robot vacuum cleaners

All the options and considerations of BC3 apply, but with different impacts. These impacts

can be seen in section 10.3. Only as regards option 1 a small modification is proposed: As

many robot vacuum cleaners should be able to wake up on a signal of their Local Area

Network (WOL=Wake-up On LAN) it is reasonable to set the maintenance mode limit at 2

W instead of 0.5 W, in accordance with the 2019 networked standby requirements.

351

https://www.researchgate.net/publication/259645071_Recycling_of_Spent_Lithium-Ion_Battery_A_Critical_Review

352

https://www.bloomberg.com/news/articles/2017-12-01/the-cobalt-crunch-for-electric-cars-could-be-solved-in-suburbia

353

Also see https://csm.umicore.com/en/recycling/battery-recycling/our-recycling-process

246

13. Task 7: Scenarios

This chapter consists of two main parts, one retrospective and one forward-looking:

• Evaluation of the existing regulations in accordance with the Better Regulation

parameters efficiency, effectiveness and relevance

• Scenarios and recommendations for amending and improving the regulations.

Scenarios for both energy and resource efficiency are included in this section.

Evaluation is a tool to help the Commission learn about the functioning of EU interventions

and to assess their actual performance compared to initial expectations. By evaluating, the

Commission takes a critical look at whether EU activities are fit for purpose and deliver

their intended objectives at minimum cost (i.e. avoiding unnecessary costs or burdens).

Since the evaluation is retrospective and based on collected market data, it includes the

previous, annulled Energy Labelling Regulation and the effect it had on the market.

The scenario section presents the options for how the Regulations can be further improved

and how they can aid in better environmental performance of vacuum cleaners. Options

are presented for two different aspects separately: energy efficiency and resource

efficiency. These aspects are analysed separately, and in the end, recommendations are

given as to what combination of energy and resource requirements are favourable based

on both cost and environmental impact.

13.1 Better Regulation evaluation

The purpose of this section is to evaluate the effect of the current Ecodesign Regulation

and the annulled Energy Labelling Regulation for vacuum cleaners, and compare the results

obtained so far with the expectations in the impact assessment. In addition, it analyses

how well the regulations have been able to solve the market failures identified in the impact

assessment.

The evaluation will focus on answering questions regarding:

• Effectiveness of the regulations. What has been the impact of the regulations so far

and have the objectives of the policy measures been achieved?

• Efficiency of the regulations. Have the regulations been cost effective and are the

costs justified?

• Relevance of the regulation. Are the regulations still relevant and have the original

objectives been appropriate?

These questions are based on the official template

354

and the Better Regulation Toolbox

355

354

https://ec.europa.eu/info/sites/info/files/file_import/better-regulation-toolbox-47_en_0.pdf

355

https://ec.europa.eu/info/files/better-regulation-toolbox-47_en

247

and the specific questions answered in each of the following sections, are the questions

from these sources.

The existing regulations are the Ecodesign Regulation and the annulled Energy Labelling

Regulation for vacuum cleaners Regulation EU 666/2013 and regulation EU 665/2013

respectively. The aim of the regulations was to provide dynamic incentives for suppliers to

improve the energy efficiency of vacuum cleaner for household and professional use and

to accelerate market transformation towards energy-efficient technologies.

According to the current Ecodesign Regulation the annual electricity consumption of

vacuum cleaners covered by the regulation was 18 TWh in the Union in 2005. Without the

regulations the annual electricity consumption was predicted to be 34 TWh in 2020. In

addition, the preparatory study showed that it is possible to significantly reduce the

electricity consumption of vacuum cleaners.

Description of the current regulations and their objectives

The Ecodesign Regulation and the annulled Energy Labelling Regulation have been

prepared in a parallel process with the aim to assess the possibilities of implementing

Ecodesign and Energy Labelling requirements for vacuum cleaners.

The two regulations are intended to work in synergy; the Ecodesign Regulation pushing

the market towards higher energy efficiency by removing the least efficient vacuum

cleaners from the market, and the energy label pulling the market towards even higher

energy efficiency by providing consumers with the necessary information to identify the

most energy efficient vacuum cleaners on the market.

The Ecodesign Regulation for vacuum cleaners entered into force in 2013 and sets

maximum limits for annual energy consumption in two tiers from respectively 1 September

2014 and 1 September 2017. In addition, the two tiers include maximum limit for rated

input power, and minimum levels for dust pick-up. The second tier also include

requirements regarding dust re-emission, sound power level, durability of the hose and

operational motor lifetime.

The annulled Energy Labelling Regulation also entered into force in 2013. According to the

Energy Labelling Regulation vacuum cleaners should have, when displayed for sale, from

1 September 2015 bear an energy label with an A-G scale and from 1 September 2017 an

energy label with three more ambitious energy classes on top of the A-class (i.e. A+, A++,

and A+++). From 18 January 2019 the energy label may no longer be displayed on vacuum

cleaners.

See a more detailed description of the current regulations in section 7.2.1.

248

The objectives of the current regulations are:

• Correcting the identified market failures in the preparatory study and impact

assessment

• Reducing energy consumption and related co

and pollutant emissions due to

vacuum cleaners following Community environmental priorities, such as those set

out in decision 1600/2002/EC or in the Commission’s European climate change

programme (ECCP)

• Promoting energy efficiency hence contribute to security of supply in the framework

of the community objective of saving 20% of the EU’s energy consumption by 2020.

More specifically, the objectives of the current regulations were to reduce the energy

consumption of vacuum cleaners by 20% (from 34 to 19 TWh/year in 2020) and CO

emissions from 11 to 7 Mt/year.

According to the Impact Assessment

356

the main market failure related to vacuum cleaners

was the lack of consumer information on energy use and cleaning performance. As a result,

most consumers took the electric power input (in W) as a proxy for cleaning performance.

The power consumption of vacuum cleaners had been rising from 1275 W in 1990 to around

1500 W in 2005 and was expected to reach 2300 W in 2020 (without measures). Non-

household ‘professional’ vacuum cleaners were more efficient (30% less power for a better

performance) but still had a potential for energy savings. At the same time a decrease in

the energy efficiency was seen, which meant the higher power ratings were not justified

by better cleaning performance. High power rating was actually a sales argument in itself.

Over the past decades this led to low price, high-power but low-performance vacuum

cleaners, mainly from China, flooding the EU market and more than doubling the societal

energy consumption of vacuum cleaners. Vacuum cleaners were therefore a significant

contributor to household’s energy consumption and a candidate for Ecodesign measures.

Baseline and point of comparison

The baseline for the evaluation is the market without the implementation of the current

Ecodesign Regulation and the annulled Energy Labelling Regulation. The Impact

Assessment accompanying the actual regulations is normally an important data source for

determination of the baseline. However, the stock and sales figures used in the 2013

Impact Assessment for vacuum cleaners is generally very high compared to the newest

available data. Therefore, the stock model from this review study is used as a baseline for

356

Executive summary of the Impact assessment accompanying the documents Commission Regulation with regard to

Ecodesign requirements for vacuum cleaners and Commission Delegated Regulation with regard to Energy Labelling of vacuum

cleaners http://ec.europa.eu/smart-regulation/impact/ia_carried_out/docs/ia_2013/swd_2013_0241_en.pdf

249

the evaluation in this section, and the data from the 2013 Impact Assessment is scaled to

match. Only vacuum cleaner types in scope of the current regulation is evaluated, and

cordless and robot vacuums are thus excluded from all analysis.

The projected stock reduction post-2015 is not due to a decrease in overall vacuum sales,

but because a larger percentage of the total sales is moved to robots and cordless. The

stock data appear from Figure 65.

Figure 65: Comparison of stock in 2013 Impact Assessment (IA) and the stock estimates used

in this study

One of the objectives of the evaluation is to compare the effect of the current regulations

with the foreseen effects when the regulations were adopted, which means the result of

the 2013 Impact Assessment. The estimations in the Impact Assessment are generally

good, but difficult to use for comparison, as none of the policy options included in the

Impact Assessment are used in the regulations. The policy options used as for comparison

is PO 5, Sub-option 1, which is the most ambitious option addressed in the Impact

Assessment

357

Furthermore, no unit prices were reported in the Impact Assessment, which only reports

total user expenditure for the stock. Since other sales and stock numbers are used in the

Impact Assessment, the results can therefore not be compared directly. Instead data from

this review study was used and scaled with the data in the Impact Assessment.

The following terminology is used in the following analysis and figures:

• BAU0 = scenario without the current regulation,

• BAU = scenario with the current regulation,

• IA SO1 = scenario predicted in the 2013 Impact Assessment PO 5, sub-option 1.

357

This sup-options includes power caps of 1350 W in the first tier and 1050 in the second tier which is less ambitious than the

actual requirements in the regulation

250

Effectiveness

Evaluation question 1: What have been the effects of the regulations?

Energy savings and reduction of CO

-emissions

The regulations have transformed the market towards a higher energy efficiency and have

resulted in electricity savings and reduction of CO

-emissions. Compared to the

expectations in the Impact Assessment this study shows for key parameters very similar

results, however for a lower stock. Comparison of results appears in Table 95 and Figure

66 and Figure 67.

Table 95: Comparison of results of this study to results from the 2013 Impact Assessment

regarding cumulative savings of key parameters

Study

Parameter

2015

2020

2025

2030

This review

study

Electric savings [TWh]

138

276

384

GHG emissions [Mt CO

-eq]

105

151

User expenditure [bln. €]

Electric savings [TWh]

153

280

377

GHG emissions [Mt CO

-eq]

107

148

Regarding the development of the total annual energy consumption and CO

-emissions the

current regulations have already resulted in significant savings and further savings is

expected in the coming years. The savings of the regulations are very much in line with

the expectations in the 2013 Impact Assessment with a tendency that the regulations could

achieve more savings in the longer-term than estimated in the Impact Assessment.

251

Figure 66: Total energy consumption for various scenarios (based on stock)

Figure 67: Greenhouse gas emissions related to electricity consumption in the use phase

Annual energy consumption (stock)

The power limits in the second tier of the Ecodesign Regulation have reduced the rated

input power of vacuum cleaners. This has contributed to a large decrease in the annual

energy consumption of the vacuum cleaners in scope of the regulations.

In the period from 2010-2016 the annual energy consumption of vacuum cleaners on the

market shows a declining trend as seen Figure 68. This is the case for all types of vacuum

cleaners covered by the regulations. Based on the market average, the annual energy

consumption decreased from around 78 kWh/year in 2010 for household cylinder vacuum

cleaners to 34 kWh/year in 2016. For upright vacuums, it declined from around 74 to 29

kWh/year. For handstick mains it decreased from 44 to 30 kWh/year. This means that the

energy consumption of the vacuum cleaners in scope declined by more than 50% on

average

358

in just 6 years.

358

Based on market data from GfK

252

The impact of the current regulations on energy consumption must be attributed to both

the Ecodesign power limit and the annulled Energy Labelling Regulation.

The Ecodesign power limit has resulted in a decrease of the annual energy consumption

from the 2010 values down to the limit values of 62 kWh/year by 2030 and 43 kWh/year

by 2017. However, the annual energy consumption for all vacuum cleaner types covered

by the Regulations have decreased further than this limit as shown in Figure 68. This

illustrates that the annulled Energy Labelling Regulation has created a market pull beyond

the Ecodesign power limits, and approximately 20% of the savings over the years are

expected to be attributed to the annulled Energy Labelling Regulation as illustrated in

Figure 69. This is based on the assumption that all savings beyond ecodesign are a result

of the label market pull. The dotted line in the graph illustrates the Ecodesign limits. All

savings from the previous BAU0 graph down to the dotted line are expected to be the result

of ecodesign, while the savings from the dotted line down to the BAU graph are expected

to be the result of the energy label.

Figure 68: Average annual energy consumption of household VC in stock and impact of

Ecodesign and Energy Labelling Regulations

253

Figure 69: Share of energy savings due to the Ecodesign regulation and the previous, annulled

Energy Labelling Regulation, based on average AE value of sales each year

Distribution of energy classes

Looking at the distribution of vacuum cleaners in the annulled energy classes, the data

analysis is difficult because useful data is only available for two years. Before

implementation of the regulations, information about energy and performance classes was

only available for a small share of the market. Even though the annulled Energy Labelling

Regulation entered into force in 2013 the labelling requirement was not mandatory before

1 September 2014. In 2014 information about the parameters on the label was only

available for 6% of vacuum cleaners on the market as seen in Table 96. The share of

vacuum cleaners bearing the label information increased to 77% in 2015 and 85% in 2016

on average for all types covered by the annulled Energy Labelling Regulation. The average

value is close to value for cylinder vacuum cleaners as they constitute 85% of the EU

vacuum market.

Table 96: Coverage of the previous, annulled energy label data for each vacuum cleaner type

in scope of the regulations

2013

2014

2015

2016

Cylinder

19%

79%

86%

Upright

16%

23%

67%

75%

Handstick mains

13%

73%

77%

Total

19%

77%

85%

Looking only at the share of vacuum cleaners provided with the annulled energy label, the

share of vacuum cleaners in energy class A increased from 47% in 2013 to 59% in 2016.

The share of A labelled vacuum cleaners was higher for uprights and handsticks than for

cylinder vacuums in 2016, but for all three types the shares of A and B labelled vacuum

cleaners dominate the market compared to the situation in 2013 (Figure 70 and Figure

71). At the same time the share of vacuums with C, D, and E energy classes has decreased

254

and the share of cylinder vacuums in energy class F and G has also decreased, while for

handstick and upright cleaners these were almost non-existent already in 2013.

Figure 70: percentage distribution of energy classes for each vacuum cleaner type in 2013,

label coverage 6%

Figure 71: Percentage distribution of energy classes for each vacuum cleaner type in 2016,

label coverage 85%

Dust pick-up and dust re-emission

The majority of vacuum cleaners sold in the EU are considered ‘general purpose’, meaning

they are intended for use on both hard floor and carpet, and the dust pick-up class should

be shown on the label for both. In addition, the dust re-emission class should be shown on

the label. The existing data regarding the development of the market share of vacuum

cleaners in the various dust pick-up and dust re-emissions classes is very uncertain

359

, and

it is not at the current stage possible to evaluate the impact of the regulations on these

359

GfK data coverage is less than 20% for the years 2013 and 2014 on these parameters, hence only 2015 and 2016 data is

available, which is not sufficient to draw any conclusion on trendlines.

255

parameters. Furthermore the parameters were not quantified for the market in the Impact

Assessment. See section 8.5 of this report regarding the current market distributions.

Evaluation question 2: To what extent do the observed effects link to the regulations?

The observed market change towards more energy efficient vacuum cleaners is likely to

be largely linked to the regulations, where the Ecodesign Regulation removed the most

energy consuming models from the market, and the annulled Energy Labelling Regulation

created further market pull beyond the power caps.

It is unlikely that the effects are in part due to other factors such as general innovation

and market trends towards more energy efficient vacuum cleaners as this is not line with

the development for vacuum cleaners seen during the last decades. Especially for

household vacuum cleaners the regulations have been able to turn the market trend from

rising rated input power, because high input power was the most important assessment

parameter for the consumers, to reduced rated input power and higher energy efficiency.

Evaluation question 3: To what extent can these changes/effects be credited to the

intervention?

Before implementation of the regulations there was no available information about energy

efficiency and cleaning performance of vacuum cleaners. The only available information

was the rated input power. Without the regulations this would still have been the case and

the consumers would still buy vacuum cleaners with the highest rated input power,

considering vacuum cleaners with high rated input power to have the best cleaning

performance. Therefore, the observed impact can to a very large extent be credited to the

regulations. If the regulations were not implemented the rated input power would probably

have increased to an even higher level leading to a further increase in the annual energy

consumption for especially household vacuum cleaners.

Evaluation question 4: To what extent can factors influencing the observed achievements be

linked to the EU intervention?

Some factors have reduced the achievements of the regulations. This is mainly a slow

implementation of the label (low coverage), lack of consumer awareness and around 70%

of consumers finding one or more parts of the information on the label unclear. Figure 72

shows the share of the 70% finding each label information unclear. It appears that more

than half of the consumers who find some of the label information unclear, find cleaning

performance and dust re-emission information particularly difficult to understand.

256

Figure 72: Share of people finding areas of the annulled label unclear, out of the 70% finding

at least one parameter unclear (source: APPLiA 2018 consumer survey)

Another factor that could have reduced the market pull effect of the label towards higher

energy efficiency is that consumers consider cleaning performance more important than

energy efficiency. As seen in Table 97, 90% of consumers consider cleaning performance

very important or important while only 67% of consumers consider energy efficiency very

important or important. This is in line with the consumers’ preference before the

implementation of the annulled energy label where the consumers chose vacuum cleaners

with high rated input power considering that high power was similar with high cleaning

performance. High cleaning performance is still the most important performance parameter

for the consumers.

Table 97: Percentage of consumers rating parameters important/very important in a purchase

situation (Source: APPLiA 2018 consumer survey)

Parameter

Percentage answering “very

important” or “important”

I expect it to last a long time

91%

Its performance

90%

The ease of use

89%

The price

87%

The ease of maintenance

86%

The type /stick, robot, canister etc.)

80%

A good filtration of the dust (allergies)

79%

The time spend cleaning

77%

257

The noise level

67%

The energy efficiency

67%

Having/not having a bag

66%

In addition, uncertainty about the cleaning performance information on the label probably

has a negative impact on the achievements. Not because it affects the end-user

understanding of the label, but because some might experience buying a label A vacuum

cleaner and not getting the expected performance.

Only the factor dealing with unclear information on the label is directly linked to the

Regulation. The other factors are more related to enforcement, consumer awareness and

preferences when purchasing vacuum cleaners, measurement standards and the

consumers’ confidence in the information on the label.

The scope of the current Regulations also reduces the achievements because not all types

of vacuum cleaners are included in the scope, such as robot and cordless vacuum cleaners.

Extension of the scope to cover (both or one of) these types of vacuum cleaners is assessed

in the second part (the scenario part) of this task.

Conclusion effectiveness

The current regulations have been very effective in reducing the electric consumption, and

GHG emission of vacuum cleaners. The Ecodesign Regulation so far seems to have been

more influential than the annulled energy label, resulting in around 80% of the total

savings.

The energy savings has also led to monetary savings for end-users, due to the savings in

electricity costs and despite the increased purchase price. However, unclear information

on the label and uncertainty about measurement standards probably reduces the

achievements of the regulations.

Also, the trend that more and more of the sale of vacuum cleaners moves towards robot

and cordless vacuum cleaners reduces the effectiveness of the Regulations because these

types of vacuum cleaners are not included in the scope.

Efficiency

Evaluation question 1: To what extent has the intervention been cost-effective?

The average price of vacuum cleaners increased from 2013 to 2016 for all the vacuum

cleaner types in scope of the regulations; cylinder prices increased from 109 EUR to 119

EUR, uprights from 168 EUR to 179 EUR, and mains handstick from 90EUR to 95 EUR

360

These prices are not corrected for inflation.

258

This indicates that the manufacturers have passed on the extra costs for innovation and

improvements of vacuum cleaners to the end-users and that the end-users are willing to

pay a higher price for more efficient vacuum cleaners.

With the annulled Energy Labelling Regulation manufacturers had the possibility to improve

the performance of their products and achieve a price premium because it is possible to

make higher performing products identifiable by the label. This is contradictory to the

situation before energy labelling where the only sales argument and consumer information

was the rated input power.

Even though the vacuum cleaner purchase price has increased the total cost of ownership

(i.e. costs for purchase and use) have decreased due to the regulations. This is the case

both for household and commercial products.

The figures below show the development in the total costs of ownership for an average

vacuum cleaner for respectively household and commercial vacuum cleaners.

As the manufacturers are able getting a higher price (increased turnover) and the total

costs of owner ship has decreased for the end-users the regulations seem to be cost-

effective.

Figure 73: Average total costs of ownership for household users

259

Figure 74: Average total costs of ownership for commercial users

The regulations apply some extra costs for testing on the manufacturers. However, as both

regulations are based on self-declaration, no excessive testing costs are put on the

manufacturers. In addition, experiences from the annulled EU energy labelling scheme

show strong evidence that manufacturers have reacted positively to the EU energy labelling

and consider the label as an important instrument to differentiate their products. This also

suggests that the extra investments needed to achieve higher efficiency levels have

generally been outweighed by the benefits

361

Dealers must ensure that vacuum cleaners bear the label at the point of sale and they will

have to cover the administrative costs for this activity. Although no quantitative data is

available, costs for dealers to show the label on displayed products is widely accepted

within the framework of the EU energy labelling scheme for energy-related products. In

addition, the dealers will benefit from higher turnover due to increased sales of better

performing and more expensive vacuum cleaners. As dealers of household vacuum

cleaners usually display other energy labelled products, and in the past vacuum cleaners

with labels, they are already familiar with the procedures and they will easily could transfer

their experiences to the re-introduced product group.

Member States need to bear the costs for market surveillance, but they will also benefit

from the energy savings and the reduction of emissions. In addition, EU wide legislation

will be more cost effective from a Member State perspective compared to national

legislation, because the costs of developing the regulation, test methods and conducting

pre-regulatory studies are shared instead of conducted for each country separately.

The costs for market surveillance vary between Member States. Some carrying out almost

no activities while others undertake both shop inspections, inspection of documentation,

361

Ecofys, Evaluation of the Energy Labelling Directive and specific aspects of the Ecodesign Directive, June 2014.

260

and testing. No EU-wide data regarding Member States costs for market surveillance with

regard to vacuum cleaners is available.

Evaluation question 2: To what extent are the costs of the intervention justified, given the

changes/effects it has achieved?

The current Regulations have resulted in substantial savings for end-users and society,

without excessive costs for manufacturers, other market actors or Member States. In total

the regulation will in 2020 have saved 116 TWh of electricity, corresponding to 45 Mt CO

eq, and users will in total have saved 19 bln. EUR.

Manufacturers have been able to pass on the extra cost for development of better

performing vacuum cleaners to end-uses, and both manufacturer and retailers have

benefitted from increased turnover compared to the situation without regulations. Both

with and without the regulations the turnover is foreseen to decrease due to the expected

sales of mains-operated household vacuum cleaners

362

. But the turnover is estimated to

be higher with the regulations than without

363

, as shown in Figure 75 and Figure 76.

Figure 75: Manufacturers turnover without regulations (BAU0) and with the current

regulations (BAU).

362

This decrease was not expected by the Impact Assessment, but is shown by the newest market data

363

Calculation of turnover in the BAU scenario is based on sales prices from GfK. Manufacturer turnover estimated by using

261

Figure 76: Retailers turnover without regulations (BAU0) and with the current regulations

(BAU).

Member States need to bear the costs for market surveillance, but they will also benefit

from the energy savings and reduced emissions due to the Regulations. In addition, an EU

wide legislation will be more cost effective from a Member State perspective compared to

national legislation. Therefore, the intervention costs seem justified given the improved

performance of vacuum cleaners and the associated benefits.

Evaluation question 3: To what extent are the costs associated with the intervention

proportionate to the benefits it has generated? What factors are influencing any particular

discrepancies? How do these factors link to the intervention?

Due to the benefits illustrated above and the low costs for implementation of the

regulations, the intervention is considered proportionate. The fact that the Ecodesign

Regulation and the annulled Energy Labelling Regulation are implemented in a parallel

process and with use of the same test procedures and calculations methods for proving

compliance (for annual energy consumption and cleaning performance) makes the

regulations more cost efficient for manufacturers.

No particular discrepancy has been identified so far.

Evaluation question 4: To what extent do the factors linked to the intervention influence the

efficiency with which the observed achievements were attained? What other factors influence

the costs and benefits?

Since the efficiency to some extent depends on the effectiveness of the Regulations, the

same factors as mentioned above also influence the efficiency.

If consumers are not aware enough of the label and/or find label information unclear

vacuum cleaners with high performance according to the label parameters will probably

262

not have a market advantage, but rather the opposite since they are often also sold at

higher prices.

In addition, the lack of reproducibility of the measurement method and the on-going work

to develop new standards probably at least in a short-term perspective decrease the

manufacturers’ incentive to develop even better performing products, until it is determined

exactly how measurements will be performed. Increasing consumer relevance of the test

methods will also increase manufacturer incentive to produce better performing vacuum

cleaners that reduce the life cycle costs of the product, both on the label and in real life.

Evaluation question 5: How proportionate were the costs of the intervention borne by different

stakeholder groups taking into account the distribution of the associated costs?

Manufacturers of vacuum cleaners bear the largest share of the costs, but they have so far

been able to pass the extra costs on to end-users, without increasing the total costs for

end-users over the lifetime of the vacuum cleaners. As shown above the total costs of

ownership have decreased significantly due to the current regulation.

End-users bear the costs for more expensive vacuum cleaners, but they will be

compensated by saved electricity costs over the lifetime of the vacuum cleaners and

increased performance.

Member States bear the costs for market surveillance for energy related products and in

general vacuum cleaners is only a small part of that.

In addition, it is important to bear in mind that it is voluntary for manufacturers to improve

the performance of vacuum cleaners beyond the Ecodesign requirements and for end-users

to purchase the products with high energy classes.

Evaluation question 6: Are there opportunities to simplify the legislation or reduce

unnecessary regulatory costs without undermining the intended objectives of the

intervention?

One opportunity for reduction of the regulatory costs is establishment of a product

registration database. This is already decided for products covered by Energy Labelling

Regulations and implemented via the Energy Labelling Framework Regulation (EU)

2017/1369. However, further reduction of the administrative costs for Member States could

be achieved if the database is extended to cover also Ecodesign Regulations (i.e. the

manufacturers should have an obligation to enter technical documentation and other

relevant documents proving compliance with relevant Ecodesign Regulations in the product

registration database). This is relevant because the Ecodesign Regulation includes various

requirement that is not included in the Energy Labelling Regulation, and the technical

263

documentation for proving compliance with energy labelling not is sufficient to prove

compliance with Ecodesign.

This would save the MSAs a lot of time in the process of first identifying the economic

operator in each country (manufacture or their representative) and retrieve the correct

documents before they can start verifying them. While document control is not considered

as important as testing, it is the step that goes before the testing, and saving time on

document control will leave more time and resources to perform actual tests. The market

surveillance could improve even more if Member States collaborated across borders to

check products, and the database is the first step in the direction of such a collaboration.

If all the necessary documentation is available in a data base the burden for Member States’

MSAs to obtain the documentation will be reduced, and the burden for manufacturers to

send documentation to each MSA, likewise.

As the Commission is already obliged to set up the database for energy-related products

covered by Energy Labelling Regulations, the extra costs for inclusion of products covered

by Ecodesign Regulations would most likely be marginal.

Another opportunity for simplification of the regulations is to differentiate which

parameters are covered by ecodesign requirements and which are covered in the energy

label. Some ecodesign limits might be set so high/low that there is no room for

differentiating the remaining products into different label classes within the frames of the

test uncertainties. In that case these parameters could be removed from the label. The

same is true the other way around, especially for requirements for new parameters, where

it might be a good solution to start with a label regulation before setting strict ecodesign

limits.

Evaluation question 7: If there are significant differences in costs (or benefits) between

Member States, what is causing them? How do these differences link to the intervention?

Member State costs associated with the current regulations are primarily related to market

surveillance.

Even though all Member States have the same the obligation to perform market

surveillance of compliance with the Regulations, the actual level of market surveillance

varies between Member States. However, market surveillance for vacuum cleaners is

probably not a high priority for any Member State. The differences in market surveillance

costs are not linked to the interventions rather to the Member State priorities and limited

budget for market surveillance.

264

Conclusions on efficiency

The evaluation assessment has shown that the benefits of the regulations seem to

outweigh its costs, both for business, end-users and for society as a whole.

The manufacturers have invested in improvements of the products, but they have been

able to pass the costs on to the end-users. In addition, the manufacturers have benefitted

from an increased turnover compared to the situation without the regulations.

The increased performance has resulted in increased purchase prices for end-users, but

this is offset by the energy savings, which results in larger savings over the lifetime of the

vacuum cleaner i.e. lower total costs of ownership.

Member State costs associated with the regulation are primarily related to market

surveillance. These costs should be reduced, to incentivise market surveillance in all

Member States at a sufficient level. In addition, the market surveillance costs will be

reduced by establishing of the product registration database for energy related products

covered by Energy Labelling Regulations

364

Relevance

Evaluation question 1: To what extent is the intervention still relevant?

The objective of the regulations was to reduce the energy consumption of vacuum cleaners

and to provide consumers with reliable information about relevant performance parameters

for vacuum cleaners allowing them to make a better-informed choice. In addition, the

objective was to address the identified market failure i.e. that consumers buy vacuum

cleaners according to the rated input power, without necessarily getting the assumed

cleaning performance. This had resulted during the last 10-20 years in design of vacuum

cleaners with a much higher input power than required.

The objectives have to a large extent been met, but the regulations are still considered

relevant. The second power cap only entered into force in September 2017 and the vast

majority of the saving related to this requirement (and the regulation) have still not been

achieved. The same is the case for the labelling where the more ambitious energy classes

A+, A++ and A+++ was introduced on the label in September 2017.

Furthermore, it is most likely that without the Energy Labelling and Ecodesign Regulations

consumers will again buy vacuum cleaners according to the rated input power and that

manufacturers will be influenced to place vacuum cleaners on the marked with still

increasing power.

364

According the Energy Labelling Framework Regulation (EU) 2017/1369

265

Consumers are buying more and more vacuum cleaners of the types not covered by the

regulations (robot and cordless vacuum cleaners). For these types a tendency to have

excessive power consumption when standing still has been observed, often exceeding the

annual energy of mains-operated vacuum cleaners. This consumption has not been

decreased based on the Standby Regulation, which also shows the relevance of the

regulations for these products. Furthermore, it shows that robot and cordless vacuum

cleaners should be included in the scope of the regulations to avoid unnecessary standstill

energy consumption and low cleaning performance in these vacuum cleaner types.

Evaluation question 2: To what extent have the (original) objectives proven to have been

appropriate for the intervention in question?

The original objectives have been appropriate and have resulted in better designed

products without unreasonably high input power of vacuum cleaners on the market. In

addition, the marked failure has been corrected for vacuum cleaner types included in the

scope of the regulations.

Evaluation question 3: How well do the (original) objectives of the intervention (still)

correspond to the needs within the EU?

The objectives regarding energy savings and increased energy efficiency are in line with

European policies such as the 2030 Climate and Energy Policy Framework, that sets targets

for greenhouse gas emissions and improvement of energy efficiency at European level for

the year 2030 (at least 40% cuts in greenhouse gas emissions, and at least 32.5%

improvement in energy efficiency)

365

Evaluation question 4: How well adapted is the intervention to subsequent technological or

scientific advances?

A significant market trend for vacuum cleaners is a technology shift to more and more

robot and cordless vacuum cleaners. However, these types of vacuum cleaners are not

covered by the current Regulations. This undermines the Regulations’ achievements and

create inadequate market conditions.

The fact that robot and cordless vacuum cleaners are not in the scope of the current

Regulations mean that there is no EU system of information to end-users for these types

of vacuum cleaners, and the end-users will probably make their purchase decision

according to the power as was previous the case for the types of vacuum cleaners now

covered by the regulations, and not be aware of their hidden energy consumption, such as

in standstill. Advertisement with words such as “equally powerful to a corded vacuum

cleaner” or similar are seen more and more for cordless cleaners, but without specifics of

365

2030 Climate and energy policy framework. Conclusion – 23/24 October 2014. EUCO 169/14.

https://www.consilium.europa.eu/uedocs/cms_data/docs/pressdata/en/ec/145397.pdf

266

what “powerful” means. Furthermore the advertised runtimes and suction power are rarely

correlated for cordless cleaners with more than 1 power mode

366

. Hence, the regulations

are not very well suited for the different functionality of cordless vacuum cleaners.

Furthermore, for both robot and cordless vacuum cleaners, there are currently no durability

or dust re-emission requirements, and especially dust re-emission is very high, up to

almost 7% for some models

367

Only a very small share of vacuum cleaners are in the top energy classes of the label (A++

and A++). This means that the energy scale is still able to differentiate new and innovative

solutions with regard to energy efficiency. However, the review study will propose a

rescaling of the current A+++ to D scale to an A-G scale in order to align with the

Framework Energy Labelling Regulation.

Evaluation question 5: How relevant is the EU intervention to EU citizens

According to the consumer survey conducted by APPLiA in 2018, the energy label is of

relevance for a large share of consumers purchasing vacuum cleaners. A share of 25 %

anticipate that the label will be a crucial consideration next time they will buy a vacuum

cleaner, while 50 % anticipate that the label will be considered among other important

items, as shown in Figure 77.

Since this reflects only how important the label as whole is to end-users, and not the

importance of each of the parameters included in the label, it is not possible to say exactly

which of the parameters that end-users look for or how important each of them are in a

purchase situation.

366

See for example the comparison between measured and advertised runtime in the test by the Danish testing organization

TÆNK: https://taenk.dk/test/ledningsfrie-stoevsugere/testede-produkter

367

Test data provided by the GTT laboratories

267

Figure 77: Importance of the energy label for future vacuum cleaner purchases

Conclusion on relevance

The regulations continue to be relevant for reducing the energy consumption of vacuum

cleaners and contributes to achieve the targets in the EU 2030 Climate and Energy Policy

Framework

368

The Ecodesign Regulation prevents placing on the market of vacuum cleaners with too high

rated input power and the energy labelling creates further market pull and functions as a

comparable information source to compare products for end-users. Together the

regulations have resulted in better designed products being placed on the market.

However, a large share of the potential savings has not been achieved yet because the

second power cap and the energy label with A+++, A++ and A+ was not introduced before

late in 2017.

The consumers that participated in the APPLiA survey find the label relevant and a large

share anticipate that they will consider the information on the label next time they would

buy a vacuum cleaner.

368

2030 Climate and energy policy framework. Conclusion – 23/24 October 2014. EUCO 169/14.

https://www.consilium.europa.eu/uedocs/cms_data/docs/pressdata/en/ec/145397.pdf

268

13.2 Policy analysis

Stakeholders consultation

During the entire study, the study team has maintained a dialog with different

stakeholders.

A first stakeholder meeting was held on the 27

of June 2018 where representatives from

Member States, testing facilities, consumer organisations and manufacturers provided

input to the first four tasks. A second stakeholder meeting was held on the 5

of December

2018 where inputs were given to the complete draft report including all tasks. Comments

and inputs were taken into account in the report.

Telephone and face-to-face meetings have taken place with some individual manufacturers

who have provided input to the first four tasks, including among others APPLiA Nilfisk,

iRobot, BSH Hausgeräte, Bissel, Dyson, Groupe SEB and th CENELEC T59X WG6.

Furthermore data was received from several stakeholders, as well as collected by the study

team from especially consumer test organisations.

Policy measures

The following policy options have been considered for the policy scenarios:

• No action (‘Business-as-Usual’, BAU)

• Self-regulation

• Energy labelling

• Ecodesign measures

No action (‘Business-as-Usual’, BAU)

If no new action is taken, the existing Ecodesign Regulation 666/2013 for vacuum cleaners

remains in force. The Energy Labelling Regulation 665/2013 was annulled, and the savings

related to this Regulation will therefore not be achieved.

Tasks 1 to 6 of this review study show that the two regulations in force have worked on

pushing the EU market towards more efficient vacuum cleaners. However, further

improvement opportunities exist offered by existing BAT. Moreover, the scope limitation to

only mains operated vacuum cleaners leaves unutilised potential for energy and

performance improvements.

Overall, it is recommended to take action and review the existing ecodesign regulation in

force and investigate whether a new energy label could be beneficial. BAU is used as a

baseline to establish the potential savings, costs and impacts on consumers, industry and

employment.

269

Self-regulation

In Art. 15.3 b) of the Ecodesign Directive 2009/125/EC self-regulation, including voluntary

agreements offered as unilateral commitments by industry, is indicated as a preferred

option. However, this is subject to certain conditions stipulated in Article 17 and Annex

VIII to the Directive (e.g. market coverage by signatories, ambition level, etc.).

These conditions are not fulfilled for vacuum cleaners: none of the relevant stakeholders

expressed interest in self-regulation because the risk of ‘free-riders’.

Consequently, self-regulation has not been further considered as a policy option.

Ecodesign

The Ecodesign Regulation 666/2013 in force has made a positive impact as presented in

section 13.1. However, further improvement opportunities exist as presented in previous

tasks, especially for an expanded scope.

While the Ecodesign Regulation has removed the far majority of inefficient vacuum cleaner

from the market, there is still a large variety in products on the market, especially when it

comes to cleaning performance. Cleaning performance is currently represented by dust

pick up from a crevice in hard floor and embedded dust from a plush carpet. This is not

necessarily the situation consumers meet in real life, and it is therefore proposed to review

the current ecodesign requirements to reflect more consumer relevant performance and

possibly expand the scope. This review could also take the opportunity to introduce

resource efficiency requirements as discussed in previous tasks.

Details about proposed ecodesign policy options are presented in section 13.4.

Energy Labelling

The previous Energy Labelling Regulation 665/2012 that has been annulled also made a

positive impact while it was in force as presented in section 13.1.

The distribution of the energy label classes of the old energy label showed that there is still

potential for improving a large share of the market towards simultaneous better cleaning

performance and lower energy consumption.

Due to the result of the Dyson vs. European Commission court case, any new energy label

regulation would need to consider part load testing of the vacuum cleaner. This will most

likely lead to an increased uncertainty of measurements. Using the opportunity of instating

a new Energy Label Regulation, energy and performance classes could be adjusted to

reflect the actual uncertainty of the measurements, by making them wider than the

measurement uncertainties.

270

Moreover, other aspects related to consumer understanding can be incorporated to make

the label easier to understand by the consumer at the time of purchase. Finally, some

aspects about resource efficiency could also be integrated, as discussed in task 6.

The effect of the energy labelling regulation on performance and annual energy

consumption is clear, and it is therefore proposed to review the current energy label to

reflect the current and future technological progress and market trends. Details about

proposed energy labelling policy options are presented in section 13.4.

13.3 Baseline scenario - BAU

In order to estimate the effect of any changes of the regulations, a base line scenario with

the current regulations was established. The baseline scenario, or Business As Usual (BAU),

reflects the market development, energy consumption and resource consumption of

vacuum cleaners if no changes are implemented to the current Ecodesign Regulation. The

BAU is used to compare the effect of the policy option scenarios presented in the next

sections. All scenarios are modelled from 2015 to 2030, including the BAU scenario.

The BAU scenario is built on the data presented in task 2-5 of the current market and

product characteristics. The following assumptions are made regarding the development

from 2018-2030:

- Sales (and thus stock) will follow the trends presented in task 2

- Robot vacuum cleaners are sold as hard floor vacuum cleaners only

- The average AE values for the various types of vacuum cleaners will develop as

shown in Table 98

Table 98: Development of average AE values for household mains-operated and commercial

vacuum cleaners 2020-2030

Product type

2020

2025

2030

Cylinder household

Upright household

Handstick mains-operated

Weighted average mains (base case 1)

Commercial (base case 2)

Cordless (base case 3)

Robots (hard floor only) (base case 4)

The annual energy consumptions in Table 98 are all based on the formulas given in chapter

3, which includes corrections for the performance. For both robot and cordless vacuum

cleaners, the maintenance mode consumption makes up around half of the annual energy

consumption. Even though the poor carpet performance is removed from the robot

calculation (because they are assumed to be sold as hard floor only), the generally high

271

energy consumption while cleaning (around 40 Wh in the 20 m

test room), gives rise to

high energy consumption

369

The total annual energy consumption for the entire stock in the BAU scenario can be seen

in Figure 78. The line depicting the total, takes into account only the four base cases and

not the dotted line for robot cleaners including carpet performance, which is also shown in

Figure 78. It should be noted that the energy consumption for commercial cleaners is based

on the actual use pattern (300 cleaning cycles per year), as described in task 3, rather

than the label AE value, which is based on around 50 hours per year, as explained in task

As seen from the BAU scenario of the energy consumption, the household mains-operated

vacuum cleaners are by far the type of vacuum that gives rise to the largest energy

consumption in the EU28, which is due to the large stock of these products. However, when

going towards 2030, the stock is slowly shifted towards cordless and robot vacuum

cleaners. With the simultaneous increasing battery size of cordless products, the energy

consumption of the cordless stock is expected to be close to that of the mains-operated

stock by 2030, partly because end-users exchange their corded vacuum cleaner with a

cordless one.

This also results in an increase of the overall energy consumption in EU28 (i.e. the purple

line showing the total consumption of the entire stock), both because of an increase of the

total stock increases and because the energy consumption of the cordless and robot

products is not covered by the current Regulations.

The energy consumption of commercial vacuum cleaners decreases only very slowly, and

is more or less linear until 2030 due to the assumed constant sales and slowly decreasing

AE values.

369

Note again the differences in how the energy consumption is measured for the different vacuum cleaner types, which are not

directly comparable, i.e.

272

Figure 78: Expected energy consumption development in the BAU scenario, 2015-2030

The corresponding emission of greenhouse gasses (GHGs) for the four base cases in the

BAU scenario can be seen in Figure 79. The GHG emissions does not increase at the same

rate as the energy consumption, because it is expected that more and more renewable

energy will be used in the electricity production in the EU.

Figure 79: Expected annual greenhouse gas emissions in the BAU scenario 2015-2030

Another effect of the changes in energy consumption is the change in costs for the end-

users. Decreasing energy consumption results in savings for the end-users in terms of

electricity costs, however, depending on the technology steps and development needed to

achieve the energy savings, the product price will increase.

273

For mains-operated household and commercial vacuum cleaners, which have been in scope

of the Regulations for some years now, data exist to make a correlation between

development in AE values and consumer purchase price. Based on the GfK data, this

correlation was calculated to be 2,1 € increase in purchase price per kWh decrease in the

AE value for household mains-operated vacuum cleaners

370

. For commercial vacuum

cleaners, the correlation is 2,7 € price increase per kWh AE decrease.

According to the MEErP method the consumer expenditure cost is calculated as all costs

paid for by all end-users in the EU28 each year. Hence, the purchase price is paid the year

the product is sold (thus based on sales and purchase price every year), while the energy

consumption is paid over the course of the lifetime of the vacuum cleaners (thus based on

stock energy consumption and electricity prices every year). Repair and maintenance costs

as well as auxiliary costs (bags and filters) is split evenly over the lifetime of the vacuum

cleaners.

Since the sales and stock of mains-operated vacuum cleaners decreases from 2016 to

2030, the end-user expenditures in EU decreases as well as seen in Figure 80. The stock

of commercial vacuum cleaners decreases only slightly, so the decrease in electricity cost

from 2016 to 2030 is a mix of decreasing stock and replacement of the stock with more

energy efficient products. For cordless and robots the increase in consumer expenditures

is mostly due to the increasing sales and for cordless to a small extend the increasing

energy consumption per product because the battery and motor size is expected to

increase. This is also seen by especially the purchase costs increasing significantly.

Figure 80: Expected development in consumer life cycle costs in the BAU scenario from 2016

to 2030

370

Calculated

274

13.4 Policy scenarios for energy efficiency and performance

In this section three policy options, PO1, PO2 and PO3, are analysed and compared to the

Business as Usual (BAU) scenario. These three options regard only the energy and

performance-related requirements. All resource related aspects are treated in policy option

PO4. The results of the scenario calculations are the impacts in EU28 of the policy options.

PO1 and PO2 include both ecodesign and energy labelling requirements, while PO3 is a

scenario without Energy Labelling, but with stricter Ecodesign limits. The key differences

between PO1 and PO2 are the AE and rated power thresholds. In PO1 the thresholds for

AE are set at 36 kWh/year and 750 W for rated power, as suggested by some stakeholders.

In PO2 the current thresholds for AE of 43 kWh/year and the power limit of 900 W are

maintained. All other thresholds are the same in the two policy options, as seen in Table

99. PO3 follows the thresholds of PO1.

All three scenarios also include a more specific division of commercial vacuum cleaners in

terms of test methods, thresholds and calculation methods (see Section 9.7.4).

Table 99 shows the ecodesign limits in each of the policy options, represented by numbers,

as well as the parameters that should be included in the energy label in each policy option,

represented by blue colour. In Table 99 below each of the requirements are presented one

by one.

Table 99: Policy Option 1, 2 and 3: Energy and performance related requirements.

Ecodesign

Parameter

Commercial

Mains-

operated

household

Cordless

Robot

Common parameters for Policy Options 1, 2 and 3

dpu

≥0.98

dpu

≥0.75

Debris hard

floor*

≥0.40

≥0.80

Debris

carpet*

≥0.75

Dust re-

emission

≤0.8%

Tier 1: ≤3%

Noise

≤78 dB(A) or

≤80 dB(A) if the

product is

equipped with a

beat and brush

nozzle

≤78 dB(A) or

≤80 dB(A) if the

product is

equipped with a

beat and brush

nozzle

≤85 dB(A)

≤65 dB(A)

Measured from

1.6 m distance

Decrease in

air flow with

≤15%

275

Ecodesign

Parameter

Commercial

Mains-

operated

household

Cordless

Robot

Motion

resistance

40N

Maintenance

power

≤0.5 / 1.0 /

2.0 W

≤0.5 / 1.0 / 2.0

Coverage

factor

≥80.00%

Policy Option 1

Annual

Energy, AE

≤36 kWh/year

Energy

Index, EI

0,8 m

/min

Rated power

≤750 W

Energy labelling

Policy Option 2

Annual

Energy, AE

≤43 kWh/year

Energy

Index, EI

0,76 m

/min

Rated power

≤900 W

Energy label

Policy Option 3

Annual

Energy

≤36 kWh/year

Energy

Index, EI

0,8 m

/min

Rated power

≤750 W

No Energy Labelling

*Based on very limited data, and require additional testing before final requirements are set.

Requirements

Annual energy and rated power for household vacuum cleaners

The limits on annual energy consumption and rated power input for mains operated

vacuum cleaners were introduced because of the ever increasing wattages used for

marketing purposes. However, for cordless and robot vacuum cleaners there is a natural

limitation to the possible motor power, because they run on batteries, and large motors

will reduce the run time per charge. The rated power is therefore also difficult to measure

for these products, because it cannot be measured as power drawn from the grid while the

cleaner is operating. Hence, setting rated power limits for robots and cordless is difficult,

and will most likely not result in the same energy savings as for mains operated vacuum

cleaners.

276

Requirements with regard to annual energy and rated power are therefore only set for

mains operated vacuum cleaners. PO2 includes the same values for AE and rated power

as the current ecodesign regulation, whereas PO1 and PO3 includes stricter requirements

for both parameters, namely at 36 kWh/year and 750 W.

Even though ecodesign requirements with regard to annual energy are not set for cordless

and robot cleaners, it is suggested for the policy options including energy labelling (i.e.

PO1 and PO2) to base the label classification on the AE value, calculated according to the

formulas presented in task 3.

Energy Index for commercial vacuum cleaners

For commercial vacuum cleaners it is suggested to replace the AE calculation with the

equations described in section 9.2.1, resulting in an Energy Index, EI, instead of an annual

energy consumption, in order to make the calculation and possible energy label more

relevant to the end-users. Based on measurement data from commercial vacuum cleaner

manufacturers such a measure has much higher relevance to the commercial customers

and the EI results in clear distinguishability between good and not so good machines. The

EI should be used both for Ecodesign thresholds instead of AE as well as for the energy

label scale in policy options where this is relevant.

Dust pick-up hard floor

The dust pick-up on hard floor is suggested to be maintained at ≥0,98 for mains operated

vacuum cleaners in all policy options. It is not suggested to increase this level, since there

are still very few vacuum cleaners that have very good dust pick-up performance and

annual energy values simultaneously.

For cordless and robot vacuum cleaners, very limited data exist on dust pick-up, however

data for cordless vacuum cleaners show that by the far majority of the models perform

well below the average mains operated vacuum cleaner. Setting too strict ecodesign

performance requirements for cordless and robot vacuum cleaners would remove many

products from the market, reducing consumer choice significantly.

For robots the test method is different (flat floor, no crevice), and the results are therefore

not directly comparable. The available results show relatively low performance, even on

flat floor (around 60% on average), measured when the robot has moved over the floor

once (first pass). It is therefore not suggested for the policy options with energy labelling,

to show the dust pick-up results on the label, rather than setting an ecodesign requirement.

Dust pick-up carpet

The dust pick-up on carpet is likewise suggested to be maintained at the current ≥0.75 for

mains operated vacuum cleaners. Since the average performance of cordless and robot

277

vacuum cleaners are very poor on carpet compared to mains-operated vacuum cleaners,

and there is no standard to measure it, it is suggested not to implement ecodesign

requirements, when there is an option for the label, i.e. instead include dust pick-up on

carpet on the label in PO1 and PO2. However, in PO3, where there is no energy labelling,

an ecodesign requirement is suggested instead.

Debris pick-up

In order to make the performance tests more consumer relevant, it is suggested to add a

debris pick-up ecodesign requirement for commercial, mains operated and cordless

vacuum cleaners. This is expected to put even more emphasis on a good nozzle design

that gives good performance in real life as well.

It is suggested to set the requirements at the same value for mains operated household

and cordless vacuum cleaners. This is not possible for dust pick-up, since cordless cleaners

are not intended for deep cleaning, as simulated by the crevice test on hard floor and the

embedded dust test on carpet. Cordless cleaners are, on the other hand, designed to be

good at visible cleaning, which is represented by the debris pick-up. Therefore, both the

mains operated and the cordless cleaners should be capable to quite easily reach the

suggested requirements for debris pick-up.

The debris pick-up is kept as a separate parameter and not included in the AE calculation

or averaged with the dust pick-up. According to manufacturers it is easier for most vacuum

cleaners to reach a good debris pick-up than a good dust pick-up (deep cleaning).

Therefore, including debris in the formula might result in some good AE values, covering

over quite low dust pick-up performances due to a good debris pick-up. Hence in order to

avoid manufacturers focusing too much on debris instead of deep cleaning, it is suggested

to keep debris pick-up separate, but included as a minimum threshold, to avoid the very

specialised nozzles developed to perform well in the dust pick-up tests.

For robots, it is a bit different, primarily because the test method is different. Furthermore,

the consumer is not performing the vacuuming, and robot vacuum cleaners tend to balance

large debris and fine debris pickup. As for the cordless they are thus designed to be good

at debris pick-up, and for maintenance of a relatively clean area rather than for intense

weekly/monthly cleaning. However, while there is limited data for mains operated and

cordless cleaners, there is not data at all for robots regarding debris pick-up. It is therefore

not recommended to set a minimum threshold for robots. Once a test is developed, it could

be required to make the information available in a data sheet, in the user manual or similar.

For commercial vacuum cleaners it is suggested to set the debris pick-up requirement

based on the specific commercial test standard described in section 9.7.4.

278

Dust re-emission

The dust re-emission is suggested to be improved for mains operated vacuum cleaners,

based on average label data received by GfK. Dust re-emission is an important parameter

for human health, especially for sensitive demographic groups such as children or people

with allergies.

This requirement, however, cannot be applied to cordless or robot vacuum cleaners. For

robots there is no test method available, and for cordless cleaners it is very difficult to

reach such low dust re-emission values as the mains. The main reason is the limited

physical size of the cordless cleaners along with the limited power. High efficiency filters

restrict the airflow and create suction performance losses. Therefore a higher limit value is

recommended for cordless vacuum cleaners.

Noise

The noise levels of most cylinder type vacuum cleaners are well below the current 80 dB(A)

limit, and it is thus suggested to lower the requirements to 78 dB(A) with the exception of

vacuum cleaners with active beat and brush type nozzles. These are primarily used for

upright vacuum cleaners, which is the reason they have had many difficulties complying

with the current 80 dB(A) limit.

For cordless vacuum cleaners, the limited data available shows higher noise levels than for

mains operated. This is again due to the limited size of the products, leaving little space

for adding sound insulating material to reduce the noise. Furthermore, adding such noise

insulation would also restrict the airflow, leading to lower suction performance. Therefore

it is suggested also for noise, to impose less strict requirements on cordless vacuum

cleaners.

For robots, on the other hand, the noise is lower on average, according to the available

data. Measured from 1.6 meter distance, the average is well below the 80 dB(A) and a

limit of 65 dB(A) is therefore recommended for robot vacuum cleaners.

Loaded air power

The biggest concern regarding consumer relevance and part loading is the big variation in

how each individual product reacts to the dust loading, and the risk that some vacuum

cleaners might vary significantly from the test values with empty receptacle when used in

real life. Since there are large difficulties in developing a part load test method that is

repeatable and reproducible, different approximations was discussed (see task 3). One of

them is the air power, or suction power, measured in watts. One way to accommodate the

problem with vacuum cleaners that decrease rapidly in performance with dust loading, is

to set an ecodesign requirement for the maximum decrease in air power when the vacuum

279

cleaner is full, compared to when it is empty. Initial measurement data from the Round

Robin Tests on part and full load air performance measurements show a large variation in

how much suction power is lost when filling the receptacle of different vacuum cleaner

models. Based on this, a limit of 15% reduction is suggested, but more data should be

collected before setting the final requirement.

Motion resistance

It is suggested to add a cap on motion resistance on carpet in order to avoid specialised

nozzles that are not practically useful for end-users. The motion resistance cap is set at 40

N. This value is based on measurements on the Wilton carpet, and measured at a constant

speed of 0.5 m/sec in accordance with the current test standards. The motion resistance

should be measured simultaneously with the dust pick-up on carpet, to avoid different

settings or detection of test conditions optimising to each of the measurements.

Maintenance power

The maintenance power requirements are applicable to both cordless and robot vacuum

cleaners. The maintenance power is the power consumption in the so-called ‘charged and

docked’ mode, i.e. when the vacuum cleaner is standing in the docking station fully

charged. This mode includes any trickle charging, standby consumption etc. that the

vacuum cleaner might need. This mode is the one the cordless and robot vacuum cleaners

are in most of the time, and as seen by the BAU scenario calculations this is also the most

energy consuming mode today. The maintenance power can range between less than 0,5

W to 8 W. Hence the average value presented in the BAU covers a large variance, and

large amount of energy is spent in this mode by some models. It is suggested to measure

this energy consumption as an average consumption over 24 hours

371

, in order to allow for

the docking station to consume more energy for short time spans to perform relevant

tasks.

The maintenance mode requirements are suggested to follow that of the 2019

requirements in the standby regulation

372

. While vacuum cleaners are in principle covered

by the standby regulation, the maintenance of battery power with so-called ‘trickle

charging’ or other functions, it can be used as a loophole to state that this is not a standby

mode.

The proposed standby requirements are 0.5 W, with info display 1.0 W, and with networked

connection 2.0 W. This includes consumption of the power supply, docking station and the

371

To avoid circumvention, it should not be allowed to program the appliance in a way that the standby consumption in the first

24-hour period is different, e.g. by not activating function such as software updates until after the first 24-hour period.

372

https://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex%3A32013R0801

280

robot itself. Since it has been argued that at least robot vacuum cleaners are covered by

the current networked standby requirements in the standby regulation, it should already

be possible to test and comply with these requirements.

Run time

The run time is a measure of how long the vacuum cleaner can be used when it is fully

charged, before it needs charging again. In order to ensure that consumers are not mislead

by different statements regarding run times, which is an important marketing parameter

for battery driven vacuum cleaners, it is suggested to develop a standardised way of

measuring (operational) run time, and include it in the energy labelling in PO1 and PO2.

The run time should be measured in the same mode as the dust- and debris pick-up.

Coverage factor

The coverage factor applies only to robot vacuum cleaners and is a measure of how much

of the floor area in a given room the robot covers in its cleaning cycle. At low coverage

rates, the cleaning performance measured in tests is not representative of the actual

cleaning, because parts of the floor are not covered by the robot at all. No minimum

requirement is recommended for the coverage factor, but it is recommended to include the

coverage factor in the energy labelling in PO1 and PO2. Furthermore, the coverage factor

is included in the AE calculation for robots.

Energy saving potentials

Based on the above requirements and the data presented throughout the study, the impact

of PO1, PO2 and PO3 on energy consumption in EU28 has been derived and compared to

the BAU scenario. As seen from Figure 81, the energy consumption in all three scenarios

is lower than in the BAU scenario, however the savings in PO3 (1.44 TWh/year in 2030) is

less than half of the savings in PO1 (3.99 TWh/year in 2030) and PO2 (3.84 TWh/year in

2030).

281

Figure 81: Energy consumption in PO1, PO2 and PO3 compared to BAU from 2018 to 2030

As shown in the BAU scenario, the greenhouse gas emissions follow the energy

consumption, but with the assumption of more renewable energy in the electricity mix in

the future. The comparison of GHG emissions in the scenarios can be seen in Figure 82.

The PO1 scenario, which has the largest savings, results in savings of around 1.3 Mt

/year by 2030, corresponding to 28% of the annual vacuum cleaner stock greenhouse

gas emissions in the BAU scenario.

Figure 82: GHG emissions in PO1, PO2 and PO3 compared to BAU from 2018 to 2030

The energy savings in both PO1 and PO2 are primarily caused by the increased energy

efficiency of cordless and robot vacuum cleaners, while only minor energy savings are

attributed to setting stricter ecodesign requirements for mains-operated vacuum cleaners

as seen from Table 100. Hence the lower motor power threshold of 750 W and the limit for

the Annual Energy of 36 kWh/year in PO1 does not contribute to significant energy savings

compared to maintaining the current thresholds, illustrated by the small difference between

282

PO1 and PO2. This small saving potential for setting the stricter limits, is because many

products are already way beyond the ecodesign limit. The reason for this can be largely

attributed to the Energy Labelling Regulation, and the fact that more than 50% of the

products sold today are labelled in energy class A

373

This is also the reason why setting only the Ecodesign limits and removing the energy label

results in only half the savings as having both regulations, as modelled in PO3. It is

assumed that without an energy label, the argument for selling more expensive products

based on performance and the incentive to develop products with performance/energy

consumption beyond the limit values, are removed. While the potential savings for cordless

and robots is only a few percentage points lower than in the PO1 and PO2 scenarios, the

increasing AE values (up to near the limit value) for household mains-operated and

commercial vacuum cleaners causes the energy consumption for these vacuum cleaner

types to increase.

The majority of the energy savings in all three policy options are achieved by including

cordless and robot vacuum cleaners in scope of the Regulations and secondarily by re-

instating an energy labelling regulation for mains operated vacuum cleaners. Since the

cordless and robot vacuum cleaners are expected to increase in market share and annual

energy consumption, it is important to include them in the regulation(s), not only for their

energy saving potential, but also to provide consumer protection and a level playing field

among products when cordless and robots starts to compete with and replace the mains-

operated vacuum cleaners.

Table 100: Energy savings for each base case in 2030 for PO1, PO2 and PO3 in EU28

2030 energy consumption,

TWh

Annual savings in 2030,

TWh

Annual savings, %

BAU

PO1

PO2

PO3

PO1

PO2

PO3

PO1

PO2

PO3

Household

mains

6.71

5.30

5.41

6.28

1.41

1.31

0.44

21%

19%

Commercial

3.88

3.18

3.23

3.78

0.70

0.65

0.10

18%

17%

Cordless

2.15

0.83

1.32

61%

62%

Robots

1.18

0.62

0.49

0.56

0.69

48%

59%

Total

13.93

9.94

10.09

11.38

3.99

3.84

2.55

29%

28%

18%

Table 101 shows the energy consumption of cordless and robot cleaners in BAU and PO1

(the strictest scenario), divided into maintenance power consumption and power

consumption for cleaning (including charging). In 2018 around half of the annual energy

consumption is associated with the maintenance power for cordless and ¾ for robots. In

373

Based on the 2017 label. Se task 2.

283

the BAU scenario maintenance mode is considered unchanged (but taking into account

2019 requirements in the standby regulation). In the PO1 (and other) scenario

maintenance consumption is reduced drastically towards 2030 to less than half of the 2018

values. This is due to the proposed maintenance mode requirement. Also the power for

cleaning and charging is decreasing due to better dpu and because the power supplies are

expected to become more efficient in order to bring the maintenance power down.

Table 101: Energy consumption of cordless and robot vacuum cleaners in BAU and PO2,

kWH/year

Base case

Mode

2018

2020

2025

2030

Cordless,

BAU

Maintenance mode, kWh/year

Cleaning, kWh/year

AE, kWh/year

Cordless,

PO1

Maintenance mode, kWh/year

Cleaning, kWh/year

AE, kWh/year

Robots,

BAU

Maintenance mode, kWh/year

Cleaning, kWh/year

AE, kWh/year

Robots ,

PO1

Maintenance mode, kWh/year

Cleaning, kWh/year

AE, kWh/year

Total consumer expenditure

While the energy saving potential is higher in PO1 and PO2 than in PO3, all three scenarios

result in roughly the same monetary savings for the end-users compared to the BAU

scenario. Figure 83 shows the total end-user expenditure for all vacuum cleaners in the

EU. The cost is composed of total purchase price each year and the electricity cost and

maintenance cost of the stock per year (i.e. vacuum cleaners sold the previous years). This

is also why in the PO3 scenario the total costs drop under the PO1 and PO2 scenarios in

2019-2026+, because without the energy label the AE values increase slightly, causing

lower purchase cost, but the energy consumption of the stock is still low. However, when

the stock is replaced with the PO3 products (i.e. no Energy Labelling, only Ecodesign), the

costs exceeds those in PO1 and PO2 (around 2027), due to the higher energy consumption.

This trend would continue to be more pronounced in the years following 2030, if the model

was forecasted further.

The graph in Figure 83 shows, that in the long term (after 2030) the end-user costs will

be lowest in the PO1 scenario, but still quite similar to the PO2 and PO3 scenarios. The

reason for the small difference is primarily due to how the energy costs are calculated:

284

with a small annual increase in electricity prices, but a discount rate of 4%, which is larger

than the increase in the electricity price, hence giving low value to energy savings.

Figure 83: Total end-user expenditure for all vacuum cleaners in EU28 each year from 2018-

2030.

For all the scenarios, the consumer expenditure is lower than in the BAU, however the

effect varies between the base cases, as seen in Table 102. For all scenarios the effect on

mains-operated household vacuum cleaners is less than 1%. Even though the energy

consumption decreases for these products, the increase in purchase cost more or less level

out the cost savings related to use of the product for end-users. For the commercial

cleaners the effect is larger in PO1 and PO2 than in PO3, since only limited additional

ecodesign requirements are set, while the energy label has a larger effect.

For cordless and robot vacuum cleaners, the effect on user expenditures is more or less

the same in all three policy scenarios, because both ecodesign and energy labelling

requirements are new to these product categories. However, the energy savings are larger

in the label scenarios (PO1 and PO2), than in the PO3. The difference between PO1 and

PO2 is very small though, showing again that when there is an energy label, the effect of

the stricter ecodesign requirements is limited.

285

Table 102: EU User expenditure for each base case

Consumer expenditure, Million €

Year

2030 Savings

Base case

Scenario

2018

2020

2030

Mains-operated

household

BAU

7,699

6,130

5,060

PO1

7,699

6,259

5,054

PO2

7,699

6,248

5,067

PO3

7,699

6,179

5,048

Commercial

BAU

2,519

2,272

2,299

PO1

2,519

2,197

2,206

PO2

2,519

2,201

2,213

PO3

2,519

2,252

2,281

Cordless

BAU

2,013

3,558

4,715

PO1

2,011

3,401

4,435

PO2

2,011

3,401

4,435

PO3

2,011

3,392

4,434

Robot

BAU

823

1,449

1,968

PO1

823

1,389

1,849

PO2

823

1,389

1,849

PO3

823

1,388

1,846

Consumer health potentials

As well as the energy savings, which can be directly correlated to the consumer

expenditures, the parameters related to consumer health are also affected by the

requirements suggested in the policy options.

Table 103 shows the effect of each policy option on the average noise of each vacuum

cleaner type. Since there is a significant difference between the different mains operated

vacuum cleaner types, these are shown separately in the table. As explained above, it is

recommended to set lower limits for vacuum cleaners that do not have beat and brush

nozzle, but maintain 80 dB for those that have. There have already been great difficulties

286

for these vacuum cleaners to have a noise level below the maximum of 80 dB, while those

with other nozzle types are typically lower.

For most cordless vacuum cleaners, the noise levels are higher than 80 dB at the moment,

due to the light construction that does not allow for much sound-insulating material, and

the requirement of 85dB s therefore suggested. It is not expected that the average values

will change with only the ecodesign requirement, based on the data available. However,

with energy labelling, it is expected to decrease a further on average. Robots generally has

lower noise values, and therefore 65dB is suggested as a limit. Again, ecodesign alone is

not expected to change the average values, but energy labelling is expected to change it

slightly. By implementing the label, more information about the noise levels of products on

the market will also be available, helping to set more realistic requirements in the future.

Table 103: Average noise levels of each vacuum cleaner type in 2018, 2025 and 2030 in the

policy scenarios

Average noise, dB(A)

Sales year

2018

2025

2030

Cylinder

Mains operated household

BAU

78.8

PO1

78.8

76.9

76.6

PO2

78.8

76.9

76.6

PO3

78.8

77.4

Upright

Mains operated household

BAU

80.0

PO1

80.0

PO2

80.0

PO3

80.0

Handstick

Mains operated household

BAU

79.9

PO1

79.9

79.4

79.0

PO2

79.9

79.4

79.0

PO3

79.9

79.7

Commercial

BAU

79.1

PO1

79.1

77.2

76.9

PO2

79.1

77.2

76.9

PO3

79.1

77.6

Cordless

BAU

83.6

PO1

83.6

83.0

82.6

PO2

83.6

83.0

82.6

PO3

83.6

83.0

Robot

BAU

60.8

PO1

60.8

59.8

58.6

PO2

60.8

59.8

58.6

PO3

60.8

60.0

287

The average dust re-emissions for the different vacuum cleaner types are shown in Table

104. The dust re-remissions are expected to decrease slightly for the mains-operated

household and commercial cleaners due to the new limit of 0.8. For the cordless products,

the decrease is much greater, because the values today are very high (for some machines

up to around 8% dust re-emission has been measured

374

). Hence, the requirement of

maximum 3% dust re-emission is expected to bring the average values down at least 1%-

point (PO3), while the energy label is expected to decrease levels even further (PO1 and

PO2).

Since dust re-emission cannot yet be measured for robot vacuum cleaners, no data is

available, and consequently no requirements have been suggested.

Table 104: Average dust re-emission levels of each vacuum cleaner type in 2018, 2025 and

2030 in the policy scenarios

Average dust re-emission, %

Sales year

2018

2025

2030

Cylinder

Mains operated household

BAU

0.35%

PO1

0.35%

0.30%

PO2

0.35%

0.30%

PO3

0.35%

0.30%

Upright

Mains operated household

BAU

0.42%

PO1

0.42%

0.36%

PO2

0.42%

0.36%

PO3

0.42%

0.36%

Handstick

Mains operated household

BAU

0.72%

PO1

0.72%

0.61%

PO2

0.72%

0.61%

PO3

0.72%

0.61%

Commercial

BAU

0.35%

PO1

0.35%

0.30%

PO2

0.35%

0.30%

PO3

0.35%

0.30%

Cordless

BAU

4.19%

PO1

4.19%

2.60%

PO2

4.19%

2.60%

PO3

4.19%

2.99%

Conclusions

Based on the scenario analyses above, the energy savings in PO1 and PO2 are quite similar

and approximately double that of the PO3 scenario. At the same time consumer

374

Data supplied by GTT laboratories in accordance with IEC draft standard

288

expenditure is roughly the same in all three scenarios (might be highest in PO3 in the long

term). The user health effects of noise and dust re-emissions are similar in PO1, PO2 and

PO3 for mains operated and commercial vacuum cleaners, but for cordless PO1 and PO2

result in the lowest user health impact.

Hence, on all performance parameters energy labelling is expected to lead to larger

benefits for the end-users. Based on this a policy option with energy labelling is

recommended. However, this is dependent on the development of a methodology to either

measure or simulate properly the effect of part loaded dust receptacle.

The difference between the two scenarios including energy labelling (PO1 and PO2) is only

the stricter AE limit values, but the effect of this is limited when there is a label pulling the

market towards better performance. At the same time, as shown in task 6, these stricter

Ecodesign limits do not lead to lower life cycle costs for the end-users. Hence, if a label

scenario is chosen, it is recommended to follow PO2. If it is not possible, however, to

implement an energy label, the stricter AE requirements (as in PO3) are still required, to

at least obtain some of the potential energy savings.

According to the standardisation group working on robot vacuum cleaner standardisation,

the test standards are still not mature to be used for Ecodesign and Energy Labelling

purposes, partly because the technology is still very new and rapidly evolving, and partly

because experience with testing is still too limited and repeatability data is not yet available

(Round Robin Tests (RTT) ongoing). However, seeing that other fast developing

technologies such as computers are also covered by Ecodesign Regulations, the speed of

development of the technology should not be an issue in itself. On the other hand, the lack

of robust testing methods could be a barrier for including robot vacuum cleaners in scope

of the regulation, but this could be solved by considering a longer implementation time

frame.

Taking the test development into account only some of the performance parameters are

suggested for robot and cordless vacuum cleaners. At the very least it is highly

recommended that both cordless and robot vacuum cleaners are included in scope of the

Ecodesign Regulation with requirements on the maintenance mode power consumption and

preferably with the range of performance parameters covered in PO2.

Label classes

For a new energy label it is suggested to use the same classes as in the now annulled

Energy Label Regulation, but use the letters A-G for the scale, as shown in Table 105.

In addition to the energy classes, the assumed market distribution of the four base cases

among the energy classes by the time of application is shown in Table 105. For the mains-

289

operated household cleaners the distribution is based on forecasting the label distribution

from GfK for 2016 to 2021. For the other base cases, where there was no data regarding

distribution, it was based on the average AE value. Cordless vacuum cleaners are assumed

to be the only vacuum cleaner type that can achieve the A+++ (or A) rating, because of

the small motors (low ASE) and the expected drastic decrease in maintenance power.

However, due to the very low performance (in terms of dpu), these machines will not be

A/A/A/A (annual energy/dpu

/dpu

/dust re-emission). Hence, no such products are

expected to exist upon entry into force of the revised regulation

375

Table 105: Rescaling of the energy label and assumed distributions

Current

label

classes

Interval

New

label

classes

Assumed 2021 market distribution

Mains-

operated

household

Cordless

Robots tier 1

Robots tier 2

A+++

≤ 10

0.0%

A++

10 < AE ≤ 16

1.0%

A+

16 < AE ≤ 22

2.0%

21%

22 < AE ≤ 28

61.0%

54%

28 < AE ≤ 34

22.0%

11%

10%

34 < AE ≤ 40

7.0%

14%

18%

40 < AE

7.0%

75%

61%

For the other performance parameters it is suggested to change the scales shown on the

label. This is primarily based on the findings of the standardisation work, which shows that

the expanded uncertainties of the measurements exceed the label class width. It is

therefore suggested to reduce the number of classes on each scale from seven to four to

make room for broader classes. The suggested performance class intervals are shown in

Table 106. As noted previously, the standardisation groups are working on a suggestion

for changing the dust re-emission scale to a logarithmic scale rather than a linear one.

Table 106: Suggested label classes for the performance parameters on the energy label

Performance

class

Dust pick up on carpet

(dpu

)

Dust pick up on hard

floor (dpu

)

Dust re-emission (dre)

dpu

>0.91

dpu

>1.11

dre≤0.02%

0.85≤dpu

<0.91

1.07≤ dpu

<1.11

0.02%<dre≤0.2%

0.80≤ dpu

<0.85

1.02≤ dpu

<1.07

0.20%<dre≤0.60%

dpu

<0.80

dpu

<1.02

dre>0.60%

375

This is also the case even if the performance classes are rescaled as suggested in Table 106, since the criteria for A remains

the same as in the previous, annulled Energy Labelling Regulation.

290

These suggestions are based on RRT measurement data of parameters with an empty

receptacle. The standardisation group is currently working on an RRT with partly loaded

receptacle, but results of these measurements are not published before the end of this

review study. These data can therefore not be shown here, but it is suggested to revisit

and re-evaluate the classes suggested in Table 106 when the data is available.

Commercial vacuum cleaners

For commercial vacuum cleaners a new EI is suggested to replace the current AE value,

and thus different classes would need to be applied. Since a higher EI value equals a higher

productivity (m

/min) and thus results in a lower energy consumption, the higher the

value, the better. This is opposite of the AE scale.

Table 107: Energy label classes for the new commercial EI scale

Label class

Interval

Estimated market distribution by

2021

≥4,3

4,3 > EI ≥ 3,6

3,6 > EI ≥ 2,9

2,9 > EI ≥ 2,2

30%

2,2 > EI ≥ 1,5

40%

1,5 > EI ≥ 0,8

20%

0,8 > EI

13.5 Policy scenario for resource efficiency

In this chapter the policy scenario regarding resource efficiency is analysed and compared

to the Business as Usual (BAU) scenario. The requirements in PO4 are all aiding in

increasing product life through durability and reparability requirements.

Table 108: Policy Option 4: resource efficiency requirements

Ecodesign

Parameter

Requirements for

mains-operated

household and

commercial

Requirements for

cordless

Requirements for

Robots

Motor life

500 hours

Hose oscillation

40,000 oscillations

when a hose is

present

Battery lifetime

600 cycles and

maintain 70%

capacity

600 cycles and

maintain 70%

capacity

Spare part

availability

8 years (household)

5 years (commercial)

6 years

291

Easy

changeable

repair-prone

parts

Hose

Power cord roll-up

Permanent filters

Handle

Active nozzles

Battery (4 years)

Hose

Permanent filters

Handle

Active nozzles

Battery (4 years)

Wheels

Brushes

Permanent filters

Information

requirements

on repair

How to repair/change

repair-prone parts

How to

repair/change

repair-prone parts

and how to best

ensure battery

longevity

How to repair/

change repair-

prone parts and

how to best ensure

battery longevity

Information

requirements

on recycled

material

Share of recycled plastic content

One of the important parameters of increasing product life is the availability of spare parts.

In task 6 is it stated that it is feasible to increase the current motor lifetime from 500 hours

to 550 hours

376

for mains-operated household and commercial vacuum cleaners since this

can be achieved at low costs (i.e. still achievable with universal motors) and this is enough

for a product lifetime of >10 years. For robot and cordless vacuum cleaners, a lifetime of

at least 600 hours is suggested as a requirement in Task 6, to ensure a lifetime of 6 years

with 100 hours of use per year. In task 6 it was not considered to be a problem since the

motor types used in cordless and robot vacuum cleaners often have much longer lifetimes

than the universal motors with carbon brushed used in main-operated vacuum cleaners.

However, based on stakeholder inputs it is argued that a motor lifetime of 500 hours is

sufficient for a lifetime of 10 years

377

so manufacturers do not see the benefit of increasing

the motor lifetime currently. For cordless and robots, the lifetime of DC-motors are below

600 hours and it would in principle exclude all DC-motors. Also, it is difficult to perform

accurate tests reflecting real life use of cordless and robots. In addition, robots and cordless

vacuum cleaners are emerging technologies and are currently present in a wide price

range. This means that some product is more suited for light duty cleaning/spot cleaning

and are only used few hours a year. A motor lifetime requirement would make these

products consume more resources and increase the cost. Hence it is suggested to not

include a lifetime requirement for cordless and robots and instead give information about

the motor lifetime in the product fiche. However in the next revision of the regulation a

requirement regarding motor lifetime should be considered as consumer organisation are

in favour of such a requirement.

376

This requirement is suggested disregarding of the motor is tested partly loaded or empty.

377

With the current assumption on usage

292

The current requirement of 40,000 hose oscillation is recommended to be maintained and

also applied to cordless vacuum cleaners when a hose is present. For robot cleaners, a

hose is never expected to be present.

The important aspect of battery lifetime is suggested to be regulated with a minimum

requirement for cordless and robot vacuum cleaners. No standard for battery lifetime

exists, but the computer Ecodesign Regulation has an information requirement of battery

lifetime based on the number of charging cycles it can last. However, the battery capacity

falls over time with the number of charging cycles, and the share of power drawn from the

battery out of its total rated capacity (also called Depth of Discharge, DoD) is crucial for

the lifetime in terms of the capacity left after a number of cycles. It is therefore

recommended to set the requirement according to a definition including DoD and threshold

for remaining capacity, for example ‘after 600 charging cycles with 90% discharge in each

cycle, 70% of the battery capacity should remain’

378

. This means that cordless and robots

will need 2 batteries on average in their lifetime of 6 years, since they are used 200 times

a year. For robots and cordless vacuum cleaners, the battery is essential for a proper

lifetime and it is important that the battery is durable and can be exchanged for a fair

price. Otherwise consumers may replace their product instead of the battery. However,

stakeholders have also expressed concerns about to strict requirements for the battery

and capacity. A large battery will consume more resources, have an increased weight and

add significant cost to the product (also in the case of replacement). In addition,

stakeholders have expressed concerns whether it is possible for the market surveillance to

control such measures.

Besides durability requirements there are other possibilities to extend the lifetime of

vacuum cleaners. Based on a Deloitte study

379

it seems like the following options have a

positive effect on the environment:

• Measures to ensure provision of information to consumers on possibilities to repair

the product

• Measures to ensure provision of technical information to facilitate repair to

professionals

• Measures to enable an easier dismantling of products

• Measures to ensure availability of spare parts for at least a certain amount of years

from the time that production ceases of the specific models

• Different combination of the above-mentioned options

378

EN 61960:2011 could be used for measuring battery endurance in cycles (part 7.6.2 or 7.6.3 in the standard)

379

Deloitte (2016) Study on Socioeconomic impacts of increased reparability – Final Report. Prepared for the

European Commission, DG ENV

293

These measures are also applicable to the different types of vacuum cleaners and

information about repair and repair instructions for certain parts are suggested. According

to the study performed by Deloitte, the most beneficial measure is to ensure the availability

of spare parts. However, the essential/critical spare parts vary. According to preliminary

results from an ongoing study on the development of a scoring system for repair and

upgrade

380

, the most important aspects that define some parts as ‘priority parts’ are (listed

in order of importance):

1. Their frequency of failure

2. Their functional importance

3. The steps needed for their disassembly

4. Their economic value and related repair operations

5. Their environmental impacts

Parts which are likely to fail and are reasonable priced are permanent filters, hoses,

handles, accessories in general. These parts are all essential for the function of the machine

and are likely to be purchased if they break

381

. Many of the same parts are assumed to be

essential for cordless and robots. However, for cordless and robots the battery is also an

important spare part, and for robots the rotating brushes.

It is therefore recommended to set requirements on the availability of critical spare parts

throughout at least one lifetime of the product. This is 8 years for household mains-

operated, 5 years for commercial, and 6 years for cordless and robot vacuum cleaners.

However, stakeholders have explained that advancement in battery technology makes it

economical unfeasible to produce “older” batteries e.g. if an original cell is out of production

new UN and IEC approvals for replacement cells will cost a considerable amount of money.

This means there is a risk that the batteries become so expensive that it is unfeasible to

buy a battery. Therefore, a 4 years availability of batteries seems appropriate as the

consumers then can replace a battery after half a lifetime. The spare parts should be

available for the specified number of years after the last unit of a specific model is

produced. The repair prone parts should be possible for the user to change, without

needing help from a professional repair person, and should therefore be possible to conduct

without the need for special tools. Furthermore, an information requirement should be

implemented regarding information to the end-user on how to change these parts, as well

as how to best maintain the capacity of the battery in cordless and robot vacuum cleaners.

380

http://susproc.jrc.ec.europa.eu/ScoringSystemOnReparability/index.html

381

For upright vacuum cleaners, also the belts in the nozzle are expected to be changed if they break

294

In PO4 an information requirement on the amount of recycled plastic is suggested in order

to promote recycling of plastic and support the65% recycling goal from the WEEE Directive.

Where the WEEE Directive targets the End-of-Life aspects (collecting and recycling) the

Ecodesign Directive targets the design phase and thus the products placed on the market.

Since metals are already recycled at high rates, this requirement is based only on the

plastic, which has much lower recycling rates. In Figure 63 a conceptual drawing of the

recycling sign is presented.

Figure 84: Conceptual drawing of a recycling sign

The main barrier for such a requirement is how to ensure compliance, since it is not possible

to easily tell apart recycled and virgin materials, neither for metals nor plastics. One

solution is paper proof to have a trail of documentation for the material used and

declarations from suppliers about the material’s origin.

In order to calculate the effect of the two resource policy options, it was assumed that all

the requirements in PO4 results in 2 years additional lifetime (increasing lifetime from 8 to

10 years) for mains-operated household vacuum cleaners, corresponding to 25%.

Similarly, the lifetime of the other base cases is assumed to increase with 25%. Note that

the consumption of spare parts is expected to increase.

Material energy saving potentials

Based on the above requirements and the data presented throughout the study, the impact

of PO4 has been derived and compared to the BAU scenario. What is compared in this

section is the material energy, i.e. the energy consumed for production and embedded

energy of materials, not the energy consumed by the vacuum cleaners in the use phase.

As seen from Figure 85, the material energy in both scenarios is lower than in the BAU

scenario from 2022.

295

Figure 85: Material energy in PO4 compared to BAU from 2018 to 2030 in EU 28

As shown previously, the greenhouse gas emissions follow the energy consumption, which

is also the case for material energy and GHG emissions. The PO4 scenario has savings of

around 0.2 Mt CO

-eq/year by 2030 as seen in Figure 86.

Figure 86: GHG emissions in PO4 compared to BAU from 2018 to 2030

The savings in PO4 are caused by an assumed increase in the lifetime of vacuum cleaners

of 25% (from 8 to 10 years for mains-operated household vacuum cleaners), and an

increased use of recycled plastic. This means that more material (spare parts) are used

per vacuum cleaner and that the vacuum cleaners will miss out a potential energy

improvement according to the longer lifetime. However, the shift to recycled plastic

ensures savings from the first year the information on the label is introduced. The rapid

decrease in 2025 is due to the reduction in sales which is a result of the increased lifetime.

296

The material energy savings for each type of tumble driers in 2030 is presented in Table

109.

Table 109: Material energy savings for each base case in 2030 for PO4 and PO5 in EU28

2030 Material energy, TWh

2030 savings, TWh

2030 savings, %

BAU

PO4

Household mains-operated

4.74

3.11

1.64

35%

Commercial

1.17

0.75

0.42

36%

Cordless

5.73

4.26

1.47

26%

Robots

2.70

2.02

0.68

25%

Total

14.35

10.14

4.21

29%

The energy saving potential is 29% in PO4, which is also reflected in the end-user

expenditures, compared to the BAU scenario. Figure 87 shows the material end-user

expenditures for all vacuum cleaners in the EU. The cost is composed of the total sales,

the purchase price, increased costs for spare parts and the cost of loss in efficiency (i.e.

higher energy costs) when the lifetime is increased. The increased cost of spare parts is

responsible for increased costs compared to BAU until the sales are reduced after 2025.

Note that as even recycled plastic currently is cheaper it is assumed the content of recycled

plastic has no effect on the purchase price.

Figure 87: Material end-user expenditures for all vacuum cleaners in EU each year from 2018-

2030.

For PO4, consumer expenditure is lower than in the BAU as seen in Table 110. However,

the purchase price of vacuum cleaners might increase if spare parts are made available for

a longer period of time as more expenses can occur e.g. higher stocks of spare parts. Also,

if the demand for recycled plastic increases it may increase the price of recycled plastic.

297

Table 110: EU Material end-user expenditures for each base case

Consumer expenditure, Million €

Year

2030 Savings

Base case

Scenario

2018

2020

2030

Mains-operated household

BAU

5,405

4,482

3,635

PO4

5,405

4,859

3,580

Commercial

BAU

1,510

1,468

1,475

PO4

1,510

1,570

1,384

Cordless

BAU

1,893

3,244

4,258

PO4

1,893

3,296

3,511

18%

Robot

BAU

743

1,286

1,718

PO4

743

1,296

1,395

19%

13.6 Parameters on the energy label

In order to include cordless and robot vacuum cleaners as new product types in the scope

of a future Energy Labelling Regulation, it might very likely be necessary to also change

the label parameters and design. Especially for robots the label might need to look

different, since they cannot be compared directly to manually operated vacuum cleaners

due to differences in measurement methods and in performance parameters.

The parameters from the annulled energy label, are still relevant:

- Energy efficiency class

- Average annual energy consumption (kWh/year)

- Dust re-emission class

- Carpet cleaning performance class

- Hard floor cleaning performance class

- Sound power level

However, for cordless and robot vacuum cleaners, the battery run time per cycle is an

important parameter for end-users and could be added as a number (in minutes) on the

label for cordless products. For robots, a similar declaration could be made, however, some

stakeholders have suggested instead to show the area the robot can cover within a given

time, as a “covered area” per half hour measured in m

. Furthermore, it has been

suggested to add a “navigation class” to the robot label, based on the coverage rate, since

this is also an important parameter for end-users. In addition, it is suggested to add

information on the content of recycled plastic on the label for all vacuum cleaners.

In summary the following parameters are suggested for each of the vacuum cleaner types

in PO1, PO2 and PO4:

298

Table 111: parameters suggested for the energy label in PO1, PO2 and PO4

Mains operated

household

Commercial

Cordless

Robots

Annual energy scale

as main scale

Annual energy scale

as main scale*

Annual energy scale

as main scale

Annual energy scale

as main scale

Dust pick-up hard

floor

Dust pick-up hard

floor

Dust pick-up hard

floor

Dust pick-up hard

floor

Dust pick-up carpet

Dust re-emission

Noise

Coverage

factor/covered area

The content of

recycled plastic

The content of

recycled plastic

The content of

recycled plastic

The content of

recycled plastic

*Possibly exchange for a productive number, e.g. area/time, if a test is ready

13.7 Sensitivity analysis

The sensitivity analysis was performed for two parameters: market penetration (i.e. sales

and stock) of robots and cordless vacuum cleaners, and the assumed lifetime extension.

The impact on energy consumption of different sales numbers for robot and cordless is

calculated for policy options 1-3, since the sales have an impact on the total energy

consumption. The impact of a different lifetime extension as a consequence of the

requirements in PO4, is calculated only for the resource parameters, and thus only for PO4.

In order to calculate the sensitivity of the assumptions, the following scenarios were

modelled:

• Change in sales in PO1, PO2 and PO3

o Double/half the sales of robots in 2030. All other sales are stable.

o 25% increase/decrease in the sales of cordless in 2030. All other sales are stable.

• Expected lifetime extension by spare parts availability in PO4

o 10%-point increase/decrease of the expected 25% increase in lifetime i.e. 35%

increase in lifetime and 15% increase in lifetime.

The impact of these changes is calculated as the difference in TWh (use phase or material

energy). In Table 112 and Table 113 the impact of changing the sales of robot and cordless

vacuum cleaners is presented. The tables include the original scenarios, a scenario which

double sales by 2030 and a scenario with half the sales by 2030.

Table 112: Change in robot vacuum cleaner sales and the effect in BAU, PO1, PO2 and PO3

2020

2025

2030

2020

2025

2030

BAU

Savings, TWh

Change from original, %

BAU (original)

16.205

16.074

17.399

299

With double robot sales

16.357

16.684

18.548

0.9%

3.8%

6.6%

With half robot sales

16.129

15.769

16.824

-0.5%

-1.9%

-3.3%

PO1

Savings, TWh

Change from original, %

PO1 (original)

0.506

2.926

4.502

With double robot sales

0.527

3.178

5.053

4.0%

8.6%

12.2%

With half robot sales

0.496

2.799

4.227

-2.0%

-4.3%

-6.1%

PO2

Savings, TWh

Change from original, %

PO2 (original)

0.521

2.790

4.350

With double robot sales

0.541

3.043

4.900

3.9%

9.0%

12.7%

With half robot sales

0.511

2.664

4.075

-2.0%

-4.5%

-6.3%

PO3

Savings, TWh

Change from original, %

PO3 (original)

-0.370

-1.337

-1.832

With double robot sales

-0.370

-1.332

-1.818

0.0%

-0.3%

-0.7%

With half robot sales

-0.370

-1.339

-1.838

0.0%

0.2%

0.4%

Table 113: Change in cordless vacuum cleaner sales and the effect in BAU, PO1, PO2 and PO3

2010

2015

2020

2025

2030

BAU

Savings, TWh

Change from 100% robots, %

BAU (original)

16.205

16.074

17.399

With double cordless sales

16.366

16.793

18.772

1.0%

4.5%

7.9%

With half cordless sales

16.044

15.356

16.026

-1.0%

-4.5%

-7.9%

PO1

Savings, TWh

Change from 100% robots, %

PO1 (original)

0.506

2.926

4.502

With double cordless sales

0.520

3.120

4.953

2.7%

6.7%

10.0%

With half cordless sales

0.492

2.731

4.052

-2.7%

-6.7%

-10.0%

PO2

Savings, TWh

Change from 100% robots, %

PO2 (original)

0.521

2.790

4.350

With double cordless sales

0.535

2.985

4.800

2.7%

7.0%

10.4%

With half cordless sales

0.507

2.596

3.899

-2.7%

-7.0%

-10.4%

PO3

Savings, TWh

Change from 100% robots, %

PO3 (original)

-0.370

-1.337

-1.832

With double cordless sales

-0.370

-1.352

-1.899

-0.2%

1.2%

3.7%

With half cordless sales

-0.371

-1.321

-1.764

0.2%

-1.2%

-3.7%

In Table 112 and Table 113 it is seen that the sales of robots and cordless vacuum cleaners

has an impact on the results. In general, if more vacuum cleaners are sold (increase in the

penetration rate) the impact of vacuum cleaners on energy consumption increases, as well

as the potential savings in PO1 and PO2. This means that the relative change with an

increase/decrease in sales are small e.g. if the sales of robots are doubled towards 2030

the overall energy consumption will increase by 6.6% in the BAU scenario, but the savings

in PO1 will increase by 12.2%. Meaning that the energy consumption in PO1 (with current

assumption) is 12.897 TWH in 2030. With increased sales (double) of robots the resulting

300

energy consumption in PO1 is 13.495 TWH, or a change of 4.6% in energy consumption.

This means that even with a relatively high change in the sales of robots and cordless, the

overall result is still valid.

In Table 114 the impact of the expected increase in lifetime is calculated. Note that BAU is

the current assumption on lifetime, PO4 is the current assumed increase in lifetime in PO4

(25%), PO4 +10% is an expected increase in lifetime of 35% and PO4 -10% is an expected

increase in lifetime of 15%.

Table 114: Change in the expected increase in lifetime in policy option 4

2030 Material energy, TWh

Savings, TWh

Savings, %

BAU

PO4

+10%

PO4

-10%

PO4

+10%

PO4 -

10%

PO4

+10%

PO4 -

10%

Household

mains-

operated

4.74

3.11

2.72

3.49

1.64

2.03

1.25

35%

43%

26%

Commercia

1.17

0.75

0.66

0.85

0.42

0.51

0.32

36%

44%

27%

Cordless

5.73

4.26

3.72

4.79

1.47

2.01

0.94

26%

35%

16%

Robots

2.70

2.02

1.77

2.28

0.68

0.93

0.43

25%

34%

16%

Total

14.3

10.14

8.87

11.4

4.21

5.47

2.94

29%

38%

20%

The change in lifetime is difficult to predict, but even an increase of 15% in the lifetime

will cause a reduction of 20% in the material energy consumption (including the increase

in the content of recycled plastic). If the increase is 35% the savings is almost 40% of the

material energy (5.5 TWh in 2030).

13.8 Conclusions and recommendations

Based on the data and analyses presented in this report, it is recommended to include

cordless and robot vacuum cleaner in scope, but with different requirements than mains

operated vacuum cleaners.

If it is technically possible and feasible to develop a reproductive and repeatable test

methods, that either measure or simulate the performance of vacuum cleaners with part

load, it is recommended to follow policy option 2. This entails maintaining the same AE and

rated power requirements, but implementing a new energy label regulation.

301

Policy option 1 with stricter ecodesign requirements is not suggested, as this would leave

no room for a full label scale (i.e. 7 classes). If the number of classes were reduced, it

might be possible, however, as showed in the scenario analysis this will give little savings

in addition to PO2.

If a part load test is not technically possible or feasible, on the other hand, it is

recommended to set stricter ecodesign requirements for AE and rated power, according to

PO3. This will ensure that at least some of the potential savings are obtained, even though

it will be around only half the savings than with an energy label.

In addition, it is recommended to include in policy option 4 resource requirements and

information requirements on the content of recycled plastic on the label. Policy option 4

can be applied regardless of the choice of other policy options in connection with durability,

reparability and availability of spare parts requirements. The information on the content

on recycled plastic can only be added to PO1 and PO2. Without a label, the potential savings

will be reduced.

If requirements PO4 is applied, it is suggested to adopt the formulation on resource

requirements from, e.g. the refrigerating appliances/washing machines to ensure

coherence across the different product groups regarding resource efficiency.

Other specific recommendations include:

- Remove the definition of “full size battery operated vacuum cleaner” and instead

use the definitions of robot and cordless vacuum cleaners. Have only one category

for cordless vacuum cleaners without any sub-division. Leave handheld vacuum

cleaners (not for floor cleaning) out of the scope.

- Include cordless and robot cleaners in scope of both the regulations, but:

o Consider the timing of when they should be included from, which might not

be the same for both product types, and might to some extent depend on

finalisation of the test standards

o Analyse in more detail, preferably with additional data, which requirements

are appropriate and consider implementing them in two tiers to give the

market and manufacturers time to adapt

o At the very least make sure that maintenance mode requirements are set

within a relatively short time frame

- Use the EI calculation instead of the AE calculation for commercial vacuum cleaners,

and make a separate label design.

- Rescale the label to an A-G scale and rescale the performance parameters scales to

only four classes (A-D).

302

o The timing of re-introducing the labelling regulation and including more

nuanced performance standards is important. Changing the standards might

influence the limit value of Ecodesign requirements and the energy label

scales.

o Make specific label designs for mains operated household, cordless, robot

and commercial vacuum cleaners.

- Set the verification tolerances according to the measured expanded uncertainties

when final results are available, and in accordance with the new label scales

303

14. Annexes

I. Annex A – Elaboration of standards

Elaboration of standards under request M/540

The responsible WG dealing with Mandate M/540 is WG 6 “Surface cleaning appliances”

that operates under CENELEC TC 59X, the broad CENELEC TC that is responsible for

standards regarding “Performance of household and similar electrical appliances”. WG 6,

Surface cleaning appliances, has subdivided specific parts of the mandate into several sub-

working groups as shown in Table 115.

Table 115: CENELEC TC 59X WG 6 sub-working groups

Sub-working group

Specific part

WG 06-01

Water filter vacuum cleaners

WG 06-02

Uncertainties for vacuum cleaners

WG 06-03

Commercial surface cleaning appliances

WG 06-04

Durability of suction hoses

CENELEC TC 59X WG 6 Surface cleaning appliances cooperates very closely with their

counterparts on IEC level within IEC SC 59F (see Table 116). IEC WGs agreed to address

considerable content of the Standardisation Request (M/540) because it is relevant

worldwide. Examples: full-size battery operated vacuum cleaners, robot vacuum cleaners

etc. Also other relevant issues are handled in the respective IEC WGs - e.g. Wilton carpet

test (in IEC SC 59F WG 9). Experts are mostly the same in both CENELEC and IEC WGs.

Meetings are held in combination or jointly as far as possible.

Table 116: IEC TC 59 SC 59F Working groups and advisory groups

Working group

Title

IEC TC 59 SC 59F/ WG2

Acoustical noise of household appliances

IEC TC 59 SC 59F/ WG3

Dry surface cleaning appliances

IEC TC 59 SC 59F JWG4

Wet surface cleaning appliances linked to ASTM-INTERNATIONAL

IEC TC 59 SC 59F/ WG5

Surface cleaning robots

IEC TC 59 SC 59F/ WG6

Commercial surface cleaning machines

IEC TC 59 SC 59F/ WG7

Cordless (battery operated) vacuum cleaners

IEC TC 59 SC 59F/ WG9

Test equipment and test material

IEC TC 59 SC 59F/ AG1

CAG Chairman's Advisor Group

IEC TC 59 SC 59F/ AG2

Hard floor cleaning

IEC TC 59 SC 59F/ AG3

Advisory group on airborne noise from surface cleaner

304

1. Durability of the hose and operational lifetime of the motor

Durability testing of the hose and operational motor lifetime are part of the new EN 60312-

1:2017 standard which was handled through a Unique Acceptance Procedure (UAP)

382

and

has recently been harmonised .

The efforts of CLC TC59X WG 6 to produce a harmonised standard implementing the

durability requirements was closely linked to the special review study on vacuum cleaners

of the European commission prepared by VHK

384

. This special review study followed Article

7(2) of Commission Regulation (EU) No 666/2013 on Ecodesign requirements for vacuum

cleaners

385

, which specified that the durability requirements on hose (at least 40 000

oscillations) and motors (at least 500 hours at half-loaded receptacle) had to be reviewed.

The study started in December 2015 and the final study report was published in June 2016.

Durability of the hose

The current test set-up and test-procedure in Clause 6.9, ‘Repeated bending of hose’ in

the harmonised standard EN 60312-1:2017 has been used for many years by industry and

consumer associations and is in principle unproblematic. For the durability test of the hoses

the problem lay with the definition of the hoses: Which hoses (primary, secondary) of

which types of vacuum cleaners (cylinder, upright) will need to be subject to the test.

Both upright and cylinder vacuum cleaners are, for the purpose of the current Regulation,

dry vacuum cleaners. Section 6.9 of the harmonised standard EN 60312-1:2017 defines

primary and secondary as follows:

“This test is only applicable to hoses that constitute the primary structural link between

the floor-supported main body of a cylinder vacuum cleaner and a separate cleaning head

or cleaning head/tube assembly that, in normal use, is used to clean a floor from an upright

standing position (see Figure Z.1).

This test is not applicable to hoses that, in normal use, remain affixed at both ends to a

vacuum cleaner with a cleaning head that, in normal use, forms an integral part of, or is

permanently connected to, the vacuum cleaner housing. This configuration can often be

found on upright vacuum cleaners (see Figure Z.2)

386

. Such hoses may be released at one

end to allow other cleaning tasks to be carried out (see Figure Z.3)

387

382

The Unique Acceptance Procedure (UAP) is a procedure which may be applied to an EN standard, in order to achieve rapid

approval. The UAP combines the 2 voting stages (Enquiry and Formal) and does not allow technical comments. The duration of

a UAP is approximately 1 year.

383

OJ publication C 267/4, 11-08-2017

384

http://ia-vc-art7.eu/

385

Commission Regulation (EU) No 666/2013 of 8 July 2013 implementing Directive 2009/125/EC of the European Parliament

and of the Council with regard to Ecodesign requirements for vacuum cleaners, OJ L 192, 13.7.2013, p. 24–34

386

A test regarding durability of such hoses is under development. https://www.techstreet.com/standards/bs-en-60312-1-

2017?product_id=1950146

387

Section 6.9.1 of the harmonised standard EN 60312-1:2017

305

Figure Z.1 – Typical cylinder vacuum cleaner

with primary hose

Figure Z.2 – Typical upright vacuum cleaner

with secondary hose (contour front and

back)

Figure Z.3 – Typical upright vacuum cleaner with secondary hose used for cleaning curtains

(left) and stairs (right).

The test is not applicable for:

• Hoses that are permanently housed within other components of a vacuum cleaner,

or that cannot be removed from a vacuum cleaner without the use of tools;

• Hoses that join two or more components, where, in all usage modes, the structural

link between those components is provided by features other than the hose itself

(an example is shown in Figure Z4);

• Hoses that are provided as additional accessories or where another primary hose is

provided for general use.

306

Figure Z4 – Example of a hose joining two or more components

Durability of the motor

Clause 5.9, Performance with loaded dust receptacle, is excluded

388

from the harmonised

standard EN 60312-1:2017. This part of the standard explains how to load the receptacle

which is needed for the operational motor-life test:

6.Z3.1 Purpose

The purpose of this test is to determine the stationary operational life-time of a dry vacuum

cleaner suction and agitation device motor.

6.Z3.2 Test method

The dry vacuum cleaner, equipped as in its normal operation with hose and tube (if

applicable) and nozzle, shall be operated as stated in 4.6. It is allowed to run intermittently

with periods of 14 min 30s on and 30 s off in maximum power setting.

This test is operated with a half loaded receptacle; hence the dust receptacle shall be

loaded with 50 % of the amount of test dust required according to 5.9. Alternatively, an

empty dust receptacle can be used during the test. In this case the recommended testing

time shall be increased by 10 % of the stated motor life value for testing with a half loaded

dust receptacle.

The tube grip of dry vacuum cleaners with suction hose or the handle of other dry vacuum

cleaners shall be held as for normal operation at a height of 800 mm  50 mm above the

test floor. The nozzle shall not be in contact with the floor, but lifted 1 cm off the floor.

If the dry vacuum cleaner is provided with an agitation device, it shall be running. If

manufacturer’s instructions require different settings of the agitation device for use on

carpets and use on hard floor, the agitation device shall be operated with the respective

settings for 50% each of the total testing time.

Test with half loaded dust receptacle: After 50 h



5h of operation, the vacuum cleaner shall

be equipped with a clean dust receptacle and new filters (see 4.5). This procedure, with

388

Clauses 5.9, 6.15, 6.Z1.2.3, 6.Z1.2.4, 6.Z1.2.5, 6.Z2.3 and 6.Z3 are not part of the present citation. In clause 7.2.2.5 read

‘A2 fine test dust’ instead of ‘test dust’.

307

the receptacle loaded with the same amount of test dust as for the first cycle, shall be

repeated in steps of 50 h



5h.

Test with empty dust receptacle: After 100 h



5 h of operation, the vacuum cleaner shall

be equipped with a clean dust receptacle and new filters (see 4.5).

Changing or maintenance of dust receptacles and filters shall be carried out in accordance

to the manufacturer's instructions and this shall be recorded, see 4.5. End of life is reached

when the suction motor and, if applicable, the agitation device stops operating or the

recommended testing time has elapsed.

NOTE The 30 second off period is not included in the calculation of overall motor life.”

2. Water filter vacuum cleaners

As water filter vacuum cleaners were not addressed in existing standards, this aspect has

been added to the harmonised standard EN 60312-1:2017. All tests were checked and

amended where necessary in order to make them applicable for water filter vacuum

cleaners. The following definitions have been added to the 2017 version:

Water filter vacuum cleaner: Dry vacuum cleaner that uses water as the main filter

medium, whereby the suction air is forced through the water entrapping the removed dry

material as it passes through.

Water filter system: removable water filter components which are in contact with the water”

3. Full size battery operated vacuum cleaners

Work on this part of the mandate is mostly done by IEC SC 59F WG 7. The new draft

standard “IEC 62885-4 Surface cleaning appliances – Part 4: Cordless dry vacuum cleaners

for household or similar use – Methods for measuring the performance” focusses on battery

operated vacuum cleaners to be used on the floor by the user from an erect standing

position and is based on the EN 60312-1 for dry vacuum cleaners. The new draft standard

IEC 62885-4 is currently at CD stage. It is subject to parallel voting on CENELEC level.

All tests were checked and amended where necessary for battery operated vacuum

cleaners. This includes specific measurement methods for the energy consumption of the

batteries. Another parameter which is considered to be highly relevant for battery operated

vacuum cleaners is “run time”. This is the duration such an appliance can be used by

customers while a reasonable suction power is provided. A new test was elaborated which

is included in the draft standard.

Handheld battery operated vacuum cleaners for above-the-floor cleaning are left for a

future edition.

4. Robot vacuum cleaners

Robot vacuum cleaner standards are developed on a worldwide level by IEC SC 59F WG 5

and in cooperation with CENELEC TC 59X WG 6 the potential Energy labelling and Ecodesign

requirements will be addressed in a new standard “IEC 62885-7 Surface cleaning appliance

– Part 7: Dry-cleaning cleaning robots for household use – Methods of measuring

performance”. The new standard amends the existing test standard IEC (EN) 62929:2014

- Cleaning robots for household use. Dry cleaning: Methods of measuring performance.

308

IEC 62929:2014 is applicable to dry cleaning robots for household use in or under

conditions similar to those in households. The purpose of this standard is to specify the

essential performance characteristics of dry cleaning robots and to describe methods for

measuring these characteristics. This standard is neither concerned with safety nor with

performance requirements.

IEC 62929 contains measurement of:

• Dust removal from hard flat floors and from carpets - box test

• Dust removal from hard flat floors and from carpets - straight line test

• Autonomous navigation/coverage test

• Average robot speed

The following additional tests are planned for the next voting stage

389

of the new draft

standard IEC 62885-7:

• Obstacle overcome capability

• Energy consumption

• Debris pick-up – box and or straight line

• Fibre pick-up – box

The overall conclusion by the Standardisation work group (TC 59X WG 6) is that the

standards are not mature enough to be used for Energy Labelling / Ecodesign purposes.

The main reasons are:

• There is limited experience with tests because standard is new (published in 2014)

or under development

• There is no data for repeatability available; RRT has yet to be concluded

• Still considerable change on the market

The forecasted publication date is July 2020

390

5. Measurement with market-representative carpet(s) and hard

floor(s)

This part of the mandate is executed in close collaboration with CEN TC 134, Resilient,

textile and laminate floor coverings. TC 134 presented figures of EU market shares for floor

coverings. As can be seen in Figure 88, carpets cover about 24% of the total floor area,

hard floors about 30% (laminate and parquet), resilient floors about 17% and ceramics

about 29%.

389

CDV is Committee Draft for Vote, similar to the Enquiry vote within CENELEC and is estimated to take place June 2018

http://www.iec.ch/dyn/www/f?p=103:23:0::::FSP_ORG_ID:1395

390

http://www.iec.ch/dyn/www/f?p=103:23:0::::FSP_ORG_ID:1395

309

Figure 88: Total EU market for floor coverings in 2015, equalling 1900 million m2 and 15% of

global market

391

Carpets

The analysis also showed that domestic cut pile and domestic loop pile, shown in Figure

89, will cover around 90% of the domestic carpet share in the EU. Therefore, two cut pile

and two loop pile carpets are chosen and will be distributed to the test labs

392

Figure 89: left: domestic loop pole, right: domestic cut pile

393

Resilient floors

For resilient floors two samples are also proposed; the Cushion Vinyl (embossed) and

Luxury Vinyl Tiles (LVT) planks. The pictures shown in Figure 90 are examples of Cushion

Vinyl and Luxury Vinyl Tiles planks and might differ from the samples distributed to the

test labs and are merely shown as illustration.

391

EU market share floor coverings (Source: presentation CEN TC 134 at the February 2017 meeting in Hartmannsdorf)

392

Status as of November 2017

393

Picture source: presentation CEN TC 134 at the February 2017 meeting in Hartmannsdorf

29%

24%

18%

15%

10%

Ceramic Carpet Laminates Vinyl Wood Others

310

Figure 90: left: Allura Vinyl Tile

394

, right: Viva Cushion vinyl

395

Laminate

For laminate floors also two kinds are proposed the Quick Step Impressive

396

and Colours

Gawler

397

Parquet

One kind of parquet is proposed the Maxistab

398

Testing

During the initial tests 7 different floor coverings were distributed to 13 laboratories. The

proposed selection of flooring includes: 2x synthetic carpets “The Noble Collection - Saxony

180" (cut pile) and Gala 13 (loop pile) and 5x types of hard floors, thereof: 2x Laminate

(impressive; Colour Gawler), 2x Vinyls (Novilon, Allura) and 1x Parquet (Maxistab).

Findings regarding carpets were presented during the standardisation meeting in March

2019 in Brussels, based on 6 laboratories:

- Saxony 180 is not suitable to become a test carpet (preliminary results showed longer

conditioning process to get a stable result, low dpu level with high fluctuation, structure

changed after few runs),

- Gala 13 could be qualified to become a test carpet (preliminary results showed higher

level dpu with sufficient stability, lower motion resistance, no change in ranking (depending

on nozzle)).

The hard floors findings showed no differences on all hard floors for dpu with debris and

very little to no differences on all hard floors for dpu with fine dust

A report on the results regarding tests on market representative floors will be presented

to the Commission as part of M/540.

394

https://www.forbo.com/flooring/nl-nl/producten/luxe-vinyltegels-en-stroken/allura/ba9wax

395

https://www.forbo.com/flooring/en-uk/products/for-your-home/novilon-cushion-vinyl/novilon-viva/bqsjty#7400

396

https://www.quick-step.be/nl-be/campagnes/impressive-laminaatvloeren

397

http://www.classen.de/en/laminate-flooring

398

https://www.meisterwerke.com/de/declaration-of-performance/markenauswahl/schulte-raeume/

311

6. Consumer organization tests

Which?

Which? is an independent consumer organization based in the UK. Every year they test

over 3600 products and cover the essential features of a product. Tests performed on

cylindrical and upright vacuum cleaners are

399

• Cleaning of fine dust and dirt: For the carpet test a machine spreads super-fine dust

over a carpet and grinds it in. The vacuum cleaner is then placed onto a test rig,

which pulls and pushes it back and forth five times as it sucks up the dust. This test

is repeated for smooth and creviced wood floors.

• Cleaning of debris: Vacuum cleaners are also challenged to pick up larger debris.

For this Which? used a large amount of dry rice.

• Dust re-emission: To test if vacuum cleaners keep fine dust safely locked away

inside, specialist machinery is used to test how much dust and fine particles the

vacuum cleaners retain.

• Suction power while the bag or canister fills up: The vacuum cleaner is put on the

test rig again, and measuring takes place on the suction power when bags or

canisters are empty, and again when they are filled with dust and debris.

• The time it takes to pick up pet hair and longer hair: Real cat and dog fur are

combed into an area of carpet and the time is measured how long it takes to pick

all of the hair up. This test is repeated for longer hair of real human and it is tested

how long it takes to remove the fluff from a cushion with the provided upholstery

tool.

• Manoeuvrability: A panel of experts was asked to assess the manoeuvrability of the

vacuum cleaner in common scenarios, from vacuuming up and down stairs to

moving it across different and uneven surfaces. They also check how easy it is to

change and use the attachments and to empty the bag or canister.

• Noise: The sound of each vacuum cleaner is tested in a lab

Certain assessments are more important than others and so Which? carried out different

weights to categories the vacuum cleaners: 75% cleaning and filtration 20% ease of use

5% noise and energy use.

Which? tested also robot vacuum cleaners

400

and the focus was on:

• Dust and dirt removal: Super fine Arizona sand is spread over thick Wilton carpet,

and chunky lentils are spread over a hard floor to test how effectively each robot

can pick up mess from different surfaces. After the robot vacuum cleaner returns

to its charging station the amount of dust/dirt pick up is measured.

Similar to the cord vacuum cleaners test above real cat and dog fur are combed

into an area of carpet and the amount of hair picked up is measured (not the time).

• Floor coverage: A specially designed room complete with tables, chairs, lamps, rugs

and low hanging curtains is built to see how well each robot gets on navigating

around a typical room. Cameras were installed in the room and sensors were

399

http://www.which.co.uk/reviews/vacuum-cleaners/article/how-we-test-vacuum-cleaners

400

http://www.which.co.uk/reviews/robot-vacuum-cleaners/article/how-we-test-robot-vacuum-cleaners

312

attached (in three places) on each robot so that what spots the robot covers and

which areas it fails to reach can be monitored.

• Navigation round obstacles: The maximum height of a ridge is tested so that each

robot vacuum cleaner can climb over. Furthermore, a wide array of everyday

obstacles is put in the path of each robot to see how they handle this. The test room

has a tangle of wires, tables and chairs, a domed floor lamp and fold out chairs to

try and trip up each robot cleaner.

• User friendliness: The out of the box setting is tested furthermore, it is tested how

easy it is to programme and schedule a cleaning cycle and also how easy or difficult

it is to do regular maintenance on your robot, such as emptying the dust container

and cleaning any filters.

Which?’s overall ratings for robot vacuums ignore price and are based on: Cleaning - 52%

Navigation and obstacle avoidance - 28% Ease of use - 18% Noise - 2%

Cordless vacuum cleaners

401

are tested focusing on:

• Dust removal and re-emission: Fine Arizona sand is used to see how much dust

each vacuum cleaner picks up, as well as how much is re-emitted. The test

continues till the vacuum cleaners’ battery is only 20% charged. Also here, the

ability to pick up pet fur is tested, both the amount as the time it takes to pick it up

is measured.

The suction of each cordless model is tested on three different surfaces - laminate,

floorboards and carpet. 25g of dust is used and a test comprises of two runs on

each surface type, the amount dust in the canister is measured at the end.

• Battery lifetime: The time is measured how long it takes to fully charge and run

completely empty. This test is performed on the most powerful setting (not standard

as most manufacturers use). To further test the battery also the pick-up capabilities

is tested when only 20% of its charge remains.

• Noise: The sound of each vacuum cleaner is tested

Overall ratings are based for 75% on suction, filtration and battery for 20% on ease of use

and 5% noise.

Stiftung Warentest

Stiftung Warentest is an independent German consumer organization who tests products

and services according to scientific methods in independent institutes and publishes the

results in their publications. The Stiftung Warentest tested corded vacuum cleaners,

battery and robot vacuum cleaners.

Corded vacuum cleaners are tested according to the following features:

• Dust absorption test: The standardized dust intake is measured in accordance with

the EN 60312-1. For the test of Duracord carpet, smooth hard floors and crevices

the receptacles are filled with 200 grams of test dust or when this is not possible

the vacuum cleaners are tested with a negative pressure of 40 percent of the initial

value. Also the fibre uptake is measured.

401

http://www.which.co.uk/reviews/cordless-vacuum-cleaners/article/how-we-test-cordless-vacuums

313

• Handling: Five persons (testers) make an everyday test and they assess the

operating instructions, set up and dismantling of the devices, as well as handles,

switches, displays and storage - additionally and the carrying of the devices. Further

test points: How well can carpet and hard floors, stairs and upholstery be cleaned,

cleaning of the nozzles, changing of the dust bag and filter or empty the receptacle.

• Dust retention capacity: the fine dust content in the inlet and exhaust air is

compared as the degree of separation. The more dust remains in the filter, the

higher the separation efficiency, the better.

• Noise: sound power level is tested according EN 60704-2-1.

• Power Consumption: during the dust absorption test described above the electricity

consumption of the vacuum cleaner is measured (the result refers to 10m

• Durability test: The lifetime of the motor is tested by letting the vacuum cleaner

run up to 600 hours; and up to 95 hours for cordless hand-held vacuum cleaners

with assessment of the battery time reduction.

Impact test are performed so will a vacuum cleaner hit 1.000 times a post and go

10.000 over sleepers. The nozzle must exceed 1.200 falls from a height of 80cm

and the cable extraction must succeed 6.000 pulls. Furthermore the hose fittings

are pivoted for 40000 and the pipes and hoses are squeezed with a load of 70 kg

for 10 seconds.

• Safety: In accordance with EN 60335-1 and -2-2, the electrical safety of the vacuum

cleaners is checked.

Battery operated vacuum cleaners are tested the same way as the corded vacuum cleaners

the only differences are that with the dust absorption test 25 grams and 50 grams are fed

to the vacuum cleaner. The battery recharge times are evaluated, and the vacuum cleaners

will undergo 67.500 cycles on the crank test instead of the threshold test. The following

features have been tested for robot vacuum cleaners:

• Dust absorption test: The tests were conducted in accordance with EN 62929 on

carpet and hard floors.

• Navigation: The navigation test is carried out in a test room in accordance with EN

62929, the inventory was slightly modified (compared to the dust absorption test

room) and an additional outdoor area of approx. 2 m² was created before the

entrance door.

• Handling: Five experts evaluated the instruction manual and tested benefits of

the cleaner controls/displays, the ease of emptying the dust box, cleaning of

filters and unit and remote-control capability, defined space and carrying the

device.

• Environmental characteristics: Sound power was tested according to EN 60704-2-

1 on carpet and hard floors

Dust re-emission was tested according to EN 60312-1 and the annual power

consumption for daily cleaning of the test room (about 20 square meters) was

calculated, including running and charging times, maintenance charging of the

battery and maintenance mode power consumption of the charger.

314

• Durability: The vacuum cleaners ran non-stop in a test room with short pile carpet

for 16 weeks

402

. They ran until the battery had to be charged. After recharging, they

continued cleaning again.

Consumentenbond

The Dutch independent consumer organization tested cylinder vacuum cleaners

403

. The

features that were tested are:

• Cleaning performance: Tests are performed on carpets and hard floors including

crevices. To test the cleaning of pet fur/hair synthetic fibres are used to mimic real

pet hair. Furthermore, the suction power is measured when the receptacle fills up.

• Durability test: Motor lifetime is tested according to EN 60312-1 chapter 6.10.

The mechanism to roll up the cable is tested by unwinding it 1.000 times and let it

roll up again.

• Dust re-emission

• Energy consumption: The energy consumption is measured while vacuuming 10m

of carpet and hard floors.

• Noise

402

Stifftung Warentest, 2/2017, page 63 it is written in the Haltbarkeit section “16 Wochen”

403

https://www.consumentenbond.nl/stofzuiger/hoe-wij-testen

315

II. Annex B – GfK data coverage

Data coverage of the data purchased from GfK.

Country

Coverage

Population

GDP (bill. EUR)

Austria

90%

8 690 076

349.5

Belgium

88%

11 311 117

421.6

Czech Republic

89%

10 538 275

163.9

Germany

74%

82 175 684

3 134.0

Denmark

83%

5 659 715

266.2

Spain

83%

46 445 828

1 114.0

Finland

82%

5 487 308

214.1

France

90%

66 759 950

2 225.0

Great Britain

95%

65 382 556

2 367.0

Greece

95%

10 783 748

175.9

Croatia

75%

4 190 669

45.8

Hungary

94%

9 830 485

112.4

Ireland

90%

4 724 720

265.8

Italy

89%

60 665 551

1 672.0

Luxembourg

70%

576 249

54.2

Netherland

81%

16 979 120

697.2

Poland

93%

37 967 209

424.3

Portugal

94%

10 341 330

184.9

Romania

90%

19 760 314

169.6

Sweden

85%

9 851 017

462.4

Slovenia

85%

2 064 188

39.8

Slovakia

89%

5 426 252

81.0

Bulgaria

7 153 784

47.4

Cyprus

848 319

17.9

Latvia

1 968 957

25.0

Lithuania

2 888 558

38.6

Estonia

1 315 944

20.9

Malta

434 403

9.9

Total

510 221 326

14 800

Total coverage

430 709 693

12 580

84%

85%

316

III. Annex C - Sales and stock data

Vacuum cleaner sales in each category, 1995 to 2030, million units.

Year

Cylinder

househol

Cylinder

commercial

Upright

Househol

Upright

Commercial

Handstick

Mains

Handstick

cordless

Robot

Total

1995

14.81

1.78

2.61

0.31

0.30

0.51

20.32

1996

14.81

1.78

2.61

0.31

0.30

0.51

20.32

1997

14.81

1.78

2.61

0.31

0.30

0.51

20.32

1998

14.81

1.78

2.61

0.31

0.30

0.51

20.32

1999

14.81

1.78

2.61

0.31

0.30

0.51

20.32

2000

14.81

1.78

2.61

0.31

0.30

0.51

20.32

2001

15.82

1.90

2.79

0.34

0.32

0.55

21.71

2002

13.71

1.64

2.42

0.29

0.28

0.48

18.81

2003

15.88

1.91

2.80

0.34

0.32

0.55

21.80

2004

15.95

1.91

2.82

0.34

0.32

0.55

21.89

2005

16.92

2.03

2.99

0.36

0.34

0.59

23.22

2006

19.02

2.28

3.36

0.40

0.38

0.66

26.10

2007

23.52

2.82

4.15

0.50

0.47

0.82

32.28

2008

25.16

3.02

4.44

0.53

0.51

0.87

34.53

2009

25.09

3.01

4.43

0.53

0.50

0.87

34.43

2010

25.01

3.00

4.41

0.53

0.50

0.87

34.33

2011

24.80

2.98

4.18

0.50

0.57

0.99

0.15

34.18

2012

25.96

3.12

4.17

0.50

0.68

1.18

0.32

35.92

2013

25.82

3.10

3.94

0.47

0.76

1.31

0.48

35.88

2014

25.17

3.02

3.64

0.44

0.82

1.42

0.63

35.13

2015

25.28

3.03

3.44

0.41

0.91

1.56

0.79

35.43

2016

25.73

3.09

3.29

0.39

1.00

1.74

0.97

36.22

2017

25.47

3.06

3.04

0.37

1.08

1.86

1.13

36.01

2018

25.90

3.11

2.87

0.34

1.18

2.04

1.32

36.78

2019

26.30

3.16

3.02

0.36

1.17

2.57

1.34

37.92

2020

25.07

3.01

2.91

0.35

1.25

4.24

1.45

38.28

2021

24.58

2.95

2.62

0.31

1.46

5.72

1.57

39.22

2022

24.03

2.95

2.61

0.31

1.56

6.55

1.78

39.80

2023

23.43

2.95

2.60

0.31

1.66

7.39

2.00

40.35

2024

22.77

2.95

2.58

0.31

1.77

8.25

2.22

40.85

2025

22.06

2.95

2.56

0.31

1.87

9.11

2.45

41.32

2026

21.31

2.95

2.53

0.31

1.98

9.99

2.67

41.74

2027

20.51

2.95

2.50

0.31

2.08

10.87

2.90

42.12

2028

19.67

2.95

2.46

0.31

2.18

11.75

3.12

42.46

2029

18.79

2.95

2.43

0.31

2.28

12.63

3.35

42.75

2030

17.88

2.95

2.38

0.31

2.38

13.51

3.58

43.00

317

Calculated stock of each vacuum cleaner category, 1995 to 2030, million units.

Year

Cylinder

househol

Cylinder

commercial

Upright

Household

Upright

Commercial

Handstick

Mains

cordless

Robot

Total

1995

14.8

1.8

2.6

0.3

0.5

20.3

1996

29.6

3.5

5.2

0.6

1.0

40.6

1997

44.4

5.2

7.8

0.9

1.5

60.7

1998

59.1

6.7

10.4

1.2

1.9

80.5

1999

73.6

7.9

13.0

1.4

1.5

2.3

99.6

2000

87.4

8.8

15.4

1.6

1.8

2.6

117.5

2001

100.9

9.5

17.8

1.7

2.0

2.9

134.7

2002

110.0

9.6

19.4

1.7

2.2

3.0

146.0

2003

118.5

9.8

20.9

1.7

2.4

3.2

156.6

2004

124.2

10.0

21.9

1.8

2.5

3.3

163.7

2005

128.7

10.3

22.7

1.8

2.6

3.5

169.5

2006

133.8

10.7

23.6

1.9

2.7

3.6

176.4

2007

142.8

11.7

25.2

2.1

2.9

3.9

188.6

2008

153.1

12.9

27.0

2.3

3.1

4.2

202.6

2009

163.2

13.9

28.8

2.5

3.3

4.5

216.1

2010

172.9

14.8

30.5

2.6

3.5

4.8

229.1

2011

182.0

15.5

31.9

2.7

3.7

5.2

0.2

241.1

2012

191.5

16.1

33.2

2.8

4.1

5.7

0.5

253.7

2013

199.6

16.5

34.0

2.8

4.5

6.2

0.9

264.5

2014

205.6

16.6

34.3

2.7

4.9

6.9

1.5

272.5

2015

210.0

16.7

34.0

2.6

5.4

7.5

2.2

278.5

2016

213.2

16.8

33.4

2.5

5.9

8.3

3.0

283.2

2017

214.9

16.8

32.3

2.4

6.5

9.2

3.9

286.0

2018

216.2

16.9

30.9

2.2

7.2

10.0

4.9

288.5

2019

217.5

17.0

29.7

2.2

7.8

11.4

5.8

291.4

2020

217.3

16.9

28.5

2.1

8.4

14.2

6.7

294.2

2021

216.5

16.8

27.2

2.0

9.1

18.3

7.6

297.5

2022

215.1

16.7

26.1

1.9

9.9

22.9

8.5

301.0

2023

213.0

16.6

25.1

1.9

10.7

28.0

9.5

304.7

2024

210.2

16.5

24.3

1.8

11.5

33.5

10.5

308.3

2025

206.7

16.4

23.6

1.8

12.3

39.2

11.7

311.7

2026

202.4

16.3

23.0

1.8

13.2

45.0

12.9

314.7

2027

197.5

16.3

22.6

1.7

14.1

51.0

14.2

317.3

2028

192.0

16.3

22.1

1.7

15.0

56.9

15.6

319.5

2029

186.0

16.3

21.8

1.7

15.9

62.8

16.9

321.3

2030

179.6

16.2

21.5

1.7

16.8

68.6

18.4

322.8

318

IV. Annex D - Calculated collection rate

Based on data collected from Eurostat the collection rate is calculated in Table 117.

Table 117: Calculated collection rate in EU 2014

404

Average EEE placed on

the market 2011-2013

Weee collected

2014

Collection

rate

405

Austria

17,270

8,415

49%

Belgium

40,998

13,028

32%

Bulgaria

2,986

3,790

127%

Cyprus

1,095

124

11%

Czech Republic

15,448

6,235

40%

Germany

172,507

126,943

74%

Denmark

13,955

5,405

39%

Estonia

1,281

331

26%

Greece

12,510

3,246

26%

Spain

48,850

14,263

29%

Finland

8,926

2,680

30%

France

158,873

34,478

22%

Croatia

3,699

317

Hungary

10,853

5,633

52%

Ireland

10,403

1,920

18%

Iceland

504

354

70%

Italy

68,298

20,983

31%

Liechtenstein

117

219%

Lithuania

2,250

1,422

63%

Luxembourg

1,604

412

26%

Latvia

1,256

400

32%

Malta

752

Netherlands

20,233

10,219

51%

Norway

16,831

5,570

33%

Poland

45,977

19,495

42%

Portugal

10,653

8,594

81%

Romania

14,240

1,021

Sweden

24,301

5,790

24%

Slovenia

2,458

940

38%

Slovakia

5,259

1,969

37%

United Kingdom

149,963

34,770

23%

Total

884,286

338,872

38%

404

Due to how the numbers are calculated it is possible to collect more than 100 % (This is also related to how the values are

compiled in each country)

319

320

V. Annex E– Test results from consumer organisations

NETHERLANDS

Consumentengids June 2017, Steeds wisselen van mondstuk?. p/52-56

55%

20%

10%

Model & Brand

Price

Test score

hardfloor

crevices

carpet dust

carpet fibres

full bag

ergonomics

dust re

-emiss

noise

dust

energy

bag (zak)or nobag

(bak)

power

Eur

180

7.5

7.7

7.8

7.1

7.6

10.0

8.0

7.5

8.3

5.9

6.6

zak

800

180

7.4

7.9

7.7

8.2

9.1

7.8

7.0

8.4

5.0

6.7

zak

800

180

7.4

7.8

7.5

8.2

8.9

7.9

6.7

8.3

5.7

6.5

zak

800

160

7.4

7.8

7.9

7.6

8.0

9.1

7.7

6.9

8.3

5.2

7.4

zak

700

170

7.4

7.9

6.6

8.7

8.3

10.0

7.7

5.0

9.9

8.0

6.0

zak

650

180

7.3

7.5

8.7

6.4

7.0

7.5

8.1

5.8

9.9

7.5

6.3

zak

650

170

7.3

7.2

6.9

6.3

7.0

10.0

7.5

6.2

9.8

8.2

6.5

zak

700

105

7.1

7.9

8.3

8.8

8.2

7.5

7.9

4.6

9.7

4.3

7.2

zak

600

240

7.1

5.5

6.5

9.5

8.0

5.6

9.9

7.6

6.8

bak

650

195

7.0

7.1

7.0

5.1

7.7

9.1

7.8

5.6

9.9

6.8

6.3

zak

750

230

7.0

7.2

6.7

7.5

6.2

10.0

7.2

6.7

8.4

5.8

6.6

zak

800

140

6.8

6.4

6.7

5.2

8.0

5.3

7.8

6.8

9.5

5.2

7.9

zak

600

160

6.8

7.4

6.8

8.8

8.2

6.4

7.8

4.8

9.9

4.3

6.2

bak

750

160

6.7

7.0

7.1

8.3

7.9

5.2

7.8

5.7

8.5

4.9

6.7

zak

600

170

6.7

5.8

7.0

2.2

7.8

3.7

8.1

6.1

9.9

10.0

7.7

zak

650

6.6

6.9

4.9

8.1

7.7

6.2

7.9

5.5

9.1

3.9

7.4

bak

700

6.6

6.9

7.5

8.4

5.7

7.5

6.9

4.7

7.7

7.2

7.7

zak

700

330

6.6

7.6

6.9

8.5

7.2

7.1

7.9

4.3

9.0

2.8

6.5

bak

750

100

6.4

6.7

4.4

7.6

8.1

6.4

7.1

5.9

8.7

3.0

7.2

zak

700

330

6.4

5.7

6.6

3.8

8.0

1.3

7.7

5.0

9.9

9.7

6.6

bak

700

155

6.2

6.0

6.1

4.0

7.8

3.4

8.1

6.4

8.3

4.0

6.5

zak

650

175

6.2

6.5

6.8

4.7

6.4

7.7

8.0

3.9

7.1

6.8

7.5

zak

620

290

6.2

5.6

7.9

4.8

5.7

2.0

7.1

6.7

9.6

4.0

7.4

zak

650

330

6.2

6.3

4.6

6.5

7.6

5.9

7.8

5.0

9.9

4.9

5.3

bak

900

205

6.1

5.8

8.9

3.9

7.7

1.3

5.6

4.7

9.7

6.1

6.8

bak

700

170

5.9

5.8

5.4

3.0

8.0

4.6

7.3

5.5

8.2

4.8

6.8

bak

750

250

5.8

5.1

2.7

4.6

8.0

1.3

8.1

6.1

9.8

7.7

1.5

zak

750

350

5.4

5.5

5.2

8.6

8.3

1.8

1.0

3.5

9.2

4.0

6.4

bak

800

5.3

6.4

7.1

4.7

6.1

7.9

7.8

6.0

4.3

3.8

8.0

zak

800

100

4.6

4.3

3.4

2.6

5.8

2.1

7.9

5.5

6.5

1.7

6.2

bak

750

average

187.8

6.6

6.7

6.6

6.2

7.4

6.3

7.4

5.7

8.9

5.6

6.6

714.0

321

NETHERLANDS

Consumentengids July/August 2018, Strengere regels voor stofzuigers. p/26-29

Model &

Brand

Price

Test score

Cleaning total

hardfloor

crevices

carpet dust

carpet fibres

full bag

ergonomics

Hardfloor

Carpets

dust re

-emiss

noise

Energy

Bag

power

Energy

efficiency

class

Eur

55%

20%

10%

180

7.5

7.8

6.9

8.6

8.4

7.0

5.4

4.7

4.0

9.4

8.5

7.8

Yes

550

A+

190

7.5

6.9

8.4

9.4

7.8

6.9

6.2

9.4

5.8

8.2

Yes

550

A+

140

7.3

7.9

7.3

8.5

8.3

7.3

5.4

6.2

3.2

9.3

6.5

6.3

Yes

700

150

7.3

7.2

6.9

6.3

7.0

7.5

6.2

8.5

4.7

9.8

8.2

6.5

Yes

700

155

7.3

7.5

8.7

6.4

7.0

7.5

8.1

5.8

7.7

4.7

9.9

7.5

6.3

Yes

650

240

7.2

6.7

7.4

8.3

8.4

1.1

8.0

5.9

6.9

5.5

9.9

6.7

Yes

650

300

7.2

6.8

6.5

8.2

2.1

7.7

6.1

6.2

3.2

8.3

8.5

Yes

650

A+

220

7.1

5.5

6.5

9.5

8.0

5.6

8.5

5.5

9.9

7.6

6.8

650

120

6.9

6.7

6.6

8.6

8.3

2.5

7.9

5.4

3.2

1.7

9.7

7.8

7.5

Yes

750

275

6.9

6.8

6.2

5.9

8.6

6.4

7.6

5.7

7.7

2.4

9.9

7.4

5.4

Yes

650

495

6.8

7.7

5.6

8.5

8.3

7.9

4.2

6.9

4.7

9.1

4.9

6.3

700

6.6

6.9

4.9

8.1

7.7

6.2

7.9

5.5

6.2

4.7

9.1

3.9

7.4

700

6.6

6.9

7.5

8.4

5.7

7.5

6.9

4.7

5.5

4.0

7.7

7.2

7.7

Yes

700

120

6.6

6.5

8.5

8.2

2.4

7.9

5.2

4.7

4.0

8.6

7.7

6.9

Yes

750

6.5

6.4

5.6

6.2

7.5

5.1

7.5

5.8

7.7

4.7

9.8

5.1

6.7

Yes

750

260

6.5

5.6

4.6

4.7

7.2

3.2

8.0

6.0

8.5

4.7

9.9

9.8

6.4

Yes

650

6.4

6.3

7.0

6.5

7.4

3.7

7.4

6.3

9.2

4.7

9.9

3.1

6.9

Yes

600

140

6.4

6.9

5.8

7.7

8.2

5.3

7.9

4.5

3.2

8.2

4.6

7.4

700

150

6.4

7.1

5.0

8.3

7.2

7.3

4.2

6.9

2.4

9.0

3.5

6.4

650

280

6.4

6.6

8.6

8.2

1.3

7.1

5.0

6.9

9.9

4.5

7.3

600

325

6.4

7.0

4.2

6.7

8.3

7.2

4.9

7.7

4.0

9.9

4.8

2.7

890

150

6.3

6.6

8.2

1.9

8.7

5.7

8.0

6.8

9.2

6.2

8.2

3.8

2.1

Yes

890

295

6.3

6.4

7.4

5.2

8.7

2.1

7.8

4.6

7.7

3.2

9.9

8.1

3.0

850

6.2

6.4

5.3

8.9

8.4

1.5

8.0

5.4

5.5

3.2

9.3

3.1

6.5

Yes

650

160

6.2

6.4

7.2

2.2

8.5

6.8

8.0

6.6

4.7

8.1

4.4

2.2

Yes

890

175

6.1

6.6

5.9

7.6

8.1

5.7

7.7

4.1

7.7

4.7

7.7

5.1

6.8

Yes

620

5.5

6.2

4.1

8.3

7.9

3.0

8.1

5.2

6.2

4.0

1.7

4.6

7.0

Yes

700

5.5

6.1

5.6

8.4

8.2

1.1

7.9

5.3

7.7

5.5

3.9

2.9

7.4

Yes

700

395

5.5

4.5

4.6

2.7

4.5

2.0

8.1

6.3

9.2

9.0

4.6

7.5

650

A+

5.3

5.9

5.2

8.3

5.8

3.9

7.3

4.7

5.5

3.2

9.3

8.1

7.6

Yes

700

5.3

5.1

2.8

4.0

7.5

4.8

7.5

5.2

8.5

4.0

8.5

4.5

6.9

650

4.8

4.4

5.8

6.2

6.8

1.0

4.6

8.5

6.2

8.4

2.4

7.2

700

average

176.5

6.5

6.6

6.1

6.8

7.7

5.3

7.5

5.4

7.1

4.5

8.8

5.9

6.4

693.4

322

BELGIUM, Test Achats, June 2017

Cylinder types

Model (all with bag)

Euro

200

800

265

700

7.2

194

800

7.4

120

600

5.6

133

600

5.6

134

750

5.8

325

750

7.8

181

750

8.1

181

750

6.7

162

800

5.8

213

650

7.5

700

5.4

161

600

5.3

319

650

7.8

163

650

6.2

108

750

5.9

157

800

6.2

150

800

6.3

700

5.8

700

6.3

116

700

6.1

700

4.8

Average

163

714

6.4

323

GERMANY Stiftung Warentest, 2017, Cylinder type vacuum cleaners

Model

Price Euro

Declared W

Measured W

Volume Receptacle

Cable m

Weight kg

Length hose cm

Previous, annulled Energy Label

classes

Energy

Re-emission

Carpet

Hardfloor

Bag

208

650

800

2.2

12.3

7.5

107

240

800

743

3.4

10.6

7.4

104

157

750

753

2.1

9.1

6.5

197

750

852

2.6

7.4

228

750

796

11.1

6.7

283

650

721

2.6

10.9

8.1

bag

278

700

786

1.7

10.7

8.7

108

286

800

899

1.5

9.5

8.6

103

310

650

743

1.9

10.7

8.1

192

750

803

1.3

8.9

7.1

187

700

782

1.5

11.9

130

800

766

2.1

9.3

6.7

225

729

787

2.2

10.6

7.5

94.8

GERMANY Stiftung Warentest, Feb. 2018, Cordless vacuum cleaners

Model

Price Euro

Volume

Receptacle L

Weight kg

Battery run-time min

Measured battery

charge time

(min.)

Battery price Euro

Maximum

power (W)

Minimum

power (W)

400

0.9

3.7

310

100

500

0.6

2.6

209

250

0.5

3.1

306

105

175

0.4

2.5

140

205

0.6

3.4

116

169

0.6

2.8

202

120

0.4

2.5

255

324

100

0.4

2.9

283

151

0.7

2.3

181

100

1.0

2.2

276

Avg.

217

0.6

2.8

228

325

GERMANY Stiftung Warentest, June 2018, Cylinder type vacuum cleaners

Model

Price Euro

Declared W

Measured W

Volume

Receptacle L

Cable m

Weight kg

Previous, annulled Energy Label classes

Energy

Re-emission

Carpet

Hardfloor

Bag

279

550

633

3.4

11.7

7.4

A+

227

650

827

2.2

12.1

7.5

165

600

684

2.4

9.8

6.2

229

700

845

2.3

11.8

7.4

130

750

764

1.6

9.0

6.4

125

800

757

2.0

9.1

6.8

159

500

474

2.3

8.9

5.6

A+

750

742

2.2

8.8

5.2

219

500

577

3.2

13.1

7.0

A+

700

687

1.8

8.8

6.2

750

770

1.8

8.6

4.9

No bag

355

700

786

3.7

10.8

8.7

340

550

598

2.6

9.5

8.6

A+

150

750

776

1.9

9.2

6.8

177

700

776

1.9

7.3

5.8

199

650

621

2.4

10.3

7.2

A+

250

750

755

2.5

11.1

8.3

700

717

2.0

9.1

5.9

149

800

758

2.4

8.0

6.7

250

600

633

2.0

9.6

7.3

Avg.

186

673

709

2.3

9.8

6.8

326

Other tests

https://robomow.jimdo.com/vorwerk-vr200/

327

VI. Annex F - Impacts over a lifetime of vacuum cleaners calculated in the

EcoReport Tool

Table 118: All impact categories for mains-operated household vacuum cleaners. The life cycle

phase with the highest impact for each of the categories is highlighted with red text.

Material

Manufacturing

Distribution

Use

Disposal

Recycling

Total

Other Resources & Waste

Total Energy (MJ)

675

192

206

2,436

-100

3,423

of which, electricity (MJ)

135

115

2,420

-24

2,647

Water – process (litre)

-8

Water – cooling (litre)

746

115

-39

876

Waste, non-haz./landfill

(g)

1,035

633

154

1,281

-247

2,901

Waste, hazardous/

incinerated (g)

-4

Emissions (Air)

GWP100 (kg CO

eq)

104

-6

155

Acidification (g SO

-eq.)

203

461

-42

712

VOC (g)

Persistent Organic

Pollutants (ng i-Teq)

-5

Heavy Metals (mg Ni eq.)

-11

PAHs (mg Ni eq.)

-22

Particulate Matter (g)

274

-19

356

Emissions (Water)

Heavy Metals (mg

Hg/20)

-22

Eutrophication (g PO

)

Table 119: All impact categories for commercial vacuum cleaners. The life cycle phase with

the highest impact for each of the categories is highlighted with red text.

Material

Manufacturing

Distribution

Use

Disposal

Recycling

Total

328

Other Resources & Waste

Total Energy (MJ)

883

284

230

8,320

-119

9,611

of which, electricity (MJ)

170

8,296

-2

8,511

Water – process (litre)

-2

Water – cooling (litre)

993

379

-12

1,438

Waste, non-haz./landfill

(g)

1,507

964

166

4,328

-347

6,667

Waste, hazardous/

incinerated (g)

132

-1

195

Emissions (Air)

GWP100 (kg CO2-eq)

355

-7

419

Acidification (g SO2-eq.)

253

1,573

-53

1,890

VOC (g)

185

188

Persistent Organic

Pollutants (ng i-Teq)

-8

Heavy Metals (mg Ni eq.)

-12

128

PAHs (mg Ni eq.)

144

-47

125

Particulate Matter (g)

342

-16

434

Emissions (Water)

Heavy Metals (mg

Hg/20)

-29

104

Eutrophication (g PO

)

Table 120: All impact categories for cordless vacuum cleaners. The life cycle phase with the

highest impact for each of the categories is highlighted with red text.

Material

Manufacturing

Distribution

Use

Disposal

Recycling

Total

Other Resources & Waste

Total Energy (MJ)

964

170

2,500

-103

3,639

of which, electricity (MJ)

551

2,495

-61

3,042

Water – process (litre)

145

-15

132

Water – cooling (litre)

342

114

-10

472

Waste, non-haz./landfill

(g)

1,119

324

136

1,294

-139

2,758

Waste, hazardous/

incinerated (g)

-4

Emissions (Air)

GWP100 (kg CO2-eq)

51.2

106.8

-6

170

Acidification (g SO2-eq.)

389

474

-46

877

VOC (g)

Persistent Organic

Pollutants (ng i-Teq)

-2

Heavy Metals (mg Ni eq.)

140

-17

163

PAHs (mg Ni eq.)

-12

Particulate Matter (g)

263

171

-30

426

Emissions (Water)

Heavy Metals (mg

Hg/20)

-12

Eutrophication (g PO

)

329

Table 121: All impact categories for robot vacuum cleaners. The life cycle phase with the

highest impact for each of the categories is highlighted with red text.

Material

Manufacturing

Distribution

Use

Disposal

Recycling

Total

Other Resources & Waste

Total Energy (MJ)

1,738

134

170

2,309

-33

4,324

of which, electricity (MJ)

1,125

2,302

-23

3,485

Water – process (litre)

287

-5

286

Water – cooling (litre)

536

107

-2

678

Waste, non-haz./landfill

(g)

1,955

441

136

1,200

-43

3,698

Waste, hazardous/

incinerated (g)

-1

131

Emissions (Air)

GWP100 (kg CO2-eq)

-2

210

Acidification (g SO2-eq.)

704

440

-14

1,198

VOC (g)

Persistent Organic

Pollutants (ng i-Teq)

Heavy Metals (mg Ni eq.)

270

-6

301

PAHs (mg Ni eq.)

-2

Particulate Matter (g)

526

171

-11

708

Emissions (Water)

Heavy Metals (mg

Hg/20)

107

-3

116

Eutrophication (g PO

)

330

VII. Annex G – Commercial vacuum cleaner Energy Index formulas and

parameters

The data and results presented in this section are based on the work of commercial vacuum

cleaner manufacturers. Multiple solutions and different equations were investigated to

arrive at a useful and representative EI measure. The following analyses are based on the

EI equations form section 9.2.1 and results from the existing measurement methods with

addition of the commercial debris pick-up test (section 9.7.4).

The results in Table 122 illustrate the sensitivity of the EI rating to each of the performance

parameters included in the equations. Each of the parameters listed in the first column was

varied from minimum to maximum (column 2 and 3) in the calculations for AE and EI,

respectively, to evaluate how large an influence it would have on the scales and how many

classes on the energy label.

Table 122: Results from commercial vacuum cleaner manufacturers on EI variation

Varied key

parameter

From..

MIN

..To

MAX

Range

ΔMAX-

MIN

Existing AE

calculation

(kWh/year)

New EI

calculation

/min)

Parameter variation

ΔMAX-

MIN

#class

ΔMAX-

MIN

#class

Nozzle

width

250 mm

400 mm

150 mm

7,9

1,3

1,34

1,9

Carpet dpu

75,0%

92,0%

17,0%

3,0

0,5

0,13

0,2

Hard floor

dpu

98,0%

115,0%

17,0%

4,5

0,7

0,13

0,2

Debris

pick-up

20,0%

100,0%

80,0%

1,25

0,9

Input

power

300 W

900 W

600 W

18,4

3,1

1,32

1,9

Sound

power

58 dB(A)

80 dB(A)

dB(A)

0,42

0,6

Table 123: Examples of EI values for commercial vacuum cleaners in the low/mid/best range

There are currently no vacuum cleaners in the “BEST” segment. Most are in the “MID”

segment, corresponding to around class D/E (section 13.4.6 on label classes). A and B are

thus empty.

331

The Comparison to the old AE-value shows the different weighting of nozzle

width and input power

same behavior and therefore same weighting of the fine dust cleaning

performance at both approaches

The Comparison shows again the larger influence of the nozzle widt

332

VIII. Annex H – Data for cordless vacuum cleaners

The data and test results presented in this annex are based on measurements performed

by the laboratory GTT (Suzhou GTT Service Co

406

). The tests are performed according to

the existing Ecodesign standards / draft standards and provided at no cost to the

Commission and Industry, in order to provide the study team with comprehensive data. In

total 13 cordless vacuum cleaners were tested by GTT and the full ecodesign test reports

were provided

407

. Below the summarised results of the tests can be seen.

• 4 out of the 13 stick vacuums (31%) achieved the minimum carpet cleaning

performance requirements of ≥ 75% per the current Ecodesign regulation. The

average cleaning of these 4 vacuums was 85% at an average run time of 12.5

minutes. Comparatively, 9 out of the 13 stick vacuums (69%) fell below the

Ecodesign performance requirements with an average cleaning performance of 57%

at an average run time of 19 minutes.

• Out of the 13 stick vacuums, 7 (54%) achieved the minimum hard floor crevice

cleaning performance requirements of ≥98% per the current Ecodesign

regulation. The average cleaning of these 7 vacuums was 103.3% with an average

run time of 20.5 minutes. Comparatively, 6 out of the 13 stick vacuums (46%) fell

below the Ecodesign performance requirements with an average cleaning

performance of 5.5% and an average run time of 19 minutes.

• 3 out of the 13 stick vacuums (23%) met the minimum dust re-emission

requirements of ≤1.0%. The average dust re-emissions of the 13 stick vacuums

was 4.0%, essentially 4 times the current limit.

• 5 out of the 13 stick vacuums (38.5%) met the noise level requirements of ≤80

dBA sound power when tested on carpet. 3 out of the 13 stick vacuums (23%) met

the noise level requirements on hard floor.

• Run time of all stick vacuums tested ranged from approximately 8 to 30 minutes

with an average run time of 18.5 minutes. There was very little difference in average

run time between carpet and hard floor.

406

http://gttlab.com/a/English/

407

Uploaded on the study website: https://www.review-vacuumcleaners.eu/documents

333

• Annual energy consumption ranged from 7.7 to 25.5 kWH/year on carpet for all 13

stick vacuums. The energy consumption calculations were based on the proposed

method for calculating effective power consumption of each vacuum per the CDV of

IEC 62885-4 cordless vacuum standard (It should be noted that no calculations of

annual energy consumption could be determined for testing on hard floor crevice

for 6 of the 13 stick vacuums because the hard floor crevice dust pickup results

were ≤20% cleaning, resulting in infinite energy consumption calculations per the

equation in the Ecodesign regulation.) The average annual energy consumption for

those stick vacuum products that met the current Ecodesign performance

requirements for both carpet and hard floor cleaning was approximately 13.9

kWH/year.

• None of the current cordless products tested meet all of the current Ecodesign

performance requirements.

The table below lists the models tested and the price segments to which they belong in

a random order. Price segments were divided as: high: >300€, middle: 150-300 €,

low: <150 €.

Model No.

Price

segment

Bosch VCA S010V32

High

Philip FC6823

High

Rowenta RS-RH5730

Mid

Bissell 2280N

Low

AEG AR180L21BCP

Mid

DEIK VC-R1093

Low

Hoover 94LD1711

Low

Vax Blade 24V DD767-2

Low

Gtech AirRam AR29

Mid

Dyson A7Y-UK-KHJ3008A

High

Hoover FD22G011

Low

Grundig VCH9630

Low

Black+Decker SVA420 H1

Low

334

Annual Energy (kWh)

Power mode

setting

Nozzle setting

Carpet

Hardfloor

Carpet

Hardfloor

Cordless no. 1

7.70

3.47

1 mode only

with brush bar

without brush

bar

Cordless no. 2

25.46

18.53

Max (3 modes)

carpet nozzle

hardfloor

nozzle

Cordless no. 3

16.96

1 mode only

brush bar open

Cordless no. 4

19.34

Max (2 modes)

brush bar open

Cordless no. 5

14.84

1 mode only

brush bar open

brush bar close

Cordless no. 6

20.92

16.79

Max (3 modes)

brush bar open

Cordless no. 7

18.15

7.52

Max (2 modes)

brush bar open

Cordless no. 8

15.27

10.08

Max (2 modes)

brush bar open

Cordless no. 9

13.06

Max (2 modes)

brush bar open

Cordless no. 10

21.61

Max (2 modes)

brush bar open

Cordless no. 11

13.47

1 mode only

brush bar open

Cordless no. 12

8.30

4.49

Max (2 modes)

brush bar open

Cordless no. 13

11.94

9.21

Max (2 modes)

brush bar open

Motor Rated

Power (W)

Battery Type

Battery

Volts DC

Battery

mAh

Number of

Battery Cells

Cordless no. 1

100

Lithium

2000

Cordless no. 2

525

Lithium

25.2

2600

Cordless no. 3

130

Lithium

22.2

2000

Cordless no. 4

Lithium

14.4

2000

Cordless no. 5

Lithium

14.4

N.A.

Cordless no. 6

Lithium

32.4

Cordless no. 7

Lithium

Cordless no. 8

Lithium

21.9

2100

Cordless no. 9

75 W

Lithium

2000

Cordless no. 10

Lithium

Cordless no. 11

Lithium

22.2

Cordless no. 12

100 W

Lithium

21.6

2150

Cordless no. 13

180 W

Lithium

21.6

2000

335

Dust pick up (%)

Dust re-

emissions (%)

Noise Carpet dB(A)

Noise Hardfloor

dB(A)

Carpet

Hardfloor

Brush ON

Cordless no. 1

74.1

105.6

7.866

83.1

84.7

Cordless no. 2

91.5

106.5

0.001

86.3

Cordless no. 3

5.7

7.025

85.2

Cordless no. 4

43.5

7.9

1.803

82.7

83.4

Cordless no. 5

44.2

3.4

4.251

79.1

78.5

Cordless no. 6

87.6

104.5

0.586

82.9

83.7

Cordless no. 7

58.4

99.7

8.648

83.4

Cordless no. 8

77.2

101.9

6.574

79.4

81.8

Cordless no. 9

8.5

6.29

79.2

81.7

Cordless no. 10

3.5

0.294

78.8

79.6

Cordless no. 11

62.1

3.7

2.852

77.2

79.1

Cordless no. 12

99.6

2.568

80.9

84.3

Cordless no. 13

85.7

105.2

2.98

84.8

Peff (W)

Runtime (min:s) on

carpet

Runtime (min:s) on

hardfloor

Carpet

Hardfloor

t90%rt

t40%rt

t90%rt

t40%rt

Cordless no. 1

160.62

114.47

07.09

29:05

19.44

42:45

Cordless no. 2

590.88

520.24

07.58

08.43

Cordless no. 3

232.09

219.50

12.42

20.20

13.37

20.45

Cordless no. 4

153.64

159.40

04.47

19.42

05.08

19.44

Cordless no. 5

126.36

119.58

14.28

15.03

16.24

16.54

Cordless no. 6

458.94

460.55

11.31

12.08

10.34

11.17

Cordless no. 7

236.99

203.71

20.42

23.51

Cordless no. 8

282.41

266.89

06.08

12.42

05.46

12.55

Cordless no. 9

165.53

153.25

06.19

20.41

06.39

18.31

Cordless no. 10

206.64

167.66

13.48

14.50

Cordless no. 11

183.47

139.06

22.53

23.24

21.30

23.35

Cordless no. 12

134.62

131.63

13.08

29:36

11.27

28:40

Cordless no. 13

243.47

243.68

07.39

15.55

07.18

14.25

336

Pictures of the cordless cleaners included in the data, in a random order:

337

338

339

340

341

342

343

344

345

346

347

348