Note on the draft RDE test protocol – Appendices 1-4; Version 7 of 26 September 2014

1. Introduction

This document details (i) key changes introduced into the Appendices 1-4 of the draft RDE test protocoland (ii) discussion points for the upcoming meeting on 7 October 2014. The changes in the protocol reflect the discussions of the RDE drafting group on 4 September 2014 and the feedback ofMember States and Industry Stakeholders received via email.

2. Key changes

Definition of noise (see also paragraph 2.6.1.5): Following the response of PEMS manufacturers we left the 10s provision for noise unchanged but provide now the possibility to account in addition for the response time of analyzers.

1.2.2 Test parameters: We specify now a measurement frequency of 1 Hz or higher and a minimum accuracy of 0.01 Hz. We also include a passage according to which ECU parameters should be broadcasted at a frequency double that of the parameters recorded by PEMS to ensure correct sampling.

In Table 1, we now remove the specifications on speed measurements from footnote 6 and introduce a new section 1.3.6 that details how vehicle speed should be measured. Following the concerns of Member States, we refrain from defining ECU speed as the reference signal. We kept the placeholders for PM, as well as PN and THC as these pollutants are regulated under Euro 6.

1.2.4. Installation of PEMS: We now specify that PEMS shall be installed into the vehicle at a location that is as far as possible free of dust and negative interference of ambient weather conditions.

1.3.1 PEMS leak check: We shift this section from the old Appendix 3 now to Appendix 1.We left the provisions unchanged.

1.3.2 Preparing the sampling system: To avoid ambiguity, we rephrase the heading and clarify that the sampling system shall be keptclean by following the instruction of the manufacturer.

1.3.5 Checking the analyzer for measuring particle emissions: Although controversially discussed with stakeholders, we decided to keep the reference to PN measurements.

1.3.7 Measuring vehicle speed: We now clarify that vehicle speed shall be measured by a sensor or a GPS (which requires further quality checks) and/or the ECU. We ask stakeholders to check provisions in the related paragraphs 2.7, 3.2.3, 4.6) and provide additional comments.

1.3.8 Check of PEMS set up: We now include here a minimum consistency check for the reported ECU engine speed. This consistency check is required because we specify in paragraph 4.4 that emission measurements shall be excluded from the analysis if engine speed is <500 ppm.

1.4.2 Test run: The second text section of this paragraph should be discussed: Can PEMS meet the drift requirements without auto zeroing, assuming that a test lasts about 2h? If not, what would be an appropriate interval for auto zeroing? Do we need to prescribe the use of N2 or ambient air for auto zeroing?

1.4.3 Test end: Following the request of stakeholders, we do not require any more that data recording shall continue until 30s after the test is completed; we refer now only to the elapse of the response of the sampling system.

1.5.2 Checking the analyzer for measuring particle emissions: We now remove the reference to the sampling frequency.

1.5.3 Checking the on-road measurements: Following the comments, we adapt the requirements and allow now that 1% of the measurements exceed the calibration range by up to a factor of 2. We would refrain from adding additional clarifications other than the ones contained in the lastsentence of this paragraph.

Appendix 2: We put the Appendices in chronological order by placing the former Appendix 2 last (now Appendix 4).

2.2.1 Linearity requirements:Following the requests, we reintroduce air flow and fuel flow into Table 3 and do not require any more an accuracy check for calculated exhaust mass flow. We decided to keep the linearity requirements for exhaust gas mass flow unchanged (in line withR49). We removed the linearity requirements for gas dividers, particle number, and for ambient pressure (the latter parameter is covered in Table 6).

2.2.2 Frequency of linearity verification: Following the requests, we specify now a verification frequency of 12 months.

2.2.3.1 General requirements: We leave the recording frequency unchanged as the linearity verification is done at steady state; a reference to the recording frequency applied later by the PEMS user may not be required.

2.2.3.3 Requirements for linearity verification on a chassis dynamometer: We now permit the use of the CVS for calibration.

2.3.2.4 Noise: Following the comments of stakeholders, we now permit an additional time allowance that accounts for the response time of analyzers. Additional comments on the existing provisions are welcome.

2.3.3.2 Efficiency test for NOX converters: Following the requests from stakeholders, we now specify that the NOX converter shall be tested no longer than one month before the emissions test.

2.3.3.3 Adjustment of the FID: We now refer in sub-paragraph 1 and 2 to R83 but kept the detailed provisions in sub-paragraph 3 for the oxygen interference check

2.3.3.5 Interference effects: Following the stakeholder comments, we now specify that interference/functionality checks have to be performed by the instrument manufacturer once per type of instrument prior to first use/release.

2.3.3.5.2 CO analyser interference check:We now specify that the analyser response shall not be more that 2% of the expected mean concentration during the test or 50 ppm, whichever is larger.

2.3.3.5.4.1 Procedure of NDUV quench check: Following the stakeholder comments, we now introduced requirements based on EPA CFR 1065. We ask for comments on the provisions (considering also other paragraphs under 2.3.3.5).

2.4.2 Gas dividers: We decided to keep the text (instead of referring to R83) to provide more detailed specifications following R49 (see text after line 4 in this paragraph). We removed, however, the specification that the primary gases used for blending shall have an accuracy of 1% as well as the accuracy requirements for the gas dividers used for the linearity verification.

2.6 Instruments for measuring exhaust mass flow: We rearranged this section; the provisions now refer to EFMs. We clarify upfront which methods are permissible to measure exhaust mas flow and which requirements have to be fulfilled by the individual methods.

2.6.3 Response time check: We left the rise time of the EFM unchanged at 3s as the time interval appears to not to be the limiting factor given the rise time of the analyser. Well-reasoned comments are very welcome; in principle JRC is open to reduce the rise of the EFM time to the initial proposal of 1s.

2.7 Sensors and auxiliary equipment: We now include accuracy requirements for air flow and fuel flow sensors. We ask stakeholders to check and comment on these.

3.2.1 Frequency of PEMS validation:Following the comments of Member States, we now recommendto validate the installed PEMS once for each PEMS-vehicle combination either before the test or, alternatively, after the completion of the test. These provisions shall be discussed once more with stakeholders.

3.2.2.2. Test conditions: Upon written stakeholder request, we now introduce the possibility to conduct validation testing next to following the R83-TA procedure also based on “any other adequate measurement method”.

3.2.3 Permissible tolerances for PEMS: We now re-introduce in Table 7 relative tolerances [%]. Upon request, we tighten the tolerance for driving distance and include a footnote that the tolerance can be met after correcting the ECU measurements for the deviations observed during the validation. Based on our recent measurements, we tighten the tolerances for THC, CH4, NMHC but leave the tolerances for CO2 unchanged.

3.3.2 Exhaust flow validation procedure: The provisions should be discussed once more, also within the RDE group. As of now, we prescribe:

  • exhaust mass flow measured by standard EFM – linearity check, accuracy etc. requirements in Appendix 2 – no validation needed
  • exhaust mass flow calculated from calibrated air flow and fuel flow sensors – linearity check, accuracy in Appendix 2 – no validation needed
  • exhaust mass flow calculated from non-traceable ECU air flow and fuel flow or non-traceable EFM: linearity on the chassis (Appendix 2), transient exhaust flow validation (Appendix 3)

The provisions should be adapted based on the outcome of the discussions in the RDE group.

4.2. Time alignment: We shortened and adapted the previously proposed time alignment approach. Suggestions for improvement of the current approach are most welcome.

4.3 Cold start: Based on the discussions with Member States it was decided to define cold start based on either of two criteria: the coolant temperature has reached 70 oC or alternatively a time of 5 min has passed since test start. The provisions should be adapted upon request of the main RDE group.

4.4 Emission measurements during engine stop: We now specify that any instantaneous emission or exhaust flow measurements obtained during engine switch off shall be excluded from the data analysis according to Appendix 5. The combustion engine shall be considered as switched off or deactivated if two of the following criteria are fulfilled: the recorded engine speed is <500 rpm; the exhaust mass flow is measured at < 3 kg/h; the measured exhaust mass flow drops to <50% of the steady-state exhaust mass flow at idling. Feedback is recommended.

4.5 Consistency check of vehicle altitude: We now clarify how GPS data should be used to verify the altitude of a trip. Comments on the newly introduced provisions are welcome.

4.6 Consistency check of vehicle speed and driving distance:This section specifies now the consistency check of GPS data only. The permissible deviation in the total trip distance is set to 4%. Comments on the feasibility are welcome.

4.7.2 NOx correction for humidity and temperature: The provisions remained unchanged but may be adapted depending on the discussions of the main RDE group.

4.8 Determination of the instantaneous gaseous exhaust components: Following the discussions on 4 September, we simplified the section and removed the more detailed calculation of instantaneous mass emissions based on exact equations.

4.8.2 Calculating NMHC and CH4 concentrations: We shortened the section and removed the reference to the calibration requirements; these are now only specified in paragraphs 2.3.3.4.1-2.

4.9 Determination of exhaust gas mass flow: Based on the discussions, we still provide the possibility to calculate exhaust mass flow from ECU data or from the air/fuel flows obtained by sensors or the ECU. Depending on the discussions in the RDE group, the provisions in this paragraph can change.

4.12 Reporting of instantaneous emissions: We have adapted the section and refer now to the parameter listed in Annex IIIA. We assume that the data calculation and evaluation would be integrated in the PEMS post-processing software. To provide sufficient flexibility, we refrain from requiring a specific structure or data format but recommend reporting the instantaneous emissions as well as any other relevant parameters in a widely compatible file format, e.g., a csv-formatted or txt-formatted data file.

3. Key discussion points

  • Measurement of vehicle speed
  • Definition of cold start
  • Correction of NOx emissions for changes in ambient humidity
  • Verification and validation of non-traceable ECU data and simplified methods to determine exhaust mass flow
  • PEMS validation
  • Time alignment
  • Zero checks of analyzers during the test

1