12 V SoC – Monitoring - Assessment of Tolerances Audi AG WLTP-DTP-LabProcICE-105

2 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Regulation R 101 ► 5.3. Test results ► The test results (fuel consumption C [l/100 km] and CO2-emission M [g/km]) of this test are corrected in function of the energy balance ΔEbatt of the vehicle’s battery. ► The corrected values (C0 [l/100 km] and M0 [g/km]) should correspond to a zero energy balance (ΔEbatt = 0), and are calculated using a correction coefficient determined by the manufacturer as defined below. ► … ► The electricity balance Q [Ah], measured using the procedure specified in appendix 2 to this annex, is used as a measure of the difference in the vehicle battery’s energy content at the end of the cycle compared to the beginning of the cycle….

3 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Regulation R 101 (ctd) ► Under the conditions below, it is allowed to take the uncorrected measured values C and M as the test results: ► 1) in case the manufacturer can prove that there is no relation between the energy balance and fuel consumption, ► 2) in case that ΔEbatt always corresponds to a battery charging, ► 3) in case that ΔE batt always corresponds to a battery decharging and ΔE batt is within 1 per cent of the energy content of the consumed fuel (consumed fuel meaning the total fuel consumption over 1 cycle). ► The change in battery energy content ΔE batt can be calculated from the measured electricity balance Q as follows: ► ΔE batt = ΔSOC(%) ·ETE batt ≅ ·|ΔAh|·V batt = ·Q·V batt (MJ) ► with ETE batt [MJ] the total energy storage capacity of the battery and V batt [V] the nominal battery voltage.

4 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Regulation R 101 (ctd) ► Fuel consumption correction coefficient (K fuel ) defined by the manufacturer ► The fuel consumption correction coefficient (K fuel ) shall be determined from a set of n measurements performed by the manufacturer. This set should contain at least one measurement with Q i 0. ► …..

5 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Equipment for Monitoring of SoC  0,1% rdg and  0,2% f.s.  0,5% rdg and  0,05% f.s. rdg accuracy  (0,1 2 +0,5 2 ) = 0,51 % f.s. accuracy  (0,2 2 +0,05 2 ) = 0,21 % f.s. = full scale rdg = reading Error depends on full scale and actual values

6 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Theoretical approach – 200 A over NEDC Max dev. = 0.48 Ah Current 200 A

7 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Theoretical approach – instationary test, 3 Ah over NEDC Max dev = 0.16 Ah (In WLTC ~ 5 Ah corresponds to  Ebatt = 1% Energy of consumed ~100 g/km CO 2 )

8 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Tolerance of CO 2 -Measurement ► The error in the determination of CO 2 is a result of tolerances in various systems within the complete emission test procedure: ► Tolerance of measurement equipment Span Gas Linearization of Analyzer Span Gas Divider Dyno specifications (tolerances) ► Tolerance of vehicle (e.g.) Battery energy management (charging/decharging) Operational strategies (adaptions etc.) Reference fuel (density) ► Tolerance of lab processes Speed trace – shifting points Soak temperature With improved WLTP procedures and reduced tolerances an overall tolerance of 2 % in CO 2 measurement is estimated. Results of Validation Phase 2 and Correlation will provide further evidence

9 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Assessment of Tolerances for Correction of CO 2 as f(SoC) Application of R101 provision

10 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Conclusion ► The error in the determination of electrical charge with an typical ammeter consists of full scale and reading error. For a typical current of a conventional engine it is assessed at 6 %. The error in the determination of CO 2 is estimated at minimum 2 % over the whole chain of measurement equipment, vehicle and lab process. ► For a vehicle emitting 100 g/km CO 2 in WLTC Cycle,  Ebatt of 1 % of consumed fuel energy is about 5 Ah. The tolerance of measured energy - depending on charging strategy - is about 0.3 Ah. ► For correction of CO 2 following R101 the ratio measurements at different  Ebatt of the low voltage batteries have to be performed (feasibility of this proposal for low voltage batteries is still to be evaluated) ► The assessed tolerances of CO 2 and SoC measurement result in an overall tolerance in the determination of the ratio CO 2 /  Ebatt of about 60 %. ► The correction process for CO 2 as f(  Ebatt ) for a low voltage battery is still very much uncertain. Further data have to be provided through VP2

11 Dr. F. Mühlberger, Dr. M. Bergmann, Dr. Kolesa AUDI AG Proposal Options: ► Make no proposal from OICA (Tuev Nord proposed following R101) ► Apply R101, but stay with a correction threshold of  Ebatt > 2 % of consumed fuel energy for low voltage batteries monitoring. Further evidence will be available after Validation Phase 2 and WLTP Correlation. ► Use of a fix correction function dependent on used battery energy (1 % of consumed fuel energy = 1,6 g/km CO2) without threshold. …..