After looking at AC charging efficiency and finding it dismal at the very low power level I have available at home, I've turned my attention to DC charging. The nearby 50kW ABB DC charger is relatively cheap compared to most in NZ, so I'm trying to weight that cost and convenience v.s. charging at home. The first interesting finding is that the OBD2 PIDs "000_Battery Current(A)" and "000_Battery Voltage(V)" appear to directly reflect the DC charger output rather than what the names suggest. This is a huge win because I already know what the battery takes on using the coulomb counter "odometer" PID CEC. While charging there are numerous items on the high-voltage DC bus. The charger, the battery, the LDC, the battery heater and parts of the climate system are all wired in parallel. What I'm getting at is that all those items are subject to the same voltage at any moment, that voltage which the BMS directs the charger to provide to maintain current between what the charger can offer and the battery can accept. So, not all the current provided by the charger goes to the battery. The PID 000_Battery Current tells me the total going in and the PID CEC tells me how much energy is taken in by the battery alone, clicking over every 0.1 kWh like a domestic electricity meter. The difference is what the other HV systems are using, in particular the battery heater and LDC (because I'm not using climate while running these tests). The only information the charger vendor provides is the total energy at session termination, with a resolution of 0.01 kWh, and the session time in seconds. I can use those to derive the average power output, adjusted by an allowance for startup and shutdown delays. The first graph shows current and voltage from the PIDs mentioned, noting the current is around 125 A, the maximum I think for this charger. The second graph shows power (V x A) from those same PIDs plotted against the billed energy plotted as an average over the same timespan, just a straight line. As you can see they are a very good match and so I'm going to accept that as evidence that those PIDs represent charger output. Further, the second graph shows energy accumulation between what the charger provides and what the battery accepts. The takeaway here is that there are startup loads that reduce power to the battery for the first few minutes, then the lines become parallel, indicating nearly 100% is going to the battery. But over the entire 20-min session the battery only gets 85% of the energy delivered by the charger, the unaccounted energy being 2.29 kWh. I'm unclear as to if the battery heater was one of those loads because my starting battery temp was higher that what I believe is the heater shut-off temp, 15°C. However, the only other plausible use of that energy is the LDC and I'd be surprised if it drew that much. So, it could be the heater and unfortunately I hadn't anticipated this and thought to log the relevant heater data. I did a second charge shortly afterwards and I'll post that graph in the next post.
The second test was carried out about an hour later and so the battery was still warm. I monitored the battery heater PID and found no activity. The overall energy accepted by the battery was 95% of the energy supplied. As before, once past the initial few minutes, efficiency heads to nearly 100% indicated by the parallel lines. The unaccounted energy lost was 0.5 kWh averaged over the 14.7 min session. That's 2.2 kW roughly over the 14.7 minutes. If all that went to the LDC that's about 120 A on the "12V system" (actually 14.7 V) at an estimated 80% conversion efficiency. That's a large amount considering there is some evidence that the "12V system" needs only 0.2 kW to stay alive while driving. We do know that the 12V battery is charged during this time but I can't quite accept (yet) that this would account for so much energy. There may however be losses to the 12V battery while charging is initiated that must be replenished. So, I'm not yet convinced I understand where this "lost" energy is going, assuming I haven't made any mistakes.
