Had to correct the image. Can't edit the post anymore. The area below the charge curve extrapolated to 100% is 28.2kWh not 27.9kWh, and the stated efficiency is 98%, not 98.9%.

 

Should I correct this in the text?
Should the picture also be corrected?

 

The picture was corrected, it says A=28.2kWh instead of 27.9kWh. I forgot to adjust to a 0 to 100% charge. If you can correct the text, it should say "27.6kWh, divided by the amount of energy used to charge, 28.2kWh. In this case, it is ~98.0%" where it says "... charge, 27.9kWh. In this case, it is 98.9%"

In the other post I estimated your degradation (not of yourself!) at 2 - 2.5%. How would such an assumption on degradation relate to the numbers found here?

I think degradation could show itself in different ways.

• A) The relationship between BMS SoC and display SoC could change, my car could be eating the reserve below 5% and would then presumably eat the reserve above 95%. As I understand, you and JejuSoul believe this. In this case, as I understand, my OCV at rest per cell at 100% SoC would be the same as any newer car, because I have 95% BMS SoC as I know you do too. My car wouldn't be charging more than any other, it would only be discharging deeper, which, after all, is less bad for degradation.
• B) The relationship between BMS SoC and display SoC could be fixed, as I think it is. Also, BMS SoC is just another abstraction. The reserve could be simply charging the car to a higher voltage. This is on my mind because I was thinking that the car charged only to 4.10V per cell, and now, since I'm recording data, I see 4.14V per cell. If the car is charging to a higher voltage, and still to 95% BMS SoC, that means that the BMS SoC to voltage relationship is not fixed, and BMS SoC is irrelevant. This would mean for us that we should charge to a lower SoC in the future to minimize degradation.
• C) The relationship between BMS SoC and display SoC and OCV at rest could be fixed. In that case, we would only see each % yielding less energy because of the battery holding less Ahs. In that case, there is no reserve (what is our source on the existance of a reserve?), and what I see on this plot could be loss of capacity already, getting 27.6kWh instead of 28 or 28.8kWh.
• other?

If hypothesis A is true, then I'm discharging deeper and the cut on Max POWER should appear earlier to me that it appears to a newer car. Sounds plausible to me.

Hypothesis B also sounds plausible... I was convinced I saw 4.10V or 4.12V at 100% before I started logging (I only have one month of data logged).

Hypothesis C also sounds plausible, one always gets the feeling that the car used to yield more energy...

Trying to guess whether a 0.1kWh difference is meaningful when the margin of error is a magnitude greater than this is pointless. Statistically you must take tens of readings and make an average of them to reduce the error. I don't recommend doing this however. Charging the car to 100% and reducing it down to empty is best done infrequently, because this process will damage the battery. Most of the time it's best to stick between 80% and 20% SOC.

Do you think the error is that bad? I know the approach is questionable, because a standard discharge curve is done at a constant C-rate and I can't do that... However, the instruments, the counters in the BMS appear to me very precise, even though drift is inevitable.

Don't worry, the plots are all from day to day use of the car. No gratuitous charge or discharge was done to obtain this data, no animals were hurt either, except a discharge to 1.0% on purpose when the SoC was already very low.

 

The discharge curves of a battery are the most informative.
.....
I'm only getting the number 27.9kWh for energy charged from 0% to 100%. I should get 28kWh from a discharge. Is my battery degraded already? What numbers do you get?

I could have about 98% SoH even though the PID published by the BMS indicates 100%. That would also mean that I either exausted the reserve capacity or that actually none exists.

In the other post I estimated your degradation (not of yourself!) at 2 - 2.5%. How would such an assumption on degradation relate to the numbers found here?

 

...I'm only getting the number 27.9kWh for energy charged from 0% to 100%. I should get 28kWh from a discharge. Is my battery degraded already? ...

Trying to guess whether a 0.1kWh difference is meaningful when the margin of error is a magnitude greater than this is pointless. Statistically you must take tens of readings and make an average of them to reduce the error. I don't recommend doing this however. Charging the car to 100% and reducing it down to empty is best done infrequently, because this process will damage the battery. Most of the time it's best to stick between 80% and 20% SOC.

 

Yes, any perceived different is likely to be as a result in change of air resistance. Drag increases as a square of the speed, so going faster increases energy draw exponentially, while the imponderable that is never considered in a real-world test is the cumulative effect of nett side-wind - the car is rarely directly into, or out of the wind, so with that and relative slope of road, that graph is close to as accurate as can be given the imponderables.

 

Yeah, given battery hysteresis, averaging makes sense. The 0.2C discharge curve is driving at a maximum of 90km/h and average slightly under 60km/h, 0.2C meaning that it would take 5h to drain the battery. At 120km/h you get the average 0.7C discharge, it would drain the battery in 1h30'.
The curves above tell that not only you spend more energy, but the battery also yields less energy. Energy loss due to hysteresis is bigger. So, it's a double loss.