Idifference in Smart and Solar mode not converging to 0

[demerstraat]

noticed Idifference never converges to 0. because integer division of 3 and 4 gives 0 as "increment" to Isetbalanced

[dingo35]

Thx for your PR, but this already came up 5 months ago.

If you can read Dutch, look af this post + the answers to it: https://gathering.tweakers.net/forum/list_message/77542200#77542200

I'll check if the behaviour is still the same in current version...

[demerstraat]

oh okay, wasn't aware. I have read and understood your concerns regarding jumps and the circuit limits. possibly linked though to my other commit regarding "calibration" of PWM vs. actual current drawn by EV. the offset there also keeps Idifference from reaching true 0.

with the offset in place, I've not seen jumps over 0.1A over the circuit limit that weren't corrected within seconds

[dingo35]

Is that offset based on any IEEE or other standard documentation, or did it just seem like a good idea to pump the current up?

[demerstraat]

it is within the tolerance of 1 percentage point as per table A.5 of IEC 61851-1

[dingo35]

Great thx, I wouldnt want to bring the manufacturer into problems !!

[mstegen]

I have seen that there can be a difference in what the EVSE reports as the (max)charge current, and what the EV actually draws from the grid. Took some measurements from my model S charging at 26A down to 6A Here are the results:	advertised	measured	reported by EV
26A	24.2A	24A
25A	23.8A	24A
24A	23.1A	24A
23A	22.0A	23A
22A	21.0A	22A
16A	15.0A	16A
6A	5.1A	6A

So it almost always draws 1A too little current, even though the car reports the correct charge current This should not be compensated for by increasing the PWM. It just the way it is, and other cars may behave different. It's helpful when solar charging as you only need 5A to keep charging.

I agree that Idifference not converging to 0 could be improved with your PR.

[demerstraat]

thanks Michael, for information, I tested this again against Skoda Superb IV (1x16A OBC) and Ioniq 5 (3x16A OBC).

"correction" only worked on the entire range (6 to 16A) on Ioniq 5, most likely because the OBC will also take 1x32A in single phase configuration, so CP dutycycle of >16A are "accepted" and acted upon by the OBC.

for Superb IV "correction" only had effect between 6 and 15A, as the OBC does not seem to take into account CP dutycyles of >16A.