johnr said:
davewill said:
It's not that's wrong so much, it's that it steps in 8.3% increments. So the actual decline is probably very small, and you just happened to cross the bar noundary.
Exactly. But why does Carwings charge gauge step in 8.3% increments? Can't it read the exact percentage from the Leaf's BMS? That's just one of the software quirks that drives me nuts.
No, it can't. Or at least it certainly doesn't. The only information Carwings uses is the number of bars. Period.
Volusiano thought he charged to 83%, but Carwings was just misleading him. He charged to 80% because that is what the timer told the BMS to do. 80% is more than 9 bars, so it shows as 10 bars, even though the tenth bar is only "partly full". Carwings saw 10 bars out of 12 and calculated 10/12 = 0.8333, so said 83%. A week later Carwings asked the car, "how many bars?" This time the logic in the car decided it was only 9. So Carwings decided 9/12 = 0.75.
Now, I have observed quite a few times that the car will be showing x bars, but if I turn it off and back on it will show x-1 bars. Yes, there is a chance that I was right on the margin, and the power required to shut down and reboot pushed it over the edge, but it happens too often for me to really believe that. I think there is indeed a software quirk
in the car that causes it to use two -- no, make that three -- different formulas for computing bars. We know that it uses different formulas when charging and when discharging. It appears to use a third formula when coming back to life after having been powered down.
But a Carwings software quirk? No, what you are seeing is a data transport design limitation, that was made much worse by a bad user interface design. Their data has one digit precision, and they are displaying it in a form that makes you believe it has two digit precision. Both are design issues, not programming quirks.
Ray