Page 1 of 1

Accuracy of state of charge indicator?

Posted: Fri Jun 02, 2023 1:28 pm
by mluisbrown
The other morning I started with 80% SOC (I usually charge to 80%), did a short trip of about 7km and parked. When I came back to the car to return home, it was showing 81% SOC 😅 I did the 7km home (all city driving) and arrived back home with it indicating 80%, the same as I had started with.

This leads me to believe that the SOC shown is not super accurate.

Re: Accuracy of state of charge indicator?

Posted: Fri Jun 02, 2023 2:01 pm
by daern
mluisbrown wrote: ↑Fri Jun 02, 2023 1:28 pm This leads me to believe that the SOC shown is not super accurate.
It's not. But none of them are.

It's all guess work really and some very complicated algorithms to attempt to work out what's left in the battery based on what went in, what's been taking out and some relatively rudimentary readings on what you can see outside (voltage and temperature, basically). This is a problem with modern batteries and it's best to assume that the percentage reading is just a "best guess" at that point in time based on the chemistry of the battery and a load of pre-performed tests calibrated into the algorithm.

As to why it varies? Temperature is a good basic variable that can affect the overall capacity of the battery and, consequently, the amount of usable charge left in it at a given point.

My experience is that the MEB does a pretty good job of guesstimating what's left and that whatever accuracy level it has, is no more variable than my driving is - i.e. I can impact it far more by the way that I drive, or the conditions on the road, than any slight inaccuracies in the reading would have.

Re: Accuracy of state of charge indicator?

Posted: Fri Jun 02, 2023 3:29 pm
by alanjs
I don't think there is any logic behind the figures that are shown,
On Wednesday morning, after charging to 80% "My Cupra" showed a potential range of 153 miles.
On Thursday, with no trips made, it showed 80% with a potential range of 96 miles.
On Friday, after a 34-mile round trip, it showed 63% with a potential range of 118 miles.

I place more faith in the Tronity app which gives much more consistent information.

Re: Accuracy of state of charge indicator?

Posted: Sat Jun 03, 2023 4:56 pm
by davmatjo
alanjs wrote: ↑Fri Jun 02, 2023 3:29 pm I don't think there is any logic behind the figures that are shown,
On Wednesday morning, after charging to 80% "My Cupra" showed a potential range of 153 miles.
On Thursday, with no trips made, it showed 80% with a potential range of 96 miles.
On Friday, after a 34-mile round trip, it showed 63% with a potential range of 118 miles.

I place more faith in the Tronity app which gives much more consistent information.
The Thursday reading will be due to a long standing bug that will sometimes run a kilometers to miles conversion on a number somewhere in the process that is already in miles. The bug has been present since day 0 as far as I know.

Re: Accuracy of state of charge indicator?

Posted: Sat Jun 03, 2023 9:53 pm
by TheBornIDentity
The Cupra app range is complete nonsense; ignore it, it’s bugged.

The in car SoC is susceptible to environmental changes such as temperature. It’s not uncommon for it to go up, especially when the temperature has gone up a lot since the car was on charge.

Re: Accuracy of state of charge indicator?

Posted: Sun Jun 04, 2023 9:37 am
by alanjs
davmatjo wrote: ↑Sat Jun 03, 2023 4:56 pm
The Thursday reading will be due to a long standing bug that will sometimes run a kilometers to miles conversion on a number somewhere in the process that is already in miles. The bug has been present since day 0 as far as I know.
Yep, I'm aware of that - it happens frequently.

Re: Accuracy of state of charge indicator?

Posted: Sun Jun 04, 2023 9:41 am
by alanjs
TheBornIDentity wrote: ↑Sat Jun 03, 2023 9:53 pm The Cupra app range is complete nonsense; ignore it, it’s bugged.

The in car SoC is susceptible to environmental changes such as temperature. It’s not uncommon for it to go up, especially when the temperature has gone up a lot since the car was on charge.
Oh yeah and then some! I just found it interesting that it changed, all by itself, sitting in the garage. But as above, it's probably due to a conversion to Kilometers, then it woke up.
It goes along with all the unplanned, unrequested changes to all sorts of stuff that the car makes up its own mind to do. Just something you get used to.

Re: Accuracy of state of charge indicator?

Posted: Sun Jun 04, 2023 9:49 am
by Elea4
Interestingly, reading the SoC value with an OBDII adapter gives slightly different values, usually a little bit higher than the in-car displays.