+1

Unusually high drive efficiency in cold weather

berticus 1 month ago updated by Wunderlag 23 hours ago 3

Hi,

I had a drive today which reported nearly 300% efficiency which seems unusual.

My drive stats shows 45 wh/km whereas in the car it showed around 200 wh/km.

The only thing I see might be different is that it was much colder this morning.

When I started, I had the snowflake icon, and no regen, but halfway through the drive, it went away and regen was back.

Is it because less of the battery was available when I started and then more of it became available after the drive which skews the results?

For example, my battery is at 80% but because it's cold, only 75% is available (5% in blue). When I get to my destination, I have 70% left. So in reality I've used 10%, but the drive stats thinks I've only used 5% resulting in unusually high efficiency.

Is that what's happening? And if so, should there be a way to account for that to more properly measure the accuracy?

Hey, I just noticed something similar today. I have a drive that shows -555.7% efficiency, I ended the drive with more range than when I started :-) It's funny because TeslaFi does show the two battery percentages at the top of the page (normal and blue lower one). It seems that the wrong remaining range value is used when calculating the consumption. 

I preheated 16 minutes, my range went from 227.58 to 224.18km. I then drove 3.76km and ended at 224.86km of remaining range. Consumption of -25Wh/km, -555.7% efficiency. The thing is, I did not have any regen. I looked at scanmytesla and maxregen was at 0 for the drive, it was -8C outside. 

I believe this is a bug. 

I see a couple of other bug reports that talk about something similar, cold soaked batteries and range / efficiency calculation errors. Hopefully eventually this goes to the top of the list if there are enough of us :-) 

I have seen that behavior too, but it might be nearly impossible to change that. The issue is probably that teslafi can't work with battery percentage because only full percentage points are provided by the API, so Teslafi uses rated range only, and Tesla is sending only currently available rated range.

Now, Teslafi could theoretical factor in the percentage difference, but using 1% to compensate could be 0.5% or 1.4% in real life since we only get whole percentages... 

tough problem IMO...