Submitted by WardT on Thu, 2019-04-04 08:13
I've heard it said the Tesla battery range prediction is based on the standard Wh/mi constant of 241.9. I think the 241.9 is the number used in the EPA battery miles calculation. If that is correct, when I take a trip with a lower Wh/mi average than 241.9, my actual trip miles should be more than the predicted battery miles. The miles I get between charges is never more than the predicted miles. I average 82% of prediction. The 82% average is the number of actual miles between charges divided by the difference between the initial predicted battery miles and the final predicted battery miles. Of course, there are the phantom losses, but things still don't add up. I have a spreadsheet recording every detail of the trip between charges: miles driven, starting predicted miles, ending predicted miles, kWh consumed (by the trip report) and date/time stamp. I calculate what the phantom losses would have to be for the battery prediction to be 100% of actual between each charge based on the actual trip Wh/mile. That phantom charge calculation averages about 12 miles/day. When the car sat at LAX for a week, the calculated phantom charge was only 2.3 miles/day, which is what I think is the real number. I've also heard the Energy graph is better, but I haven't checked it.
My question is how are the predicted battery miles calculated? It can't be based on a Wh/mile constant. If it were, when I have a trip with a low (200 Wh/mile) energy use, I would get more actual miles than predicted miles. Given the way we drive and our climate, my all time Wh/mile is down to 237 over the 12,000 miles and one year of ownership. My last 10 trips between fill-ups of electrons has averaged 225 Wh/miles, yet my actual range is only 82% of battery prediction.