Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Energy accounting

This site may earn commission on affiliate links.
My MS60 was sitting in the shop for several weeks waiting repairs, so I plotted up the rated range as a function of battery capacity as the vampires drank the juice. As you can see (if the image upload worked), the rated range definitely goes to zero when the battery reaches 10% capacity.
Jaggies on the line are the 1% charge increments.

View attachment 29887
Interesting, but this is just one dataset, correct? So not many many dataset overlaid ontop of each other? That would explain why you have no overlap at all.
 
2.5 million rows
Well, the vast majority of them were of course redundant (since this is from the streaming API, so only full miles). But the resulting graph looks remarkably similar to mine.
This clearly confirms that there is no straight simple relationship between rated miles and reported SOC.
Thank you so much for sending these - this really helped me make sure that my data isn't just a fluke.
If a few more people would be willing to do so (especially someone with another S60 and someone with an S85) that would be most appreciated.

DistanceOverChargeBrian.png
 
Yup thats correct, just one dataset with the SOC measured every 10 minutes for about 13 days. Its a fairly clean dataset though, since the environment was fairly constant and the rate of drain was also fairly constant
 
That is an interesting dataset. Do you by chance have a plot of the same but during driving rather than sitting? It would be interesting to know if they deal with vampire draw alone differently than driving, or if your car is different than the others.

Peter



My MS60 was sitting in the shop for several weeks waiting repairs, so I plotted up the rated range as a function of battery capacity as the vampires drank the juice. As you can see (if the image upload worked), the rated range definitely goes to zero when the battery reaches 10% capacity.
Jaggies on the line are the 1% charge increments.
 
That is an interesting dataset. Do you by chance have a plot of the same but during driving rather than sitting? It would be interesting to know if they deal with vampire draw alone differently than driving, or if your car is different than the others.
Think about it. He has ONE dataset, not many many data sets like Brian or myself. So it is quite easily explained why his data shows no overlap. The overlap comes from "today 200 mile range corresponds to 80% SOC - yesterday it corresponded to 81% SOC". It's the fact that Brian and I have lots and lots of datasets going from charged to less charged and back up again that causes the overlaps.

Makes sense?
 
It's not the lack of overlap that I find interesting, but rather it's a full dataset from a car just sitting and slowly discharging due to vampire draw. The data shows a "reserve" of 10% when the battery is at 0 range left that is why I'm interested in a driving dataset.


Think about it. He has ONE dataset, not many many data sets like Brian or myself. So it is quite easily explained why his data shows no overlap. The overlap comes from "today 200 mile range corresponds to 80% SOC - yesterday it corresponded to 81% SOC". It's the fact that Brian and I have lots and lots of datasets going from charged to less charged and back up again that causes the overlaps.

Makes sense?
 
I would suppose that for your particular car, from looking at your numbers, that your "rated mile" energy unit is ~287Wh/mi, and your reserve is built up at ~20Wh/mi. It would be interesting to know if that matches your own "rated mile" energy unit calculations done by the car for projected mileage.
Staring at my screen today (hey, that makes for really safe driving...) I can tell you for certain that the Energy App assumes 298Wh == 1 rated mile. Looking at the collected telemetry data, I can see that for roughly every 267Wh my rated range decreases by one. So 31Wh per mile go to the troll? That seems a lot higher than what people in 85s are reporting (usually around 20/21Wh). Very odd.

- - - Updated - - -

Oh, and that of course brings up another fun fact: 209 rated miles (what the display shows me right after finishing a max rate charge) times 298Wh (what the energy app says is my rated mile equivalent) multiplies out to 62.3kWh.
Those same 209 rated miles multiplied by the 267Wh that are used to decrease the rated miles gets us to 55.8kWh - which sounds about right for a 60kWh car with a 4.2kWh anti-bricking reserve - or a 4.5 miles (@267Wh/mile) driving reserve (that's in line with what I hear from people who actually ran their car until it stopped moving) and the oft quoted ~3kWh (i.e., 5%) anti-bricking reserve.
But mind you, it is utterly impossible to drive @267Wh without serious, serious hypermiling. I have been driving ridiculously conservatively over the last three commutes to work and can make 285Wh/mile. But to go down to 267Wh/mile I need to limit my speed to about 45.

So there are even more lies hidden in here somewhere...
 
Last edited:
I have a bunch of data from the car while driving, (about 3 months worth), so I will take a look at it when I get the chance. The nice thing about the vampire drain is that the battery temperature was (presumably) fairly constant during this time (around 25C). When driving the battery temperature is a pretty big unknown variable, and we know that this temperature figures into the estimated range, and could possibly account for most of the variation that you have seen.
 
Reviving a little old thread. But I have been noticing these issues as well. I was wondering if there was any conclusion on any of this. I also am only seeing ~224 miles on a 90% charge every morning. These two things combined have me a little concerned about getting to SC for Thanksgiving this year. I drive a S85.

I left for work today with 224 miles 'rated range' on the dash.
I drove almost exactly 30 miles to work and ended up with 194 miles 'rated range' on the dash.
The REST API is reporting 194.3 'rated range' (via AHimberg's WindowsPhone App).
My display on the dash also says 266 Wh/mi.

My 17" display says I have 223 miles 'projected range'.
My 17" display says I have 264 Wh/mi over the last 30 miles.
It was a nice cool, if not a little chilly, 60F this morning on my drive to work.

When I was driving to work I was generally between 260 and 280 Wh/mi the whole second half of the trip. And my rated miles were dropping pretty much in lockstep with my actual miles driven. They were never more than one mile off.

I have read this thread again today. It seems that I should be banking (versus actual miles driven) rated range, with a 266 Wh/mi energy consumption.


My concerns are two fold. One I was expecting that at such low power consumption that I could beat rated range (if only barely). And that my rated range is significantly reduced from the ~238 it should be at (90% of 265) on a daily basis. And that my one way trip is 320 miles to where I need to go in SC (I plan on charging about 220 miles into the trip, and have another closer charge point if necessary).

Now this being said about a month ago I drove 222 miles, using 65.4kWh, averaging 295Wh/mi. I had 29 'rated range' remaining at the end of the day. On that trip I dropped an additional 10 miles of 'rated range' in the first 7 of driving (lost 17 miles total, my average at that point was ~400 Wh/mi though). And I lost another 8 parked (only for about 2 hours) at destinations. And I was able to recover some while driving conservatively at stretches. I don't remember what my starting 'rated range' was. I think it was ~262 miles. It was a good 95F this day. I didn't have any worries (except after parking and seeing the 8 mile drop, I went easy for about 20 miles before I kicked my driving back up to ~70mph).

So my questions:

Does the top of the battery read 'rated range' loss faster than the bottom?
Is this inline with some of the behavior that you are seeing?
Is it reasonable to expect higher than initial 'rated range', if you are able to drive in the 260 Wh/mi range?
Based on my long trip it looks as if I could use about 75kWh, is it reasonable to expect that I could drive 320 miles on that if I was averaging 231 Wh/mi?

All of this would go away if there was a Supercharger in Augusta (are you listening Tesla?). I could care less as long as I could go ~250 miles (on a 100% charge) without much thought (or significantly low highway speeds).

EDIT: I am aware of this thread: http://www.teslamotorsclub.com/showthread.php/18189-You-might-consider-charging-slower so I will try to get a cosmetically larger number, for comfort before I embark on my trip to SC for Thanksgiving.
 
I too see the non-linearity of the rated range loss...with the majority of it happening at the very beginning of a trip. I assume this is a byproduct of the calculations initially including the high "startup energy" we all see consumed when you first start a trip after charging.

Your experience in beating the assumed W/mi numbers (283 or 308) and yet not "banking miles" is interesting, however...
 
Last edited:
Your experience in beating the assumed W/mi numbers (283 or 308) and yet not "banking miles" is interesting, however...

It is. I don't know if the software doesn't 'bank' miles until say after 60% charge or something. My work commutes give me really weird numbers. I have never been able to bank more than 2 miles going to work (and I only seem to see a massive drop upon starting in 'rated range' when range charged), And have twice made it with ~250 Wh/mi.

I feel very comfortable getting 250miles out of a range charge (even driving more than 55). As I have done a few trips where I was cruising 65-70 and gone over 200 miles, with enough rated range left over.

But on my morning commute I have tried and tried to "bank miles" on the 'rated range' display without any luck. And parking at work the vampire gets me, and I don't think the combined trip is usable as data.
 
It is. I don't know if the software doesn't 'bank' miles until say after 60% charge or something. My work commutes give me really weird numbers. I have never been able to bank more than 2 miles going to work (and I only seem to see a massive drop upon starting in 'rated range' when range charged), And have twice made it with ~250 Wh/mi.

I feel very comfortable getting 250miles out of a range charge (even driving more than 55). As I have done a few trips where I was cruising 65-70 and gone over 200 miles, with enough rated range left over.

But on my morning commute I have tried and tried to "bank miles" on the 'rated range' display without any luck. And parking at work the vampire gets me, and I don't think the combined trip is usable as data.

Using the tools we have for analysis of the API data you should be able to analyze this in pretty solid detail.