-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Need to find time to summarize some data on the allegedly "dead" horse, but here are my stats:
2013 P85+ 100% range when new - 272 miles. 272 miles x 0.295 kW/mile = 80.24kWh usable, 84.24kWh total including buffer. Where are the missing kWh?
Now, after 4.5+ years and 86,890 miles the 100% range is 253-254 miles (74.6 - 74.9 kWh usable, 78.6 - 78.9kWh total including buffer)
2013 P85+ 100% range when new - 272 miles. 272 miles x 0.295 kW/mile = 80.24kWh usable, 84.24kWh total including buffer. Where are the missing kWh?
Now, after 4.5+ years and 86,890 miles the 100% range is 253-254 miles (74.6 - 74.9 kWh usable, 78.6 - 78.9kWh total including buffer)
As has been mentioned probably 50 times in this very thread, 100% charge when new means nothing. The BMS isn't even remotely calibrated at that point and it has no idea what the real capacity of the pack is. I guarantee had you attempted to run down to 0% on the very first charge (or even first few cycles) you would have been in for a bad day.
And if we're splitting hairs, even if by some miracle you got one of the best 85 packs ever seen, and somehow your BMS was spot-on from the factory without ever cycling the pack (we'll ignore the fact that this is impossible for the sake of argument).... I still kind of have to point out that you answered your own question: 84.24 is in fact still less than 85.
(Queue a round of rounding up defenses)
As has been mentioned probably 50 times in this very thread, 100% charge when new means nothing. The BMS isn't even remotely calibrated at that point and it has no idea what the real capacity of the pack is. I guarantee had you attempted to run down to 0% on the very first charge (or even first few cycles) you would have been in for a bad day.
And if we're splitting hairs, even if by some miracle you got one of the best 85 packs ever seen, and somehow your BMS was spot-on from the factory without ever cycling the pack (we'll ignore the fact that this is impossible for the sake of argument).... I still kind of have to point out that you answered your own question: 84.24 is in fact still less than 85.
(Queue a round of rounding up defenses)
So, where is the “meaningless” 272 mile number coming from? Is it random?
Interesting information in this post. Not sure what would be deemed normal or acceptable in degradation / loss of milage capacity. I have S85 and just turned 107,000 miles. I believe the total range charge was 265. I have seen that drop over time and now fully charged I top out at 248 miles / 6.4% degradation.
That does not answer my question. Is it a random number? Why Tesla picked that number?
Another question is what's required for this number to be re-calibrated to a number you would trust?
It doesn't matter why. They could initially program it to be a 2 kWh pack or 100 petawatt-hour battery pack from the factory and it'd make basically no difference to what the actual usable capacity of the pack is. I'm not going to waste time speculating as to why Tesla decided to pick some initial value for the BMS capacity that is above the actual capacity (but still below advertised capacity, funny enough).
And "trust" is simply irrelevant here. Completely, totally, and utterly irrelevant. Trust has zero to do with when the system actually has enough data to make a reasonably accurate capacity determination. It is going to vary considerably based on actual usage, and doesn't change the fact that the initial and marketed values are completely bogus.
And now I'm remembering why I unsubscribed from this thread originally. Pointless questions that just attempt to distract people from the truth of the situation.
It very much does matter. It demonstrates that your story does not add up. And when this happens, the posts, as usually, descend in rudeness and condescension.
So here are the questions that you do not care to answer so far.
There is no “missing” kWh in the 85kWh battery, as multiple data points indicate that usable capacity is 80 – 81kWh, and total capacity including brick protection is 84 – 85kWh. Nothing is missing
So here are the questions that you do not care to answer so far.
- You have fleet wide Tesla data on battery capacity which you quote, but never answer the question **which** of the tesla energy capacity parameters you've used - there is more than one. I have no idea why you chose not to answer this question, but optics of it does not help you case of "stolen" capacity.
- You dismiss the 272 mile range number, do not believe it has any basis whatsoever, and do not want to answer question or speculate of where this number comes from, because it does not fit your story. You just prefer to dismiss it as Tesla is full of it…
- Then there is the story of changing capacity and different numbers of “missing” kWh. Initially it was 4.3kWh (OP in this thread) but now it is 2.38 kWh (85 kWh - 82.62 kWh = 2.38 kWh). You’ve dismissed it as having 2% margin of error for your original claim, but it is 45% reduction in “missing” capacity. You never explained why the difference if original kWh were “proven” by you several different ways, including your own bench testing.
There is no “missing” kWh in the 85kWh battery, as multiple data points indicate that usable capacity is 80 – 81kWh, and total capacity including brick protection is 84 – 85kWh. Nothing is missing
Umm... no it doesn't? How so? Care to explain? No, you're just throwing nonsense around.
I don't hold punches. When someone spews nonsense, I call it what it is. Can't take it? I suggest you play elsewhere.
Why I'm going to bother wasting time explaining why you're full of crap here yet again is beyond me, but here it goes.
No, there isn't. The diag_vitals data that Tesla logs only reports one capacity value, and I noted that the values I reported were usable energy as reported by the BMS. I suggest you go back and read those posts before making baseless claims.
No, I fully explained that the number is the result of a pre-calibration value programmed into the BMS at the factory. This will yet again be me saying the same thing. "Hello! Anybody home? Think McFly! Think!" You want to keep hammering at this because it's basically the only thing that would seem to support your position, yet doesn't stand up to any scrutiny whatsoever. You're the one who prefers to harp on it because the truth doesn't fit your story. And just to be clear, I tell no "stories", either. Everything I write is backed up by the facts and the numbers. If I'm speculating, I specifically say so.
Of course I dismiss it, and have explained exactly why. In this case, you won't even explain why such speculation is worthwhile. Why? Why does it matter why some engineer put a starting value in for capacity calibration that's never correct? I'm pretty sure the speculation wouldn't support your case at all anyway, so let's waste everyone's time on it for a moment.
<speculation>
Hmm, so let's pretend I'm an engineer at Tesla. I'm working on the BMS software. I need a way to get the calculated capacity as close to the actual value in as few cycles as possible. I know that the software is designed to downwardly adjust the available capacity estimate much more quickly towards an accurate value since capacity loss is a normal aspect of the battery pack and magical capacity increase is not. Capacity can be lost due to imbalances, degradation, etc. But all of the battery packs are different. I know that 265 miles of rated range is the target. (RWD variants, since this is engineering that occurred at that point.) 265 miles is 78.175 kWh of capacity, plus the 4 kWh of bottom buffer is a target of 82.175 kWh. Ok, good. So we'll set the initial calibration capacity to some arbitrary amount above that... say, 2.5%? Then the BMS will quickly calibrate downward over the first couple of cycles to towards the correct capacity. And it's pretty unlikely that anyone will drive the car to 0% on the first charge or two. After a few cycles everyone will settle to around 265 miles, and everyone will be happy. Win win!
</speculation>
<speculation>
Marketing Department: We need it to look like these have 85 kWh
Engineering Department: But it will just correct itself anyway.
Marketing Department: Just do it.
(Engineer still puts in 84.6 kWh)
</speculation>
<speculation>
Engineer 1: These are the specifications targeted for the battery pack.
Engineer 2: Cool, but the batteries don't actually have that much capacity.
Engineer 1: Well, our software will correct it quickly, so doesn't matter and it keeps marketing off our backs.
Engineer 2: Good point. git commit (or actually svn, since Tesla used svn in the early days)
</speculation>
<speculation>
Panasonic: Our batteries have this much capacity
Engineer: Great, we'll program the BMS with this
* Panasonic walks away with sack of cash
</speculation>
Obviously the speculation can go on and on... but for what? What good does this do? There is no scenario that matters nor makes the initial calibration value relevant because it doesn't match any real world value.
So again, as much as you might want to scream that the reason they chose the number matters... it just doesn't. The engineers put a number in that works for the project and that's basically the end of the story. It has no bearing on actual capacity.
lol. Again, actually read my posts if this matters so much to you. I'm not going to hold your hand through that by copy/pasting things for you. I clearly explained where the initial data came from and my methodology. I clearly explained where the updated data came from (containing multiple orders of magnitude more data points). Obviously even if I went and bought multiple brand new 85 kWh cars from Tesla, towed them from the factory myself with the HV systems disabled, then dismantled the packs for testing..... I still wouldn't end up with as much data as Tesla themselves have in the database that I was provided by an insider. Why is this even a question?
So, let me get this straight. Your argument that there is no "missing" kWh is based on a couple of things.
First, you claim total capacity is 84-85 kWh. Let's start with that. I'll reiterate that I have zero.... that's read z-e-r-o or "0"... data points from THE ENTIRE TESLA FLEET WORTH OF DATA that shows a single vehicle having a usable capacity of 81 kWh or more (total of 85 kWh or more). Zero. ZERO. ZERRRRRRRO.
Next, you cite "multiple data points" where usable capacity is between 80 and 81 kWh. Sure, I never said that wasn't the case. But you're taking some extreme edge cases as if they were normal. Yes, there are a handful (literally less than 200 or so out of something like 88,000) of "85" vehicles that have capacities that would almost not be criminal to round up to "85". As for basically everyone else... the other 99.9% of "85" vehicle owners.... yeah, sorry, you don't have anything close to 85 kWh of battery capacity and you never did.
And the EPA testing is just pointless. I've no idea how they come up with their numbers. They literally don't match real world data for any car. 95.507 kWh recharge from the wall event on an 85 kWh pack? That's only possible at like 5A on 120V or something ridiculously inefficient. Or with extreme pack heating needed. Or running HVAC at full blast while charging. Or any other number of reasons that make that number completely outrageous. That's a wall->wheels number of over 360 Wh/mi for a RWD 85's rated mileage.
FACT: I just drove my car down to 1 mile of range a few weeks ago and then charged to 100% from AC power. This is a newish good condition "90" pack that shows 83 kWh usable (87 kWh total capacity). The recharge event pulled 93.8 kWh from the wall. So, my "90" pack that has ~281 miles of rated RWD range at 100% charge pulls less power from the wall than the EPA calculated for an "85" with, for the sake of argument lets say it showed 272 miles of range. lol. Come on now really?
Finally, I'll apologize to everyone else here on this forum who benefits from and enjoys my technical posts and knowledge sharing. Suffice it to say @vgrinshpun has once again reminded me why I can't post such things here and why TMC still is still not the correct venue for technical discussions about anything that can possibly shine a bad light on TSLA. I simply don't have the time to bother with battling people who will defend Tesla to the death no matter what they do wrong, and this post is proof of that again. I'm 100% certain that there will be yet another response trying to take attention away from the facts, but I'm not going to bother anymore, so don't take my lack of further response here as lack of support for the actual facts. They still stand regardless of what spin people may try to put on them. (Statistics show that teen pregnancy rates drop dramatically after age 20!)
Until TMC decides to finally implement some form of self-moderation so that technical discussions can be kept clear of baseless and useless speculation and other pointless discussions that detract from the usefulness of the data and technical points themselves, or finds some better way to partition technical topics (there isn't one).... I'm going back to limiting my TMC use to the classifieds section and responding to reasonable PMs and highlights.
Last edited:
That sucks. I enjoy reading your posts!
thefortunes
Active Member
I have an S85 with 98k miles and I also top out at 248 miles. It has been +/- 1 since I bought it with 72k miles last year.
Please post more data if you can!
Are all 90 packs the same? Same chemistry? No other packs (75 or 100) have been using this chemistry?
My car is about a year old now, after obtaining cable and can-bus interface after a month or two after getting the car I have been following my own pack:
Pack kWh Kilometers Date
83,5 2728 15.feb
83,6 3081 16.feb
83,8 18 may
83,4 10513 15 june
83,4 5 july
83,4 17380 16 oct
83,1 20772 5 Dec
Screenshots, February:
The highest I have seen in May:
In October:
A few days ago (December):
Do you happen to know the layot of the pack and where the temperatures are measured?
Is it two temps per module (coolant temp in and out perhaps?)
This is from my car when it was charging at a SuC last February:
While driving in February:
While driving a few days ago (December):
The highest temp is always measured at position 17...
I'm not sure you can count the "as-new" range as reliable. I think at the point the BMS is calibrated and hits the EPA 265 (for a P85), then that is your number set in stone. Not sure how long that calibration takes though, a few full charge/discharge cycles maybe?
Misc. voltage readings or a "starting point" from Tesla? Unsure. I thought if you took the max capacity an 18650 could have voltage wise and multiply it out by cells then divide by EPA watts / mile, you get that magic 272 number. But still I know less what I'm talking about than most.
Please post more data if you can!
Are all 90 packs the same? Same chemistry? No other packs (75 or 100) have been using this chemistry?
My car is about a year old now, after obtaining cable and can-bus interface after a month or two after getting the car I have been following my own pack:
Pack kWh Kilometers Date
83,5 2728 15.feb
83,6 3081 16.feb
83,8 18 may
83,4 10513 15 june
83,4 5 july
83,4 17380 16 oct
83,1 20772 5 Dec
Screenshots, February:
View attachment 266063
The highest I have seen in May:
View attachment 266064
In October:
View attachment 266065
A few days ago (December):
View attachment 266066
Do you happen to know the layot of the pack and where the temperatures are measured?
Is it two temps per module (coolant temp in and out perhaps?)
This is from my car when it was charging at a SuC last February:
View attachment 266076
While driving in February:
View attachment 266077
While driving a few days ago (December):
View attachment 266078
The highest temp is always measured at position 17...
@bjornb - Great data. Is that the same spike that @islandbayy was seeing in his data? And do you have weird supercharging rates at times? As if the pack isn't charging as fast as it could or once did?
Edit: Just looked at both posted screenshots, the same spike at position 17. What does that mean, if anything?
I originally speculated that the spike was because of uneven heat/coolant distribution, but I dont know that for sure (insight from @wk057 is appreciated! ).
I have not experienced any problems regarding supercharging on my car. I dont supercharge that often, but I did see 110kW+ a few weeks ago.
My DC vs AC charge data:
I have also retrieved the supercharging sessions from my data at TeslaFi, and created this supercharging curve:
(The curves that start at lower power is because of shared supercharger and/or low battery temps).
My values also seems to correspond with the BTX4 curve posted by @blincoln here:
A Better Routeplanner
).I have not experienced any problems regarding supercharging on my car. I dont supercharge that often, but I did see 110kW+ a few weeks ago.
My DC vs AC charge data:
I have also retrieved the supercharging sessions from my data at TeslaFi, and created this supercharging curve:
(The curves that start at lower power is because of shared supercharger and/or low battery temps).
My values also seems to correspond with the BTX4 curve posted by @blincoln here:
A Better Routeplanner
@bjornb you should be able to determine the cause by looking at your cell voltage spread while at low state of charge or during supercharging. While driving, it is normal for the cell spread to widen but not while idle. If the cell voltage of one of your bricks is significantly lower and this corresponds to the module that is the warmest, then that would provide clear indication that you are experiencing what @wk057 has described here.
Similar threads
- Question
