-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
How many Kwh usable for driving are you getting on an 85 or 90 car? (I only get 75Kwh on my 90D)
- Thread starter marcmerlin
- Start date
-
- Tags
- Model S
No, I'm only using the trip meter to show Kwh used since last start. I do not use range for anything or any calculation.
Yes. 77kWh is well-known to be the real accessible capacity of an 85kWh pack.
The issue is that nobody seems to be able to beat that figure with a 90kWh pack!
Does anyone have a video, or photo, or anything showing a 90kWh pack as having provided more than 77kWh of energy during a full discharge?
Ok, the thread is a bit old, but I got some information from a Tesla Engineer after reporting the possible problem to my service advisor.
As a reminder, out of 4 longish trips, I got a calculated 75-76Kwh of fully usable battery by calculating energy used while driving (no stops) vs percent of battery lost during the drive.
Here is the summary of the discussion, and why one can't use the percent charge left number to extrapolate full battery capacity:
1) As you know there is an anti bricking buffer which is estimated from testing by end users to be around 4Kwh.
2) Then there was the other thing that people on an 85Kwh pack could only get around 77Kwh on a 100% to 0% drive, pointing to another missing 4Kwh.
This is because Tesla advertizes the marketing theorical capacity of the batteries in ideal discharge conditions. In real life due to internal resistance, you get less at higher draws, and Tesla also (rightfully) choose not to use the batteries from one extreme voltage range to another, keeping a bit of buffer on each side.
So that's why an 85 or 90 pack is really 81 or 86Kwh before you add the anti bricking buffer.
One could argue that Tesla could advertize the real usable number, but the engineering person rightfully pointed out that other car manufacturers do not give you true usable horsepower of your engine either, so fair enough.
Well, almost. This teardown/report shows that actually an 85 pack sems to really only be 80.7Kwh by testing and adding up the individual cells: Tesla's 85 kWh rating needs an asterisk (up to 81 kWh, with up to ~77 kWh usable)
So Tesla is probably still to blame for claiming that the base pack capacity is really more than it is, but that wasn't my main concern, I just wanted to make sure that my 90 pack had more energy than an 85 pack, even if both packs fall short of their marketing rating.
3) Now we're still left with my main issue, and the reason I contacted tesla: why was I calculating 76Kwh of usable battery for driving when I should be getting 82Kwh or so.
The engineer explained that the BMS basically gets the percentage number wrong when you're not discharging from full, and keep charging/discharging from the middle range of the battery.
It's true that you cannot really estimate a battery capacity from its voltage unless it's empty or full, so the computer keeps track of how much it pushed into it, and how much it took out.
That method gets more and more inaccurate over time and basically he pointed out that when I thought I had 95% at the supercharger, maybe I only really had 90%, and similarly when the car was showing 4%, maybe I still had 8%. There is no good way to really know because the percent number is an estimate that can become more off over time (actually a recent software update is supposed to make it more accurate).
If you add those errors up that would then mean my extrapolation to 100% battery, was wrong, and that the only way to know how many driveable Kwhy you get from 100% is really to drive from full to empty.
Also, just to make things a bit more complicated, when you charge to 100% at a supercharger, it doesn't have the time to properly balance the cells (i.e. you charge them in series and at the end of the charge, you actually balance the cells properly and that takes a fair amount of time).
So the only way to really have 100% is to charge to 100% at home overnight, where you get a full balance and get a chance to the battery computer to reset its counters for 100% charge.
Another way to explain how the error adds up: in a small plane there is no real usable precise gas gauge. You know how much fuel you have when it's full or empty. So you fill it up and then have a fuel totalizer that measures how much fuel came out. You then measure how much fuel you put back in if you don't fill up, and put that in a computer to keep track of partial fuel load.
As you can guess, if you do this 10 times in a row, you are going to have more compounding errors, and the amount of fuel shown by the computer will get more and more wrong. This only gets fixed if you empty the tank fully, or fill it up fully.
It's not entirely the same for batteries in the tesla, but it's the same overall general concept, the errors add up until you empty fully or charge and you cannot use my method of saying 80% of battery used with xkWh used means 100% is yKwh.
Hope this helps others.
As a reminder, out of 4 longish trips, I got a calculated 75-76Kwh of fully usable battery by calculating energy used while driving (no stops) vs percent of battery lost during the drive.
Here is the summary of the discussion, and why one can't use the percent charge left number to extrapolate full battery capacity:
1) As you know there is an anti bricking buffer which is estimated from testing by end users to be around 4Kwh.
2) Then there was the other thing that people on an 85Kwh pack could only get around 77Kwh on a 100% to 0% drive, pointing to another missing 4Kwh.
This is because Tesla advertizes the marketing theorical capacity of the batteries in ideal discharge conditions. In real life due to internal resistance, you get less at higher draws, and Tesla also (rightfully) choose not to use the batteries from one extreme voltage range to another, keeping a bit of buffer on each side.
So that's why an 85 or 90 pack is really 81 or 86Kwh before you add the anti bricking buffer.
One could argue that Tesla could advertize the real usable number, but the engineering person rightfully pointed out that other car manufacturers do not give you true usable horsepower of your engine either, so fair enough.
Well, almost. This teardown/report shows that actually an 85 pack sems to really only be 80.7Kwh by testing and adding up the individual cells: Tesla's 85 kWh rating needs an asterisk (up to 81 kWh, with up to ~77 kWh usable)
So Tesla is probably still to blame for claiming that the base pack capacity is really more than it is, but that wasn't my main concern, I just wanted to make sure that my 90 pack had more energy than an 85 pack, even if both packs fall short of their marketing rating.
3) Now we're still left with my main issue, and the reason I contacted tesla: why was I calculating 76Kwh of usable battery for driving when I should be getting 82Kwh or so.
The engineer explained that the BMS basically gets the percentage number wrong when you're not discharging from full, and keep charging/discharging from the middle range of the battery.
It's true that you cannot really estimate a battery capacity from its voltage unless it's empty or full, so the computer keeps track of how much it pushed into it, and how much it took out.
That method gets more and more inaccurate over time and basically he pointed out that when I thought I had 95% at the supercharger, maybe I only really had 90%, and similarly when the car was showing 4%, maybe I still had 8%. There is no good way to really know because the percent number is an estimate that can become more off over time (actually a recent software update is supposed to make it more accurate).
If you add those errors up that would then mean my extrapolation to 100% battery, was wrong, and that the only way to know how many driveable Kwhy you get from 100% is really to drive from full to empty.
Also, just to make things a bit more complicated, when you charge to 100% at a supercharger, it doesn't have the time to properly balance the cells (i.e. you charge them in series and at the end of the charge, you actually balance the cells properly and that takes a fair amount of time).
So the only way to really have 100% is to charge to 100% at home overnight, where you get a full balance and get a chance to the battery computer to reset its counters for 100% charge.
Another way to explain how the error adds up: in a small plane there is no real usable precise gas gauge. You know how much fuel you have when it's full or empty. So you fill it up and then have a fuel totalizer that measures how much fuel came out. You then measure how much fuel you put back in if you don't fill up, and put that in a computer to keep track of partial fuel load.
As you can guess, if you do this 10 times in a row, you are going to have more compounding errors, and the amount of fuel shown by the computer will get more and more wrong. This only gets fixed if you empty the tank fully, or fill it up fully.
It's not entirely the same for batteries in the tesla, but it's the same overall general concept, the errors add up until you empty fully or charge and you cannot use my method of saying 80% of battery used with xkWh used means 100% is yKwh.
Hope this helps others.
Last edited:
Johan
Ex got M3 in the divorce, waiting for EU Model Y!
What I can't wrap my head around though is how two cars, identical with identical software except for one having an "85 kWh" battery and one a "90 kWh" battery would estimate remaining or total capacity differently, relative to each other, so that the estimate on the "90" somehow ends up being the same or slightly lower than the "90"???
Two 90D cars can effectively give you a different battery charge percentage on the dashboard while having the actual same battery charge. It's because the percentage is software calculated from a bunch of approximations and can grow to be wrong over time (within bounds).
But the main point being that interpolating a portion of the battery discharge to 100% battery is a flawed method as a result of these approximations and the only way for me to really know my usable Kwh is to fill my car to 100% and drive it to 0%.
Well, actually if I fill it to 100% overnight and discharge it to let's say 20% I might get a slightly better number but the main point is that you need to start from 100% at least.
Johan
Ex got M3 in the divorce, waiting for EU Model Y!
I understand this but it has nothing to do with the fact that a large sample of "90" cars should show better total battery capacity than an large sample of "85" cars assuming the estimation/calculation methods are the same in both groups. The large sample should take care or variability within the groups.
I agree with you, and have no data that shows that a 90D car driving from 100% to 0% isn't getting 81 or 82Kwh.
Note, I'm not saying that the uncertainty in the percentage display on the dashboard proves that my battery has more driving capacity than an 85 car, I'm just saying that it's a valid reason that I can't claim my car only has 76Kwh of drivable battery capacity from the way I measured it.
There aren't that many 90D cars around. Did you find anyone who did a 100% to 0 ot 5% test?
Johan
Ex got M3 in the divorce, waiting for EU Model Y!
No, sorry I don't. But I would love to have more conclusive data on this. You're not the first one with a "90 kWh" to bring up this
Someone needs to do the test
I seldom do long drives and the one I did twice required that I charge in the middle (going up a mountain at the end), so it probably won't be me
Please post here if you find someone who has done a close enough test.
int32_t
Tesla Spotter
There is a LOT of magic to calculating a tidy percent to represent the amount of energy stored in a battery. The bottom line (all that really matters) is this: can your 90D go farther on a charge than an 85D?
Well, honesty matters too and it would be nice if Tesla would be a bit more honest with their battery numbers. But then if you don't take the ideal discharge number and paste it on the back of the car, what numbers do you pick? Real-world conditions are so variable.
Well, honesty matters too and it would be nice if Tesla would be a bit more honest with their battery numbers. But then if you don't take the ideal discharge number and paste it on the back of the car, what numbers do you pick? Real-world conditions are so variable.
Let's calm down
1) you don't have to figure it out. I was curious to know but let's be honest, you don't need to do this to drive your car
2) we already know that your 85Kwh pack is only about 81Kwh an my 90Kwh pack is also less than 90Kwh since tesla sadly does overstate the usable battery capacity (including the anti bricking buffer).
Tesla does guarantee the mileage under EPA testing conditions, so that is real, but the battery capacity is somewhat less than the Kwh number stated.
Brand new packs show around 84 kWh available according to the diagnostic screens, however some of this is quickly lost as inefficiencies build up. The cells are generally less efficient at delivering their rated capacity when the load is appreciably increased. Thus, cells that were rated at 10 mA draw will invariably "lose" capacity as the average draw is increased to 100 mA.
That's correct. That's what I was saying with internal resistance. You don't get full capacity from any battery if you're drawing a fair amount of current from them. So the way Tesla uses the cells, they are easily 4Kwh fewer than their marketing rating (and apparently, they may not quite have the right number so that the marketing rating really adds up to 85 or 90Kwh)
There seems to be little doubt in Australian samples that 90s get very close to the same capacity as 85s. This seems to invalidate the claimed 6% range gain and also price premium over 85 at the time. This appears to hold true with the same model specifications with similar mileage, settings and weather conditions. There are several online records being kept on these results, it would be great to see a definitive consolidated and statistical result on range differences between 85 and 90.
How do you know they are very close?
Did you drive 2 of them from full to close to empty side by side?
Or are there numbers from different people taken in different conditions?
Did anyone drive a 90D from full to near empty and record the kwh used by the trip meter?
RogerHScott
Active Member
The point of having a large numbers of samples of each (85 and 90) is these issues cease to matter, since they can be
expected to cancel out between the two sets of trials.
This is from many reports of the rated /typical range 85s vs 90s. They appear quite similar. I have also personally charged two 85 and 90 cars at the same time / conditions at same mileage. The difference was just a few kms.
I am happy to drive from full to close to empty . What would you say is an unacceptable number for a 90 considering 85 looks to be about 77?
Similar threads
