TimothyHW3
Active Member
Do you have SR+ or without plus? The SR+ should yield 49-50kWh, if it is SR only that will make more sense. Once I know that I will explain why there is a difference between the charge and discharging rates.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
V10 and Battery "Degradation"
- Thread starter Thunder7ga
- Start date
AlanSubie4Life
Efficiency Obsessed Member
While knowing the actual battery capacity is interesting, in the simplest distillation, as already discussed, you just need to know two key things:
1) Your rated miles will decrease by one (1) count for every 209Wh used. (SR/SR+ only)
2) If your projected 100% charge shows fewer rated miles, that means you have less available energy for use at a given charge percentage.
It really is that simple, in general. Of course there are special cases of battery malfunction, very cold battery, etc., but that is not the norm.
Last edited:
TimothyHW3
Active Member
In that case you should get close to 49kWh usable.
Can you turn off dark mode and try to drive around the typical line and match it and post a sharper picture to see what your typical constant is? And maybe on your next 100-10% run, try keeping the car around the straight typical line and see how many kWh used(since last charged shows that in your trip meter) you can pull. Should be 44-45kWh.
It's a bumpy road, so these are the best I can do with a phone mount, one picture I had to heavily crop, so it's not that high resolution.
When the my avg line matches exactly with the rated line, appears to be at 225wh/mi, so AlanSubie4Life was right:
You can click for full picture resolution.
Note that rated range does not match projected range at 225wh/mi. 169mi projected at 225wh/mi, vs. 173 mi rated range.
But rated range does seem to match the graph's projected range when at 219wh/mi or 220wh/mi:
But then if you reverse calculate my starting range based on trip consumption and distance driven, the 209wh/mi constant proves correct. So the only thing that makes sense is the graph is just wrong period. It might be projecting 174 mi range, but it is not with 220wh/mi.
-Jimmy
AlanSubie4Life
Efficiency Obsessed Member
Thanks for the follow-up. No surprises. The only piece of information to clarify here is that the "avg" line on the graph IS consistent with the trip meter (that's where it is coming from). If you had reset the trip meter and showed a 5-mile interval (rather than 5.8 miles), it would show 220Wh/mi for that interval (matching the graph avg over 5 miles). (This is easy enough to verify in the future, but it is not necessary to do so - take my word for it.)
The point is that there is no scaling (by 219/209 or whatever) of that avg line or anything like that, to explain the incorrect projected range.
Correct, I'm not doubting that 220wh/mi was my actual real consumption as calculated using the trip numbers, I've seen those numbers agree at 15 mi and 30 mi intervals. Where my confusion lies, is that it must be true that the projected range number is in fact not using the avg consumption rate calculated from the trip consumption.
So in that picture of 174 mi projected range, it is not using the 220wh/mi avg real consumption, it is using the rated consumption constant, since 174 mi projected was equal to 174 mi rated range left. Rated range uses 209wh/mi.
So is the projected range number on the graph just equal to: (battery capacity) / (avg consumption - 11) ?
There's some arbitrary offset used for reasons that escape me?
AlanSubie4Life
Efficiency Obsessed Member
It's not an offset, it's a scaling. Very different. Projection will be very optimistic for a very full battery, and nearly correct for a nearly empty battery.
But yes, the reasons escape me as well.
No. That formula is not correct.
Projected range is: (Battery Trip kWh * (219Wh Wh Charge / 209 Wh Trip )) / Trip Meter Average Consumption (Wh Trip)
Where "Battery Trip kWh" is calculated using the 209Wh/rmi constant.
(You can see the units don't cancel in the formula above - that's why the numerical result is incorrect.)
TimothyHW3
Active Member
Your typical is at 219Wh/km. This is the number you Tesla uses to calculate the rated range inside the car. This number never changes, unless Tesla decides to tweak it like they did with the LR RWD or I guess with the "new" 250 miles SR+(we shall see).
The formula is full pack capacity in kWh/ typical = rated range.
If we reverse that formula and use the rated at 240 and take typical 219 with rounding errors we get about 52.5kWh. But we know that Tesla pushes a little more kWh to offset the degradation in the first months, so it probably had about 53-54kWh.
But!
Your 0-100% is being calculated by substraction of a buffer inside the pack. In an LR it is about 3.5kWh and I assume it is at least 3kWh on SR+. S
that makes your available 0-100% roughly 50kWh and NOT 53.
If we take the same typical of 219, then you can only drive about 228 real miles until you hit 0%. You will still have your buffer below 0% for another 10 miles or so, but it will not be counting in the 0-100%.
Hence the formula Allan is suggesting.
But please barein mind that with degradation that buffer will get smaller and your capacity also, so the buffer will be, relative to the capacity, higher, hence less real % left. Therefore that 209Wh he mentions will need to be even lower once degredation kicks in, if you wanna drive 240miles... The number is not a constant, this is why the formula is never accurate.
The rated range and typical consumption is and this is why with degradation your rated miles will go perfectly in sync with that algorithm.
I explain all of these findings, regarding the buffer that Tesla uses to mislead the rated range, in this video.
AlanSubie4Life
Efficiency Obsessed Member
On the contrary, I have found this formula to always produce (for my AWD) 230Wh/rmi for the discharge constant, even as my full capacity has gone from 310 rated miles to ~305 rated miles. It is very, very accurate (I usually see within 0.3% or so for a valid test).
I have also measured it on a friend's car, who has 289 rated miles at 100% - and it also produces 230Wh/rmi.
I'm willing to accept that the constant COULD change, but we would need data to prove that, from a car with significant "degradation." So far - see the example here - the available data clearly suggest it does not change when capacity is reduced.
Take @earthwormjim for example. His full range is currently 226 rated miles (5.8% degradation from 240 rated miles). If the constant changed with loss of capacity ("degradation"), then we would expect a change in my 209Wh/rmi predicted value (to something like 200Wh/rmi???...not clear how you think it would change...). But this value has been verified by users who have 240 rated miles at 100% (see my post from Sunday to a new user) - and @earthwormjim himself has verified it is STILL 209Wh/rmi in his car, with 226 rated miles.
Last edited:
Just to clarify some things with regards to the “charge” and “discharge” constants being discussed here based on observations of my own data and other data I have looked at related to this.
The charge constant is truly an actual constant. That is, it is a fixed number in the firmware that Tesla uses to convert from energy capacity to rated miles. It is a very precise number that doesn’t change unless Tesla changes the firmware value. So, knowing the full pack value in kWh, you can always use the constant to calculate exactly how many rated miles your dash will display at any percent SOC. Or, going the other direction, you can take rated miles and percent SOC, and determine your full pack capacity in kWh.
The discharge constant on the other hand, is not truly a constant, but more of a performance number for the vehicle that from a first order can be considered a constant. It is related to the charge constant in that it is usually about 94% to 95% of the value of the charge constant. But it can vary some; for instance, if driving very aggressively or hauling a heavy load, the internal losses in the battery will increase, causing the discharge value to drop.
The discharge constant then is not an exact value and is not a value used in the firmware for any calculations.
Hope this didn't muddy the waters too much.
AlanSubie4Life
Efficiency Obsessed Member
I'm willing to concede that for huge values of consumption (say 1000Wh/mi, or maybe even a bit less) it may deviate. However, over a reasonably broad range of useful efficiencies, it is quite constant. See here my data from a recent trip - spans a range of efficiencies from 225Wh/mi to 344Wh/mi for my AWD (and that produces 232Wh/rmi to 228Wh/rmi, which may just be partially rounding error, because when you look at subsequent segments and overall averages it works out very close to 230Wh/rmi - outbound average of 332Wh/mi produced 230.2Wh/rmi and return at 251Wh/mi produced 230.0Wh/rmi ).
The problem with the discharge constant I believe is primarily that it is difficult to measure properly - battery temperature can throw it off, recent regen can throw it off (a LOT, because the added energy is not immediately displayed), etc. - a change in the energy truly available from the battery should introduce error in the calculation. Whereas the charging constant is simply effectively displayed on the charging screen directly (if you swap between energy and distance display).
Last edited:
TimothyHW3
Active Member
We have that data. Of course it will be round or about 230, but it will change and does change and this is why we can't call it a "constant".
I don't have your exact numbers as this is hear say from your friend, but let us use this example:
At 289miles or about 465km again rounding errors, the full pack is roughly 71.145kWh(again rounding errors)
At this degradation we observe a buffer of roughly 3.3kWh( see Bjørns Battery data)
This means the real consumption to drive the 465km is (71.145-3.3/4.65)*1.6 (1.6 is to get into miles )= 233Wh/km
If we take your battery, or 488km we get 74.564 with 3.5or 3.4 buffer. Or (71/4.88)*1.6 which is 227Wh/km
If we take a bigger degradation down the road and the buffer stays roughly the same at 3kWh.
Let's say 10% degradation of 69kWh full, will yield 451 rated km so 69kWh-3kWh )/4.51km *1.6 you see how that number is now closer to 234Wh/m.
So the number moves.
This is why you can't say this is a constant and this is because the buffer is not resizing in % proportional to the battery degredation.
Of course the difference is small and at 5 miles degradation you can't even measure it, but the math is clear.
But the typical consumption of 153Wh/km or 245Wh/m is a constant unless Tesla update it.
Precisely!
AlanSubie4Life
Efficiency Obsessed Member
Not hearsay - rode along with him and measured myself over the summer.
I have never measured anything close to this.
Have not got this result either.
There are some issues with decimal places here, but in any case I have not been able to obtain either of these numbers.
This seems like a good opportunity for people to measure it themselves (carefully!) on a degraded battery.
My money is on the constant not changing. I don’t see why it would need to. I don’t believe they need to change the constant. It makes a lot more sense to subtract the buffer (whatever it is), then convert the remaining energy in the degraded battery to a certain number of rated miles using a constant. That way the behavior of the rated miles is consistent.
But hard data will tell us the answer. I have presented mine, and have not seen any contradicting data so far. I have an open mind to good data, proving how it works.
The "charge constant" will not change unless Tesla changes it. But there is no equivalent "discharge constant".
To be more precise, they should probably be called "charge constant", and "discharge rate".
AlanSubie4Life
Efficiency Obsessed Member
We can call it that if you want, but the way I see it (so far) is it is an amazingly consistent predictor. Happy to see examples of contradictory information. Have not seen any yet! Would be interesting to see data from someone who tracks the car, for example - that would be an extreme example where I would not be surprised if the scalar did not hold.
It is perfectly consistent with Tesla’s philosophy for each rated mile to correspond to a certain fixed amount of energy - and I have not yet seen evidence to the contrary, or that the amount of energy per rated mile for a given vehicle changes over time.
Last edited:
n2mb_racing
Active Member
Sorry, I haven't been following the details. What software version was the new buffer introduced? And is it a buffer at the top of the battery or the bottom? I'm trying to decide if I should update...
TimothyHW3
Active Member
There is no such update - the buffer remained the same after updates. At least the buffer the car knows of, BMS, there is possibly a buffer not allocated to the car's BMS and is hidden. But the actual 3.5kWh visible buffer below 0% remained the same. It only changes with battery degradation and is probably an arbitrary formula baked into the firmware. The buffer is pretty low as it is, I doubt Tesla will touch it in any way. I think the lowest I have seen it is 3.3kWh after 30,000miles.
What they can do is lower the topic Voltage from 4.2V to 4.1V which will give less capacity, but it hasn't happened yet on a 3.
n2mb_racing
Active Member
OK, cool. So, nothing has come of this thread then? No one has confirmed a change?
V10 Loss of Range: 5% Range Loss in 2019 Model 3 After Install of V10 Update
Similar threads
