I'm showing the same. hovering at 465 to 471. 16500 km . Rarely charge beyond 85%, been to 100 twice . Below 30 once or twice . Mine is LR RWD.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
Range Loss Over Time, What Can Be Expected, Efficiency, How to Maintain Battery Health
- Thread starter KK_RedM3
- Start date
I'm showing the same. hovering at 465 to 471. 16500 km . Rarely charge beyond 85%, been to 100 twice . Below 30 once or twice . Mine is LR RWD.
Yeah, there seems to be no rhyme or reason to which batteries seem to degrade more or less. My battery has shown range estimates of between 307 to 313 miles, recently. Note the average is 310. I used to charge to 80% SOC, but the last few months, I've decided to experiment and charge to 58%. That means if I see 180 miles of range in the morning, it works out to 310miles rated. I don't really see any big difference to w
Those Teslafi charts always confuse me because the X-axis data doesn't use consistent increments.
Strange, that seems to be the opposite of what I see from Stats. Stats always reports higher SOC than the car or the Tesla app. In this case Stats shows 61% SOC, while the Tesla app shows 55%, as the battery is also showing a blue snowflake. When you divide the SOC into the Rated Range of 172 miles, 61% gives me 282 estimated miles, which is what the Battery Health chart will show. If you divide the 172 miles by 55% SOC, I get 313 estimated miles. Which one is right? When the battery is not showing a blue snowflake, my estimated Rated Range always shows between 307 miles and 313 miles. The 282miles which shows up in the Battery Health chart is spurious. That can't be the intent of the developer.
AlanSubie4Life
Efficiency Obsessed Member
That was an LR AWD as far as I can tell.
Why do you say this does not look like degradation, in your professional estimation?
In any case, thanks for the data. Not sure where your range was at exactly when you took the capture of your kWh remaining (and whether that 466km number was for a relatively low SoC which would introduce error), but the data lines up pretty well anyway (I assume you have a 2019 AWD):
245Wh/rmi * ~470rkm * 1mi/1.6093km = 71.6kWh
All in keeping with what we understand within 1% or so.
vsansal
Member
According to this VLOG there is an owner with 100K miles that is almost exclusively supercharging everyday and frequently to a 100 percent. His 100% is 297 on a LRRWD so down from 325 (or 310 from brand new). Also no reduction in supercharging speed apparently. There is a link to the original discussion on the video description but you need to be member. It is a facebook group.
Your example is exactly what I was referring to. The Stats app will show a lower range capacity(when cold)....because it doesn’t use the temp adjusted SOC. Developer seems that is more realistic. IMO it’s not because you gain back range as the battery warms up. You’re range of 307-313. That’s from Stats? Try only opening the app right after completion of charges...you’ll find that the data will be much more consistent with less variability. Unless you’re charging to 100%...everything else is a calculation. Depending where in the percentage point the calculation is done...can cause increased variability(on top of cold battery variability).
Stats for Tesla (iOS) - Tesla Owners Online
Last edited:
Are you sure of that? Maybe they just aren't aware of the discrepancy between the car reported value of SOC and their app.
I think it should match what the car reports.
From a member of the model 3 FB group:
This weekend, I attended a Tesla Owners Event, at which one of the presentations was on service and proper battery charging and maintenance. At this session, the two Tesla gentlemen acknowledged that where charging and proper battery maintenance there was a lot of misinformation on social media, and that this was not helped by materials on Tesla's site and even from Elon. He said that his information comes straight from the engineering team. The major points he discussed were:
1. Despite what you might have heard, the best thing to do for your battery on a regular basis is to charge up to 80% and not go below 20%. If you have to dip below 20% on occasion, that's OK, but charge up ASAP. If you are going on a trip, charge up as much as needed, but don't let the battery sit above 80% for long. Definitely don't let it sit above 95% for long at all - drive immediately when charging is complete.
2. Set percentage instead of range on the display. The range number is NOT set to EPA figures for the given battery percentage, but is calculated off the estimated charge of the pack AND your driving habits (See the discussion on the BMS below). Basically, a degradation of that number most likely does NOT indicate a battery problem. If you get in touch with service, they will use a sophisticated diagnostic to check the state of the pack and will most likely not see any major faults (as more than one person on FB has complained - "they aren't taking me seriously/they are hiding something/they don't want to fix it").
3. The BMS - battery management system. First, a little background is required:
You don't have one battery. Depending on the model you have, you can have upwards of 4000 individual battery cells all wired together in 3 or 4 "bricks", which combine to make one battery pack. The cells discharge at varying rates, from 4.2v (100%) down. However, when the car charges, it charges until the FIRST one of those thousands of cells reaches the charge limit percentage REGARDLESS of the charge level of the other cells. So if some batteries are 4.1v, 3.9v, etc, they will NOT get a full charge, and this will impact range.
If you are familiar with lithium ion battery packs, you may be familiar with the term "balance charging" - which is a way to ensure that the individual cells have the same voltage on them. These cells do NOT like to be overcharged and can catch fire - which is why the BMS shuts down when the first cell reaches capacity. So obviously you "lose range" because you cannot charge the other cells to capacity. Repeated short drives (and short recharges) don't give the BMS the data it needs to calculate battery health - both gentlemen stressed repeatedly that the algorithm used for the "miles remaining" figure gets more and more inaccurate over time, and that allowing the BMS to balance the cells "solves" most cases of "lost range". One gentleman stated that it's like filling a glass of water to a point and asking how much water is in the glass. You can't truly tell until you pour it into a measuring cup and measure it. It's the same with the BMS - if you use 10%, add 10%, use 5%, add 5%, etc the BMS is using the amount of current added during the charge cycle to decide what the capacity of the battery is. As mentioned in #3, you may not be adding a true 10% of capacity because one cell may not have discharged 10% - maybe it only discharged 9%, and therefore really only 9% is added in before charging stops. Those little discrepancies add up over time, and are rectified by balancing the cells:
a. Drive the car on a long trip - down to 20% is optimal.
b. Charge up to 90% and leave the charger connected overnight. The BMS will discharge the cells until they are all at the same voltage and then recharge them so that they are all the same voltage at a 90% overall state of charge. Then your remaining miles should be back to spec or thereabouts. The BMS will not balance the cells unless the battery is charged to at least 90%.
4. For optimal battery health, let the car sit for an hour before charging and for an hour after charging. They suggested using the scheduled charge feature for that. For some people like me who don't commute to work or have a set driving schedule, this won't work as well since the scheduling turns on the climate control for departure time and therefore wastes power if you aren't going anywhere.
5. To get a MUCH better estimate of your remaining miles, use the Energy app, set to Average at 30 miles. The algorithm there uses your driving style (average Wh/mi over the specified number of miles) and the percentage of battery remaining to perform its calculation. The algorithm on the main display is NOT as accurate according to them. They stressed that several times during both their presentation and the Q&A session.
6. Charging via a 110v circuit (for those of us who have it) is more efficient on the Model 3 than the Model S or X due to the improved charging circuit and battery technology.
(However, in my personal testing, it's about 6x slower than 240v/32a 14-50 charging, so I can only imagine how slow it is on an S or X....)
7. Supercharging is NOT good for the battery when done regularly. If you are relying on it for much/most of your charging, you WILL experience faster degradation of the pack. (NOTE: This *MAY* only apply to Model S/X and NOT the 3 per other information. The Tesla folks did not differentiate between models, so I can't say if this is true or not.)
8. ABC - Always Be Charging. The battery pack does much better when it is charged up after each use.
9. They are trying to get updated materials on the web site that reflect this information. Unfortunately, as you may have guessed, there are several higher priorities that the company is focusing on, so it will take time.
Hope this is helpful and explains some things.
This weekend, I attended a Tesla Owners Event, at which one of the presentations was on service and proper battery charging and maintenance. At this session, the two Tesla gentlemen acknowledged that where charging and proper battery maintenance there was a lot of misinformation on social media, and that this was not helped by materials on Tesla's site and even from Elon. He said that his information comes straight from the engineering team. The major points he discussed were:
1. Despite what you might have heard, the best thing to do for your battery on a regular basis is to charge up to 80% and not go below 20%. If you have to dip below 20% on occasion, that's OK, but charge up ASAP. If you are going on a trip, charge up as much as needed, but don't let the battery sit above 80% for long. Definitely don't let it sit above 95% for long at all - drive immediately when charging is complete.
2. Set percentage instead of range on the display. The range number is NOT set to EPA figures for the given battery percentage, but is calculated off the estimated charge of the pack AND your driving habits (See the discussion on the BMS below). Basically, a degradation of that number most likely does NOT indicate a battery problem. If you get in touch with service, they will use a sophisticated diagnostic to check the state of the pack and will most likely not see any major faults (as more than one person on FB has complained - "they aren't taking me seriously/they are hiding something/they don't want to fix it").
3. The BMS - battery management system. First, a little background is required:
You don't have one battery. Depending on the model you have, you can have upwards of 4000 individual battery cells all wired together in 3 or 4 "bricks", which combine to make one battery pack. The cells discharge at varying rates, from 4.2v (100%) down. However, when the car charges, it charges until the FIRST one of those thousands of cells reaches the charge limit percentage REGARDLESS of the charge level of the other cells. So if some batteries are 4.1v, 3.9v, etc, they will NOT get a full charge, and this will impact range.
If you are familiar with lithium ion battery packs, you may be familiar with the term "balance charging" - which is a way to ensure that the individual cells have the same voltage on them. These cells do NOT like to be overcharged and can catch fire - which is why the BMS shuts down when the first cell reaches capacity. So obviously you "lose range" because you cannot charge the other cells to capacity. Repeated short drives (and short recharges) don't give the BMS the data it needs to calculate battery health - both gentlemen stressed repeatedly that the algorithm used for the "miles remaining" figure gets more and more inaccurate over time, and that allowing the BMS to balance the cells "solves" most cases of "lost range". One gentleman stated that it's like filling a glass of water to a point and asking how much water is in the glass. You can't truly tell until you pour it into a measuring cup and measure it. It's the same with the BMS - if you use 10%, add 10%, use 5%, add 5%, etc the BMS is using the amount of current added during the charge cycle to decide what the capacity of the battery is. As mentioned in #3, you may not be adding a true 10% of capacity because one cell may not have discharged 10% - maybe it only discharged 9%, and therefore really only 9% is added in before charging stops. Those little discrepancies add up over time, and are rectified by balancing the cells:
a. Drive the car on a long trip - down to 20% is optimal.
b. Charge up to 90% and leave the charger connected overnight. The BMS will discharge the cells until they are all at the same voltage and then recharge them so that they are all the same voltage at a 90% overall state of charge. Then your remaining miles should be back to spec or thereabouts. The BMS will not balance the cells unless the battery is charged to at least 90%.
4. For optimal battery health, let the car sit for an hour before charging and for an hour after charging. They suggested using the scheduled charge feature for that. For some people like me who don't commute to work or have a set driving schedule, this won't work as well since the scheduling turns on the climate control for departure time and therefore wastes power if you aren't going anywhere.
5. To get a MUCH better estimate of your remaining miles, use the Energy app, set to Average at 30 miles. The algorithm there uses your driving style (average Wh/mi over the specified number of miles) and the percentage of battery remaining to perform its calculation. The algorithm on the main display is NOT as accurate according to them. They stressed that several times during both their presentation and the Q&A session.
6. Charging via a 110v circuit (for those of us who have it) is more efficient on the Model 3 than the Model S or X due to the improved charging circuit and battery technology.
(However, in my personal testing, it's about 6x slower than 240v/32a 14-50 charging, so I can only imagine how slow it is on an S or X....)
7. Supercharging is NOT good for the battery when done regularly. If you are relying on it for much/most of your charging, you WILL experience faster degradation of the pack. (NOTE: This *MAY* only apply to Model S/X and NOT the 3 per other information. The Tesla folks did not differentiate between models, so I can't say if this is true or not.)
8. ABC - Always Be Charging. The battery pack does much better when it is charged up after each use.
9. They are trying to get updated materials on the web site that reflect this information. Unfortunately, as you may have guessed, there are several higher priorities that the company is focusing on, so it will take time.
Hope this is helpful and explains some things.
AlanSubie4Life
Efficiency Obsessed Member
Did they provide any evidence of this? It’s explicitly on Tesla’s site that it is not dependent on driving habits, and oodles of data from SMT reads, etc. show that it is clearly directly related to how much energy the BMS thinks the pack contains.
That is not correct. To calculate the projected miles, it takes (for 2018 AWD) 245Wh/rmi, multiplies by rated miles remaining, and divides by recent efficiency. This can be verified yourself.
There really is no algorithm for the main display. It is simply a display of what the BMS thinks remains in the battery above the buffer.
Also not helped by the Tesla gentlemen. This is not the first time employees have not known what is up!
I’ve posted a lot here and with everything I have posted I have encouraged people to “trust but verify” with your own empirical tests using the proposed framework. I’ve also posted extensive data to support the theory. You can verify it, or get lost again in the weeds.
The only thing about rebalancing I will say is that most people who have SMT (ScanMyTesla) readbacks that show low kWh have very good cell balance. I have no idea whether this information about balancing is correct.
As far as degradation and it “not being real” is concerned, monitor the available energy per the trip meter on a friend’s car with good battery health and compare to your own (on a long continuous trip of course). Pretty sure it is mostly real, though of course some small amount of BMS confusion can presumably occur (though there is a pretty huge incentive for Tesla to continue to improve this as it can leave people stranded).
Battery degradation is real, normal, and not really an issue. If you expect 10% you’ll probably be in that ballpark after a couple years. If you’re unlucky you might have a bit more, if you are lucky you might have a bit less. But it is all fine; as long as you don’t lose more than about 15-20% your car is likely perfectly usable for most conceivable uses. Hopefully in the future batteries will degrade less but right now 10% is a state of the art result after a few hundred cycles.
Last edited:
AlanSubie4Life
Efficiency Obsessed Member
I’m all for not throwing out the baby with the bath water, but based on the other easily disproven “facts” provided by these folks, I do worry about the truth behind these other “facts” (which are much, much harder for an end user to determine).
Last edited:
I posted the link of the thread between myself and the developer. He’s well aware. As he states he uses “real” SOC and not adjusted..as seen in this post. Stats for Tesla (iOS) - Tesla Owners Online
I am not sure what he means by "real" SOC. There are 2 variables reported by the API related to SOC; "battery_level", and "usable_battery_level". One value is not anymore "real" than the other; they're both real, they just mean different things.
The car and the Tesla app both report the "usable_battery_level". This gives you more consistent numbers in your rated miles stats data because it doesn't calculate an artificially low number of full rated miles when the battery is cold. In other words, less noise in your data which makes it easier to analyze, and less confusing.
I think that is what most people would want to see, something that is consistent with what the car reports, and also gives the best statistical data.
I just don't see any advantage it gives the app user to be using the other variable "battery_level" in this case.
insaneoctane
Well-Known Member
While the weather in Southern California hasn't been getting significantly warmer (other than this weekend) over the last 30 days, my nominal full pack has been slowly increasing daily since my lowest reported 70.5. It's now at 73 kWh for the last 2 weeks. I'm guessing it's just the BMS recalibrating. It does make identification of degradation trends very difficult! I'll keep monitoring NFP vs extrapolated range and temperature and see where things go...
@AlanSubie4Life im just reposting. In terms of validity he did say Tesla personnel was there but I hear what you are saying that even they can be mistaken.
AlanSubie4Life
Efficiency Obsessed Member
Good to hear some recovery is possible. Your specific battery has always been a strange one though. I need to keep referring to it as “loss of capacity” rather than “degradation.” Makes hope spring eternal...
Same. M3 AWD, less than 15K kms and less than a year and I already see 6 / 7% capacity loss.
Charging pattern 30%->80% and 100% once per month (to rebalance). Very little SuC, car sleeps in garage.
It sucks.
This is uncannily similar to what one of our cars has experienced. About a 10 mile range loss in the first year and then all of a sudden another 10 Plus miles over a very short period of time. Tesla is useless on this stuff. They just tell us this is average loss. It's not.Maybe High average but still something's going on. One idiot in their service department insisted that this range loss was because our average Watt hours per mile was higher than the Benchmark 245 number. We have another car with just slightly less mileage that has shown really no loss at all. Tesla does not seem to have a true ability to parameterize these cases of high average loss. Which is okay if they would just admit that. Our experience is often times I come in and get treated as an idiot and I'm told things that are of course counterfactual and ignorant. It's not impressive actually.
AlanSubie4Life
Efficiency Obsessed Member
If you're above 270-280 rated miles on your 2018/2019 AWD, your results are pretty normal as far as I can tell (just from all the anecdata here...). Do you have two identical vehicles or is one a RWD, I can't remember...? In any case 10% appears to be totally normal BMS-perceived loss of capacity for Model 3 after a year, and there doesn't appear to be any reliable method to get some recovery.
I'd tell anyone buying an electric car (my family members, etc.), to count on 10% after a year, and hope for slower decay from there. The long term is really tough to know at this point. But definitely expect capacity loss - there are very few reports of cars over a year old with normal mileage without ANY capacity loss. And those that are still at max may have started 4% above the start of visible loss (79+kWh, 3kWh above 76kWh for 2018/2019 LR) so even they have probably lost some energy (would have to check that with CAN reader though).
I'm feeling pretty awesome with my 300 rated miles with 14k miles after 15 months, and it may have even ticked up to 302 after 2020.4.1, not clear yet. I probably started at 78kWh or so, so that's only 6% loss of capacity so far, which is great. I figure this time next year I'll be at 10%; 286 rated miles. I figure in 10 years with 100k miles on the car I'll be at about 230 miles at 100%. The warranty threshold has to be set somehow!
Last edited:
Similar threads
