Battery Management System - What I Learned At Tesla Service Center

[schrodarkian]

I thought the battery only turned red when below 10% not above 10%? Was your car showing 10-15% then suddenly dropped below 10% at the time you took that screen shot? Has your range increased since or still at 211?

Right - so it was down to < 10%. Range was fine as soon as it had charged.

Because of this you should try to be as high as possible SOC to estimate the "full" range.

Thanks hugh_jassol for the explanation on why the BMS can't estimate when the car is at a low SOC.

 

[AlanSubie4Life]

For the LR AWD it's 248 Wh/mi.

248 Wh/mi (I think it's 237 Wh/mi for LRAWD)

Note that these numbers are for the 2020 AWD 18" only, and predated the 2020.4.1 update, which changed the constant (a very rare event) to a value which would result in a match to the EPA rated miles when using 77.6kWh as the maximum energy value. These are no longer the correct values when running 2020.4.1 or later.

The latest post gives the updated values as ~241Wh/rmi and ~226Wh/rmi (again, for 2020 AWD with 18" only), as predicted a few weeks prior to the update.

 

[lencap]

I very much appreciate the detailed posts, and the work that people have put into addressing battery range/health/etc. I didn't intend to create controversy. I just wanted some type of explanation as to why my range decreased by 9% or so from when I bought the car, especially since I've followed the "30-80" charging suggestions. While I can follow most of the comments, I admit that "getting into the weeds" is challenging, and often I find myself getting lost in the details.

So, in a nutshell, what's the best way to look at this? Is my range permanently impaired below the 322 EPA range that was listed for my car? Is the roughly 300 mile range that shows up now on a full charge relatively accurate? Again, my intent isn't to nitpick over range, it's to better understand how to minimize the need to make an extra charging stop on longer trips. I really don't want to stop at a SuperCharger late at night in unfamiliar areas when traveling - knowing my range will let me better avoid that possibility.

 

[AlanSubie4Life]

Is the roughly 300 mile range that shows up now on a full charge relatively accurate? Again, my intent isn't to nitpick over range, it's to better understand how to minimize the need to make an extra charging stop on longer trips.

If you have to stop you have to stop. The trip planner will take care of all of that for you; just give it some extra margin. It's just fine to just use the trip planner and %; there won't be any surprises. But a trip you could barely make when the car was new, you might not be able to make after a couple years, is my guess. But the trip planner will tell you that immediately.

If you want to assess whether your range is "permanently" impaired (and I realize for you and others this is likely not possible), using a metered Wall Connector, do a significant charge at 240V/32A (really important to get that power close to right) - say from 5% to 90% - except change to rated miles and track those for better resolution, with no accessory use (turn off HVAC first), and compare to the AC charging event data in my spreadsheet.

That charging event includes recharging the buffer. So you have to make some estimations, but we can make some estimations:

We know that a full charge including the buffer, when vehicle is new, AC wall energy, is ~89kWh+/-1kWh.

Reduce that by 4.5% (buffer). So you should see ~85kWh from the wall for a 0-100% charge at 7.7kW.

If you do a 5-90% charge, I would expect to see 72.3kWh from the wall. (0.85*85kWh).

But if you really are at 300rated miles rather than 310 rated miles (that's only for 2018/2019 vehicles - use 322 for 2020), I would expect you would see:

72.3kWh * 300/323 = 67.2kWh (5-90% @ 300rmi max) I used 323rmi because that's what the 2018 test vehicle actually started with, not 310...complicated...

So, there are a lot of sources of error here, but a 5kWh difference in the amount of energy you have to put into the vehicle should be quite measurable and will be apparent even with those error sources. If you only have a couple miles of loss of capacity this method will not be helpful.

If someone has a meter this would be really easy to verify, TBH. I've never actually done a careful measurement myself over a very large charge event. All I know is that the efficiencies line up with the values in the Tesla's EPA documents. If you have significant capacity loss it will be really evident. We know the charging efficiency is 88.5% or so at 7.7kW, when charging to 100% (you can see for the SR it is slightly better due to the lack of taper - 89.3% - so use that value if you want if not charging to 100%).

 

[SSedan]

I very much appreciate the detailed posts, and the work that people have put into addressing battery range/health/etc. I didn't intend to create controversy. I just wanted some type of explanation as to why my range decreased by 9% or so from when I bought the car, especially since I've followed the "30-80" charging suggestions. While I can follow most of the comments, I admit that "getting into the weeds" is challenging, and often I find myself getting lost in the details.

So, in a nutshell, what's the best way to look at this? Is my range permanently impaired below the 322 EPA range that was listed for my car? Is the roughly 300 mile range that shows up now on a full charge relatively accurate? Again, my intent isn't to nitpick over range, it's to better understand how to minimize the need to make an extra charging stop on longer trips. I really don't want to stop at a SuperCharger late at night in unfamiliar areas when traveling - knowing my range will let me better avoid that possibility.

As others have said when it doesn't get to the extremes where it can measure voltage variation it kind of loses its place and begins to mildly underreport.
The battery does tend to actually lose a bit the first year, but not 8%. What you have going on is a little actual degradation and some BMS miss calibration.

Use scheduled departure and charge to 90% a few times and it should find its place a little. Tesla recommends 90% and keep it plugged in so while an argument can be made this isn't best for the pack it is not bad for it either.

 

[im.thatoneguy]

That would help explain why my 2018 Model 3 that I always street park and regularly use ~300+wh/mi due to heat and low speeds still has 315 miles of range. I almost always am running up to 90% and down to 10-20%.

 

[TimothyHW3]

I have been saying for a while now that the reason most people see "degradation" (mostly TeslaFi extrapolated data) is because they never charge pass 80%-90% and mostly keep the car in the upper 60%-90% daily for most of the time. There have been numerous reports of this over the year. And before someone says that "on my X/S this didn't happen" - well it is an S or an X different battery and chemistry and BMS. It seems the Model 3 BMS works differently and needs these deeper discharges and doesn't like it when it is kept in those higher SOC %.

So it is probably best to run it from 20%-80/90% and 10%-90% from time to time and NOT charge always and especially not keep the car between 70% to 90% like some people do.

 

[hcdavis3]

I have been saying for a while now that the reason most people see "degradation" (mostly TeslaFi extrapolated data) is because they never charge pass 80%-90% and mostly keep the car in the upper 60%-90% daily for most of the time. There have been numerous reports of this over the year. And before someone says that "on my X/S this didn't happen" - well it is an S or an X different battery and chemistry and BMS. It seems the Model 3 BMS works differently and needs these deeper discharges and doesn't like it when it is kept in those higher SOC %.

So it is probably best to run it from 20%-80/90% and 10%-90% from time to time and NOT charge always and especially not keep the car between 70% to 90% like some people do.

Timothy, can you put “not charge always” in context of best state of charge?

 

[kkillebrew]

For some months my range has been decreasing very slowly from 333 down to 320 after 18K miles. The last seven mile loss happened very quickly. Well aware of the points made in this thread I was not too bothered by this, as there are also occasions where range goes up and I realize that these are all estimates, often based on fluctuating algorithms.

So I had an opportunity to test the advice given the OP by the Tesla tech. I charged to 99% and did a round trip that left me with 8% upon return. When I put the car in park in my garage I got a warning message on the big screen that range was dangerously low and to charge. Before I got to plugging in the charger, less than 15 minutes, I got a notification from the app that my charge was dangerously low and to avoid damage to the battery and vehicle I needed to charge immediately. Yes the word damage was used, and the range at 4% was in red, so this was a 4% drop in a few minutes while parked. So I plugged in and charged to 90%. I will add that the temperature was rapidly dropping at the time although my car was put in the garage.

Charging went fine and yesterday I drove just a few miles taking the car down to 87%, parked it for the night without plugging it in and this morning it is showing a 93% charge and range is down from 320 to 318 miles. What is up?

 

[hugh_jassol]

Timothy, can you put “not charge always” in context of best state of charge?

he means - don't charge until the car gets below 20%... so in other words, don't charge every night if you only use 10% of your battery per day.

Again, this is only a *display* thing... it does not change the actual capacity of the battery. It just helps calibrate the "fuel gauge". Your car will drive the same distance either way as when it "shuts off" is based on actual battery output rather than when your display hits 0.

 

[SSedan]

For some months my range has been decreasing very slowly from 333 down to 320 after 18K miles. The last seven mile loss happened very quickly. Well aware of the points made in this thread I was not too bothered by this, as there are also occasions where range goes up and I realize that these are all estimates, often based on fluctuating algorithms.

So I had an opportunity to test the advice given the OP by the Tesla tech. I charged to 99% and did a round trip that left me with 8% upon return. When I put the car in park in my garage I got a warning message on the big screen that range was dangerously low and to charge. Before I got to plugging in the charger, less than 15 minutes, I got a notification from the app that my charge was dangerously low and to avoid damage to the battery and vehicle I needed to charge immediately. Yes the word damage was used, and the range at 4% was in red, so this was a 4% drop in a few minutes while parked. So I plugged in and charged to 90%. I will add that the temperature was rapidly dropping at the time although my car was put in the garage.

Charging went fine and yesterday I drove just a few miles taking the car down to 87%, parked it for the night without plugging it in and this morning it is showing a 93% charge and range is down from 320 to 318 miles. What is up?

Battery cooled off.

 

[mvsmucker1971]

I agree with this and drdumont, cycling BMS is almost pointless. In the long run, you are not doing the battery any favors.

I think the problem with people obsessing about range is they have a “gas tank” mentality about electric cars. Electric car batteries do not function like a gas tank.

Most important thing about the car is know your daily driving distance. This applies to Model 3 LR... If you drive 50 miles or less a day, charge it to 60%, 100 miles or less per day, charging between 70%-80% is plenty. If you drive 200 miles, charge it to 90%. If you drive more than that routinely every day.. god help you.

For road trips, Superchargers are spaced about 150 miles apart – 2 hrs of driving/restroom break… charge to about 60% and go down to 10%. Most locations and times, your car will be ready to go after you use the bathroom.

Don’t scare new over the road drivers: I have been driving long distance trips (1100 mi one way) for 6 years on my S 85. Superchargers used to be 150 mi apart, most places in rural midwest that distance is now 80-100 miles. As one friend says who has tried shorter trips in non Teslas: “Don’t try that with any other car, it’s not pretty out there”. Charging for 90 minutes at a stop is an unnecessary thing of the past.

There are exceptions, like the state of Arkansas.

 

[TimothyHW3]

he means - don't charge until the car gets below 20%... so in other words, don't charge every night if you only use 10% of your battery per day.

Again, this is only a *display* thing... it does not change the actual capacity of the battery. It just helps calibrate the "fuel gauge". Your car will drive the same distance either way as when it "shuts off" is based on actual battery output rather than when your display hits 0.

I am not quite sure that it is just a display thing. It does change the capacity the car thinks it has. So it is basically thinking you have 40 litres where you might have 60 litres, but the BMS doesn't see the extra capacity.

I don't know what happens on a car with a messed up BMS. I can only assume that if the car thinks it has 40 litres it will drive for 40 litres and stop even if you have 20 more litres in the tank the car isn't aware of.

In other terms. If the BMS thinks you only have 70kWh even though the capacity of the battery might be actually more, 75kWh, once you reach the 70th kWh and it says 0% it might shutdown or use the buffer up to a point (let's say another 2-3kWh) and then shutdown.

Same goes if the BMS thinks you have 75, but you actually have 70. It has happened before on a few cars that stopped at about 15km remaning, Bjorn Nyland being one.

But since most people will never or almost never use their full "capacity" available or otherwise, this doesn't really matter to 99%. So unless you go to below 5% you will never notice the difference.

By the way, did you manage to check your cell voltage above 4.00 on Scan My Tesla and match it to the SOC % on a Mid Range? Like I said in the other thread - according to the internal documents the car starts balancing once you are above 4.V so on a LR 4V is about 73%, wonder what it is on a Mid range?

 

[hcdavis3]

he means - don't charge until the car gets below 20%... so in other words, don't charge every night if you only use 10% of your battery per day.

Again, this is only a *display* thing... it does not change the actual capacity of the battery. It just helps calibrate the "fuel gauge". Your car will drive the same distance either way as when it "shuts off" is based on actual battery output rather than when your display hits 0.

Do you agree with this practice Hugh?

 

[hugh_jassol]

I don't know what happens on a car with a messed up BMS. I can only assume that if the car thinks it has 40 litres it will drive for 40 litres and stop even if you have 20 more litres in the tank the car isn't aware of.

In other terms. If the BMS thinks you only have 70kWh even though the capacity of the battery might be actually more, 75kWh, once you reach the 70th kWh and it says 0% it might shutdown or use the buffer up to a point (let's say another 2-3kWh) and then shutdown.

In one of the links posted earlier in this thread, there’s info from tesla that the “empty” is determined by voltage/power output of the battery, not the estimate of the BMS... so theoretically, it should drive until actually empty.

 

[hugh_jassol]

Do you agree with this practice Hugh?

I don’t know what me agreeing/disagreeing with this would mean

I recently did a month of 20% -> 90% charging followed by a 100% on a supercharger (only may second ever supercharger event). Doing that raised my “Nominal Full Pack” by 0.4 kWh (about 2 miles). This is after my ~ year of daily charging to 80% or 90% where my range reduced from 264 new to as low as 244, then back up to 248 before my experiment.

So I think all of that is to say it appears the BMS is doing a pretty good job of tracking my battery’s actual capacity. As trying to recalibrate it did make an actual change in the car’s estimate of my pack capacity... it was negligible. And I think I am, in fact, seeing a real degradation of about 5.3% after 12k miles.

 

[TimothyHW3]

In one of the links posted earlier in this thread, there’s info from tesla that the “empty” is determined by voltage/power output of the battery, not the estimate of the BMS... so theoretically, it should drive until actually empty.

Can you post me to that link? As far as I understand it - the BMS is the one calculating the voltage/power. Otherwise where will the system get its data from and why would it display the SOC based on the BMS info we are reading?

So you can't check the SOC % at mid cell >4.0V

 

[hcdavis3]

I don’t know what me agreeing/disagreeing with this would mean

I recently did a month of 20% -> 90% charging followed by a 100% on a supercharger (only may second ever supercharger event). Doing that raised my “Nominal Full Pack” by 0.4 kWh (about 2 miles). This is after my ~ year of daily charging to 80% or 90% where my range reduced from 264 new to as low as 244, then back up to 248 before my experiment.

So I think all of that is to say it appears the BMS is doing a pretty good job of tracking my battery’s actual capacity. As trying to recalibrate it did make an actual change in the car’s estimate of my pack capacity... it was negligible. And I think I am, in fact, seeing a real degradation of about 5.3% after 12k miles.

Thanks

 

[hugh_jassol]

Can you post me to that link? As far as I understand it - the BMS is the one calculating the voltage/power. Otherwise where will the system get its data from and why would it display the SOC based on the BMS info we are reading?

“This does not affect the true range of the vehicle, as the end-of-drive conditions are based on real-time battery measurements of reducing battery power rather than software estimates.” Source.

So you can't check the SOC % at mid cell >4.0V

Working on it haven’t gotten that low since my recent charge - I’m tracking it though and ready to report when it happens.

 

[SSedan]

As has been said the top and bottom 10% voltage actually changes so the BMS is measuring the state of the battery then.