How I Recovered Half of my Battery's Lost Capacity

[TimothyHW3]

So is there a simple (or complicated?) way for us 2021 LR owners to know if our cars have an LG or Panasonic battery? I'd like to know out of curiosity

In the US Tesla is delivering Panasonic only for EPA purposes. You can't achieve EPA with LG.

 

[MikeS]

It's not supposed to prevent the car from sleeping, but some users have posted complaints about it. Is there a setting? No. Depending upon what you are interested in, there are other apps, that may meet your needs

Thanks. It comes down to - I want to monitor battery health without encountering the issue of the app not allowing the car to "sleep" and get the OCV data points. Would you suggest an app that can do that?

 

[TimothyHW3]

monitor battery health

For battery health and other stuff the only proper way is scan my tesla+adapter and OBD II.

Apps don't have access to battery data.

The only thing you need API-Apps is if you want to log stuff, like trips, charge curve etc.

For anything else you can either do the math yourself based on rated km and typical constant or use scan my tesla.

 

[KenC]

Thanks. It comes down to - I want to monitor battery health without encountering the issue of the app not allowing the car to "sleep" and get the OCV data points. Would you suggest an app that can do that?

Yep, I agree with Tim, the best solution is to use an OBD2 reader and SMT. Barring that, I would query the developers of TeslaFi and Stats.

 

[Swordtail]

yes. rated miles is a unit of energy (its directly linked to a constant of watthourage). % is not.

This becomes really important when you have degradation as the closer you get to 0% the closer the car appears to have no degradation. Or in other words, range per % increases the lower your state of charge is - if you have degradation.

I'm struggling a bit to get my head around all this . So to work out range, I'd have to charge the battery to 100% and look at how many rated miles that equates to on the battery meter. Let's say for ease of math, it says 400 miles.

If I then drive 50 miles, say, but find the rated miles has dropped by 100 miles to 300 miles, that would mean the actual range of the car would be 200 miles if driven at the same Wh/mile. Given this, I could also use the Wh/mile reported by the trip for the 50 mile journey to work out the usable capacity of the battery.

Is all that correct?? I know nothing is 100% accurate but I'm interested in getting as close as I can without using all of these analysis tools which I expect are a bit beyond me.

Cheers!

 

[Candleflame]

I'm struggling a bit to get my head around all this . So to work out range, I'd have to charge the battery to 100% and look at how many rated miles that equates to on the battery meter. Let's say for ease of math, it says 400 miles.

If I then drive 50 miles, say, but find the rated miles has dropped by 100 miles to 300 miles, that would mean the actual range of the car would be 200 miles if driven at the same Wh/mile. Given this, I could also use the Wh/mile reported by the trip for the 50 mile journey to work out the usable capacity of the battery.

Is all that correct?? I know nothing is 100% accurate but I'm interested in getting as close as I can without using all of these analysis tools which I expect are a bit beyond me.

Cheers!

thats not how it works. rated miles is a unit of energy. It is not related to the distance you have driven etc and its one of the most common misconceptions about the Tesla GUI. Rated miles is a direct reflection of what the BMS thinks is the kwh remaining in the battery. It is linked to a constant Tesla determines. That also means that if your battery degrades or the BMS thinks you have less capacity or Tesla batterygates you that your rated miles will go down. 1% is't just the rated range divided by 100, 1% is an equation where if your SOC is 100% then 1% = 1% of your current rated range and if your SOC is 0% then 1% equals the rated range the car had when it was new/what it was rated for. This means that your last 10% have a lot more energy than your first 10% if you have heavy degradation.
Which is also the reason you cant use % to do range and degradation calculations unless you completely drive it down to 0. Of course all of this is irrelevant because you are essentially just chasing a few km here and there.

If you drive 50 miles and find your rated miles have dropped by 100 miles then that means you have used twice the amount of rated miles which essentially means twice the amount of energy.

Given this, I could also use the Wh/mile reported by the trip for the 50 mile journey to work out the usable capacity of the battery

Cheers!

You sort of could but its quite inaccurate and the car already reports the useable capacity of the battery to you via rated miles which is just another way of displaying remaning useable kwh. There is a slight caviat - that being that if you have absoloutely no degradation your rated range is always capped at the maximum rated range tesla dictates. On some of the older Model 3s it hides the initial 3% of degradation or so. In that case you can do the following below:

If you want to use the wh/mile to calculate the capacity of the battery rather than use rated miles you can just charge to i.e. 90% or 100% or whatever and then go into your energy meter, set this to "average" and then multiple the wh/mile by the remaining range and then divide by the % SOC i.e. if it says an average of 250wh/mile and you have 200 miles remaining at 100% thats 250 x 200 / 1 = 50kwh. This does not include the buffer or brick protection (and neither do your rated miles).

Is all that correct?? I know nothing is 100% accurate but I'm interested in getting as close as I can without using all of these analysis tools which I expect are a bit beyond me.
Cheers!

Then charge to 100%, let the car sleep for 3 hours. Then look at your rated miles and time that by the rated consumption of your car (as in the watthours per rated mile. You cant look this up on the EPA because that includes the buffer and tesla sometimes modifies this a bit, you have to work it out or ask someone else. I.e. for my model 3 its 153wh/km i think).

I personally wouldnt charge to 100%, i would just charge to 90%, then let it sleep for 3 hours and then look at the rated miles and extrapolate from there by dividing by 0.9.

If you do not know what your rated range constant is (with all the new batteries nowdays it seems to change all the time!) then the best way to know how many kwh you have is to just go in the energy screen and while it is set to average is to multiple your average consumption by the remaining range displayed and then divide that by the % left. Obviously more accurate the higher you charge. If you charge to 100% you can also then extrapolate from that how many wh your rated miles are!


edit: I might also add here that you will find many videos on youtube from socalled youtube pros and tesla experts which i.e. proclaim that you have to drive the battery down to 0 and beyond to find out how many kwh you truly have remaining and how "off" the rated range is by showing how their measured degradation is less than the cars calculated degradation - I find most of them either have flawed testing methodology or they make some error in their calculations. The ones where the test is done properly the rated range always matches the displayed consumption of the car when accounted for some heatloss by <1%. The rated range remaining is extremely accurate provided that the car has time to rest for 3 hours to take a high voltage reading.

 

[Itsuo-DC]

Thanks to all for a great initial post and thread. Does Stats for Tesla prevent the vehicle from sleeping while parked at night and thereby prevent the OCV readings to which the original post referred? If it does, is there a setting in the app that can remove the interference? I'm thinking about buying the app but don't want to create range indication problems for myself.

I’ve never experienced this. The vehicle is often in sleep mode after its been sitting for a while. I really enjoy and recommend it.

 

[KalJoMoS]

I'm struggling a bit to get my head around all this .

Is all that correct?? I know nothing is 100% accurate but I'm interested in getting as close as I can without using all of these analysis tools which I expect are a bit beyond me.

Cheers!

Perhaps this photo gives better understanding, not to mention very detailed descriptions by @Candleflame. To calculate your consumption in kW/h will give clearer understanding if you can follow it during your trips. This is official EPA statement:

 

[TimothyHW3]

I'm struggling a bit to get my head around all this . So to work out range,

Basically the energy trip meter in your car will give you a pretty good estimation of your range.

This video might help you out too. Every Tesla car has a rated range which is a representation of the capacity in the car. What you see under the battery is not a GOM as in every other EV(guess-o-meter based on let's say last trip consumption), but an exact representation of kWh in miles.

Every Tesla has a so called constant that Tesla uses to work this backwords from kWh to miles.

For example on a 2019 model it shows 310 miles when new and you have around 76-77 kWh capacity.

So you can divide 77/3.1 and get the constant which is around 248wh/m (it is not, but this is just a rough estimation without diggint too deep).

And if you know that you will drive 300Wh/m on avg, then you take 300/248, let's say around 1.2 so you will have 1.2 less miles as displayed so around 310/1.2 = 250 miles

And then you have a thing Tesla does, shown in the video, that they hide about 10 miles while you drive below 0%, so to 0% you have about 240miles.

This calculation is explained towards the end of the video. But basically if you know how much kWh you have, you can work your way backwords to get miles.

 

[TimothyHW3]

But back to topic. I see a lot of people gaining back "miles" by charging to 60% consistantly over 2 months or more. MIght be worth a try. Basically charging 10%-60%.

 

[Candleflame]

Basically the energy trip meter in your car will give you a pretty good estimation of your range.

This video might help you out too. Every Tesla car has a rated range which is a representation of the capacity in the car. What you see under the battery is not a GOM as in every other EV(guess-o-meter based on let's say last trip consumption), but an exact representation of kWh in miles.

Every Tesla has a so called constant that Tesla uses to work this backwords from kWh to miles.

For example on a 2019 model it shows 310 miles when new and you have around 76-77 kWh capacity.

So you can divide 77/3.1 and get the constant which is around 248wh/m (it is not, but this is just a rough estimation without diggint too deep).

And if you know that you will drive 300Wh/m on avg, then you take 300/248, let's say around 1.2 so you will have 1.2 less miles as displayed so around 310/1.2 = 250 miles

And then you have a thing Tesla does, shown in the video, that they hide about 10 miles while you drive below 0%, so to 0% you have about 240miles.

This calculation is explained towards the end of the video. But basically if you know how much kWh you have, you can work your way backwords to get miles.

he wants to work out how many kwh he has in his battery, not what his remaining range is. AFAIK.

 

[Candleflame]

But back to topic. I see a lot of people gaining back "miles" by charging to 60% consistantly over 2 months or more. MIght be worth a try. Basically charging 10%-60%.

that would be similar to the OT where the car cant get enough readings. I would presume that that only happens with people who shallow cycle the battery frequently below 90% - as frequent cycling between 40 and 70% might confuse the BMS slightly.
Unlikely to apply to most users.

 

[AAKEE]

Perhaps this photo gives better understanding, not to mention very detailed descriptions by @Candleflame. To calculate your consumption in kW/h will give clearer understanding if you can follow it during your trips. This is official EPA statement:
View attachment 628047

I guess EPA( or the Tesla quite close to EPA) range indication is quite overoptimistic( better then WLTP) but still overoptimistic.

Charging up to 100% would only be needed for a more exact knowledge of the maximum range that the car (BMS) will show. This for simple battery degradation calculation I guess.

As this is not very close to the true range a very cold winter day so to know the real range at a specific time I guess the Energy App on the Tesla screen is better. I have not used it at low SOC yet. It seems to calculate to a real empty batt and not take the brick protection in count ? Ye

he wants to work out how many kwh he has in his battery, not what his remaining range is. AFAIK.

I have Scan My Tesla and a ’21 M3 performance ( 82.1kwh branded battery).



Today scan my tesla reported 80.9Kwh nominal full pack. Calculation from on screen values gave me 80.2kwh.
I have done the check [ *Energy app 50km range* times *energy app consumption* divided by *charging state(SOC) on the screen*. about five times. Today, 54% SOC, 264wh/km, range 164km. OAT -21 C degrees and the car has been outside for more then 24h due to work on my daughters Audi.
Each time I have ended up with maximum 0.5 Kwh delta from SMT ”nominal full pack”, except for todays 0.7 kwh ( = 1% of total value).
Depending on the last month charging cycle the BMS could be more of than this anyway, so even with scan my tesla the value could be way more wrong.

I would say thar if the only thing youre interrested in is the nominal full( calculated number of Kwh from the BMS) you can do the on screen value calc, and be happy with that. Scan my tesla app+ ODB2 bluetooth dongle+ odb2 harness is not that cheap if you dont like to use it for other reasons.
I absolutely recommend SMT, its good to have

 

[TimothyHW3]

It seems to calculate to a real empty batt and not take the brick protection in count

Not the brick protection, but the software "battery protection" - the buffer in SMT. This is gradually removed while driving without you noticing and put down below 0%.

The energy tab as the battery indicator use the full nominal capacity with the buffer included, hence you not noticing the difference unless driving at steady speeds non stop like I did in the above video.

Basically for each 100 or so km, Tesla hides about 5km below 0%. Or 700Wh, 1/5 of the buffer.

Each time I have ended up with maximum 0.5 Kwh delta from SMT ”nominal full pack”,

That could be due to the above plus taking into account the different SOC%. The lower you get to 0% the lower the difference is, because the buffer has been eaten up already for the most part.

Also, in cold conditions you should look at the value "expected remaining", not nominal remaining, to do the calculations. This is the value the BMS uses to calculate range and when the battery is cold there is difference between the two.

 

[AAKEE]

Not the brick protection, but the software "battery protection" - the buffer in SMT. This is gradually removed while driving without you noticing and put down below 0%.

The energy tab as the battery indicator use the full nominal capacity with the buffer included, hence you not noticing the difference unless driving at steady speeds non stop like I did in the above video.

Basically for each 100 or so km, Tesla hides about 5km below 0%. Or 700Wh, 1/5 of the buffer.


That could be due to the above plus taking into account the different SOC%. The lower you get to 0% the lower the difference is, because the buffer has been eaten up already for the most part.

Also, in cold conditions you should look at the value "expected remaining", not nominal remaining, to do the calculations. This is the value the BMS uses to calculate range and when the battery is cold there is difference between the two.

Just saw that video -very good explanation. Was wondering about how the energy app handled the buffer(also thought buffer and brick protection was the same thing.)

 

[Swordtail]

Then charge to 100%, let the car sleep for 3 hours. Then look at your rated miles and time that by the rated consumption of your car (as in the watthours per rated mile. You cant look this up on the EPA because that includes the buffer and tesla sometimes modifies this a bit, you have to work it out or ask someone else. I.e. for my model 3 its 153wh/km i think).

I personally wouldnt charge to 100%, i would just charge to 90%, then let it sleep for 3 hours and then look at the rated miles and extrapolate from there by dividing by 0.9.

If you do not know what your rated range constant is (with all the new batteries nowdays it seems to change all the time!) then the best way to know how many kwh you have is to just go in the energy screen and while it is set to average is to multiple your average consumption by the remaining range displayed and then divide that by the % left. Obviously more accurate the higher you charge. If you charge to 100% you can also then extrapolate from that how many wh your rated miles are!

Thanks for this, makes sense and I'll try this out soon. I give up trying to find out the exact full capacity of the battery as I read different things everywhere. Calculating the usable capacity using this method sounds like a more useful approach.

I did try using the rated miles to try work out range at a particular Wh/mi. So I drove 19 miles which used 24 rated miles @ 268 Wh/mi. So that's about 0.8 real miles per rated mile. The battery showed 271 rated miles at the start. So from this I can say I'll get 271 * 0.8 = 216 miles range with a battery charged to 271 rated miles if driving at 268 Wh/mi.

I can also say that in 19 miles I used 19 * 268 = 5.1 kWh energy. So 1 rated mile = 5.1 / 24 = 212 Wh.

I read the 100% rated miles for my car is 332, but I'll test it at some point. If it is, 332 * 212 = 70.4 kWh. I read somewhere the battery on my car is 74 kWh with a usable capacity of 70 kWh, so this sounds about right (so much for the 82 kWh I read in other places!).

However, I also read that the rated consumption of my car is 235 Wh/mi, which makes me question my 212 Wh result, so still a bit confused

 

[elptxjc]

The only real number you need is how many Wh/mi your car uses the way YOU drive it. On our current long trip, traveling in cold weather at 80+ mph, my car uses 340 Wh/mi... which in real life terms equates to 2.5 miles per 1% of charge, so that's all I care to know . In addition, I don't like to charge more than 90% (takes too long), and arrive with 30% charge (in case there are head winds, or other energy consuming 'issues', like it happened twice), which leaves 60% of the charge to consume, or 150 miles. And that worked perfectly, since superchargers are located from 98 to 147 miles. But just like I suspected, the LR is the minimum I'd consider on an EV, since real range is FAR from the 353-mile non-sense reported by Tesla. At least if you want to travel on the damn thing like on a normal car. Ha ha.

 

[Candleflame]

I guess EPA( or the Tesla quite close to EPA) range indication is quite overoptimistic( better then WLTP) but still overoptimistic.

Charging up to 100% would only be needed for a more exact knowledge of the maximum range that the car (BMS) will show. This for simple battery degradation calculation I guess.

As this is not very close to the true range a very cold winter day so to know the real range at a specific time I guess the Energy App on the Tesla screen is better. I have not used it at low SOC yet. It seems to calculate to a real empty batt and not take the brick protection in count ? Ye


I have Scan My Tesla and a ’21 M3 performance ( 82.1kwh branded battery).



Today scan my tesla reported 80.9Kwh nominal full pack. Calculation from on screen values gave me 80.2kwh.
I have done the check [ *Energy app 50km range* times *energy app consumption* divided by *charging state(SOC) on the screen*. about five times. Today, 54% SOC, 264wh/km, range 164km. OAT -21 C degrees and the car has been outside for more then 24h due to work on my daughters Audi.
Each time I have ended up with maximum 0.5 Kwh delta from SMT ”nominal full pack”, except for todays 0.7 kwh ( = 1% of total value).
Depending on the last month charging cycle the BMS could be more of than this anyway, so even with scan my tesla the value could be way more wrong.

I would say thar if the only thing youre interrested in is the nominal full( calculated number of Kwh from the BMS) you can do the on screen value calc, and be happy with that. Scan my tesla app+ ODB2 bluetooth dongle+ odb2 harness is not that cheap if you dont like to use it for other reasons.
I absolutely recommend SMT, its good to have

that does not make sense as the rated range and remaining range are both directly computed from the nominal full pack (rated range is just the full pack displayed and divided by the constant).

 

[Swordtail]

The only real number you need is how many Wh/mi your car uses the way YOU drive it. On our current long trip, traveling in cold weather at 80+ mph, my car uses 340 Wh/mi... which in real life terms equates to 2.5 miles per 1% of charge, so that's all I care to know . In addition, I don't like to charge more than 90% (takes too long), and arrive with 30% charge (in case there are head winds, or other energy consuming 'issues', like it happened twice), which leaves 60% of the charge to consume, or 150 miles. And that worked perfectly, since superchargers are located from 98 to 147 miles. But just like I suspected, the LR is the minimum I'd consider on an EV, since real range is FAR from the 353-mile non-sense reported by Tesla. At least if you want to travel on the damn thing like on a normal car. Ha ha.

Yes, I agree, at a high level I'm concluding I can do about 200 miles in my M3 LR in the cold (just above freezing) without preconditioning the battery starting at 90% and ending around 10%, using about 320 Wh/mi. I'm expecting this to increase in summer, particularly as the UK summers are generally around the ideal temperature for an EV, so I understand. It's just there's so many variables that impact energy usage that the real range from 90% to 10% could be anywhere between 200 to 350+ miles from what I've read!

However, I'm grateful to CandleFlame for explaining the intricacies mainly from an interest point of view as I'm new to EVs. I do like to understand the detail, what numbers can be trusted and what they mean etc. before reverting to the high level approach.

 

[Battpower]

uh. maybe they only limited it to 4.15V on the Roadster and the early Model S then?

That's correct iirc. For Roadster. Must be very early S? I understand that the current cells can withstand as high as 4.4v under ideal conditions, so their is still a safety margin. 4.2v seems fairly common in a range of applications.

A Bit About Batteries

Remember that bettery chemistry and support systems have moved on since this blog post, but still good general info.