Battery temperature and preconditioning questions

[CKwik240]

So I took a drive up to Big Bear Lake on Sunday. Noticed some odd behavior. First off, I navigated to a SC near the bottom of the mountain to to of a bit before driving up. As soon as I entered the SC in the nav, preconditioning started. Its nearly an hour away. Is that normal behavior. It remained preconditioning the entire drive there. Outdoor temp started at about 67F. Is this normal?

On the way back down the mountain, Regen maintained the battery level, but near the end where it was relatively flat for a bit, it bumped up slightly. I would have expected it to go up during steeper downhills.

Lastly, during the drive home, I kept seeing the battery percentage go up at times even though the drive is fairly flat. I was monitoring in percentage at this point, so it seemed odd. Not sure if it was somehow picking up a higher state of charge as the battery was discharging. This is less of a concern, but just something I observed.

Got home with about 5% left so I let it sit for a few hours then plugged in. Set it to charge to 100%. At first it indicated it would reach 300. Then after a bit of time, started 283. Then settled at 273 which is what I woke up to and had timed it so it had at least three hours of time to sit after charging before I left for work. I've never seen the charge limit change while charging. Tessie indicates 271 is my max range. I had another post earlier this year where the max range dropped abruptly from some 287-289 to about 267. It's gone up a tad since, but not significantly. Weather has finally started to get somewhat warm here in socal.

When I look at the overall behavior, it almost seems like one of the battery temperature sensors is reading low. The near hour preconditioning was particularly concerning as I don't think I've ever seen it come on that early. But I don't supercharger often. Anyone else see anything like this before?

I'm ordering myself an OBD2 adapter and going to check it with Scan my Tesla but it may be a while before it arrives. Just trying to get some thoughts in the meantime. TIA.

 

[GtiMart]

The battery needs to be over 100F for proper supercharging, and augmenting the temperature of such a big mass takes a long time. It is not surprising that it would start 1h before the SC, especially at 67F, and your battery might even be cooler than that. If you have a single motor heating takes even longer by the way.
After supercharging, the car needs to cool the battery back down. It might spend significant energy doing that... maybe that's why your SOC wasn't going up so much while going down the hill.
For the rest, bringing the battery way down and charging it way up does help the BMS recalibrate. It's possible that some of the changes you have seen are caused by that.

 

[derotam]

So I took a drive up to Big Bear Lake on Sunday. Noticed some odd behavior. First off, I navigated to a SC near the bottom of the mountain to to of a bit before driving up. As soon as I entered the SC in the nav, preconditioning started. Its nearly an hour away. Is that normal behavior. It remained preconditioning the entire drive there. Outdoor temp started at about 67F. Is this normal?

Yes it is normal. The car would be perfectly happy getting the battery up to 40-50C before charging but it can only do so much depending on the car and specific scenario...for example...the car will only generate so much excess heat to warm up the battery between using the heat pump(if equipped) and or providing non-motive power to the motor/s to generate excess heat to pump into the battery, Tesla has an algorithm to determine what it is going to try and do. I am pretty sure that algorithm is a mystery to anyone but Tesla.

Lastly, during the drive home, I kept seeing the battery percentage go up at times even though the drive is fairly flat. I was monitoring in percentage at this point, so it seemed odd. Not sure if it was somehow picking up a higher state of charge as the battery was discharging. This is less of a concern, but just something I observed.

That can happen as the car's BMS makes adjustments to the estimates based on past driving. For example, if you drive down a long hill, it may still take a few minutes for the car to adjust the display to reflect that energy gained from regen.

I'm ordering myself an OBD2 adapter and going to check it with Scan my Tesla but it may be a while before it arrives. Just trying to get some thoughts in the meantime. TIA.

This can give you lots of insights for lots of things...and it can also not tell you things you would like to know, haha...like how much energy if being used by various systems specifically for heat generation for preconditioning(while driving).

 

[derotam]

The battery needs to be over 100F for proper supercharging, and augmenting the temperature of such a big mass takes a long time. It is not surprising that it would start 1h before the SC, especially at 67F, and your battery might even be cooler than that. If you have a single motor heating takes even longer by the way.
After supercharging, the car needs to cool the battery back down. It might spend significant energy doing that... maybe that's why your SOC wasn't going up so much while going down the hill.
For the rest, bringing the battery way down and charging it way up does help the BMS recalibrate. It's possible that some of the changes you have seen are caused by that.

I'll argue a bit here...do you have a definition for "proper supercharging"? If you want to make a battery health argument ok I guess, but as far as getting the max rated supercharging capacity out of a supercharger, you can do that at less than 100F. I feel like I used to have data for this, though with all the changes over the last couple years it would need to be updated.

 

[GtiMart]

I can rewrite to "the car wants to get to over 100F for supercharging" if you prefer. I believe I've seen mine go way over that before arriving to charge, and going over 140F during supercharging. That's what the car wants to do, I'm not talking about battery health or anything. I don't remember if I got full 250kW at lower temperatures but I certainly got 250kW at those temperatures.

 

[derotam]

I can rewrite to "the car wants to get to over 100F for supercharging" if you prefer. I believe I've seen mine go way over that before arriving to charge, and going over 140F during supercharging. That's what the car wants to do, I'm not talking about battery health or anything. I don't remember if I got full 250kW at lower temperatures but I certainly got 250kW at those temperatures.

Haha, yeah. I was just on a trip and it got my battery up to I think 47C while preconditioning in route to a supercharger which was in my opinion completely unnecessary. I have always been bothered by the seemingly unnecessary amount of preconditioning the car does in preparation for a supercharger.

 

[AAKEE]

Haha, yeah. I was just on a trip and it got my battery up to I think 47C while preconditioning in route to a supercharger which was in my opinion completely unnecessary. I have always been bothered by the seemingly unnecessary amount of preconditioning the car does in preparation for a supercharger.

For V3 supercvharges, my M3P 2021 always precondition to 48C (about 120F). Always, every time.
Before a change in the softare in september 2022 or so, the car did the same preheating (48C) independent of supercharger v1,v2,v3. But since, they have optomized it to not preheat more than needed. End temp for a not very short V3-session is about 58C.
(I have a lot of logs of this, via Scan My Tesla and teslalogger).

Preheating has two purposes, if you ask me. It shortens the charging time by mimimizing the internal resistance. But it also reduces lithium plating, which is the bad part of fast charging. So, when we see 48C it is on the right side of the limit that causes lithium plating. Probably some lithium plating occur anyway (probably at higher SOC levels), but the high temperature is the weapon against lithium plating.

That said, I always precondition completely if it is possible. A few yea back my car did start the prehetaing much later, like 30 mins or so in winter time, these days its about one hour before.

 

[CKwik240]

Thank you gentlemen. I just hadn't had it take that long to precondition the battery before so I wanted to be sure. I should receive a harness for an OBD2 port. I have a hand full of OBD2 dongles somewhere around here that I'll try. I'll probably try my Innova scanner just to see what it does as well as a Bluedriver scanner. Hopefully I won't have to buy another, but I'm anticipating I will.

 

[AlanSubie4Life]

Set it to charge to 100%. At first it indicated it would reach 300.

These massive errors are normal since it is just extrapolating and the API has no idea what your max charge will be. At very low SOC these estimates can be off by 100 miles due to rounding errors.

Very normal and nothing to do with BMS.

 

[CKwik240]

Yes it is normal. The car would be perfectly happy getting the battery up to 40-50C before charging but it can only do so much depending on the car and specific scenario...for example...the car will only generate so much excess heat to warm up the battery between using the heat pump(if equipped) and or providing non-motive power to the motor/s to generate excess heat to pump into the battery, Tesla has an algorithm to determine what it is going to try and do. I am pretty sure that algorithm is a mystery to anyone but Tesla.

Mine is resistive.

That can happen as the car's BMS makes adjustments to the estimates based on past driving. For example, if you drive down a long hill, it may still take a few minutes for the car to adjust the display to reflect that energy gained from regen.

I noticed there was a delay in reporting coming down from higher elevations last year, but it always occurred while still on the downhill portions. I had been in a flat area for a bit before it jumped up this time.

This can give you lots of insights for lots of things...and it can also not tell you things you would like to know, haha...like how much energy if being used by various systems specifically for heat generation for preconditioning(while driving).

From what I've read, it sounds like the data logging has a high sample rate. This could be fun...

 

[CKwik240]

These massive errors are normal since it is just extrapolating and the API has no idea what your max charge will be. At very low SOC these estimates can be off by 100 miles due to rounding errors.

Very normal and nothing to do with BMS.

Understood. Thanks. The sudden drop in range occurred sometime around December I think. I also last supercharged in December and I hadn't observed this behavior before so I want sure if it was related or not.

 

[drtimhill]

This can give you lots of insights for lots of things...and it can also not tell you things you would like to know, haha...like how much energy if being used by various systems specifically for heat generation for preconditioning(while driving).

Except it does tell you exactly those things on the new Energy screen.

 

[AlanSubie4Life]

The sudden drop in range occurred sometime around December I think. I also last supercharged in December and I hadn't observed this behavior before so I want sure if it was related or not.

Yes. This is a different issue. Presumably the charge limit estimate in the car is also way off. Looking in the app not sure if it even still displays that erroneous info. Even in miles mode I can only have it show the % charge limit. Probably to avoid this problem.

 

[derotam]

Except it does tell you exactly those things on the new Energy screen.

I was talking about specifically about SMT that doesn't break out preconditioning energy...last I checked anyway.

 

[derotam]

For V3 supercvharges, my M3P 2021 always precondition to 48C (about 120F). Always, every time.
Before a change in the softare in september 2022 or so, the car did the same preheating (48C) independent of supercharger v1,v2,v3. But since, they have optomized it to not preheat more than needed. End temp for a not very short V3-session is about 58C.
(I have a lot of logs of this, via Scan My Tesla and teslalogger).

Preheating has two purposes, if you ask me. It shortens the charging time by mimimizing the internal resistance. But it also reduces lithium plating, which is the bad part of fast charging. So, when we see 48C it is on the right side of the limit that causes lithium plating. Probably some lithium plating occur anyway (probably at higher SOC levels), but the high temperature is the weapon against lithium plating.

That said, I always precondition completely if it is possible. A few yea back my car did start the prehetaing much later, like 30 mins or so in winter time, these days its about one hour before.

I definitely agree that Tesla has not optimized to prevent unnecessary preheating...at least from a charge rate standpoint. Even with 250kW chargers, 48C is unnecessary from a charge rate standpoint.

As for lithium plating issues, you wouldn't happen to have any references that would give any kind of data on lithium plating vs battery temperature would you? The other issue is that there is always a trade off right, you can go higher temperature and reduce lithium plating but you can also damage the electrolyte or cause other issues. Obviously Tesla has don't a lot of research in this matter and I would generally side with Tesla on the matter, except that they don't seem to be optimizing the preheating. Maybe 48C is the sweet spot for anything over a certain charge rate, but I would love to see some kind of data or white paper that shows something about battery temperature, charge rate, AND lithium plating.

 

[derotam]

Mine is resistive.

I noticed there was a delay in reporting coming down from higher elevations last year, but it always occurred while still on the downhill portions. I had been in a flat area for a bit before it jumped up this time.

From what I've read, it sounds like the data logging has a high sample rate. This could be fun...

Mine is resistive as well, 2018 LR RWD. In the beginning Tesla couldn't put non-motive power to the rear motor when going above about 30mph(I was able to show this with SMT data in the past), but they figured that out, you just can't see it in SMT...well it is applied in the data from the car and displayed in SMT in the power number, you just can't figure out that preconditioning power unless you can get in a controlled(or semi-controlled) testing environment. I haven't had the time or will to actually attempt to get it via SMT using cruise control and a specific stretch of road and turning on/off preconditioning.

As long as you have a good OBDII adapter, then the sample rate will be limited by the car's reporting rate. Different data is sent at different intervals.

 

[AAKEE]

I definitely agree that Tesla has not optimized to prevent unnecessary preheating...at least from a charge rate standpoint. Even with 250kW chargers, 48C is unnecessary from a charge rate standpoint.

Well, they did optimize it when that update fid bring lower preheat for ”slower” superchargers.

As for lithium plating issues, you wouldn't happen to have any references that would give any kind of data on lithium plating vs battery temperature would you? The other issue is that there is always a trade off right, you can go higher temperature and reduce lithium plating but you can also damage the electrolyte or cause other issues.

The high temperature is for a short time and will not cause any noticable damage, but the researchers have found that preheating the battery to quite high temperature ( 40C or more) is the thing that is needed to conquer lithium plating. Lithium plating itself is a quick way of killing lithium batteries, so it should be avoided, either by reducing the charging power or increasing the temperature.

Obviously Tesla has don't a lot of research in this matter and I would generally side with Tesla on the matter, except that they don't seem to be optimizing the preheating. Maybe 48C is the sweet spot for anything over a certain charge rate, but I would love to see some kind of data or white paper that shows something about battery temperature, charge rate, AND lithium plating.

Well, I have a few research reports on this, but not available here. From my own studies I found that 250KW (about 3C) would need at least 40 or 45C temp to avoid lithium plating at low SOC, and despite the C-rate reduces with increasing SOC.

I quickly gogled this report, but it is not the absolute best one for this subject.(I just briefly looked at it). At least we can see that temperatures up to 60C is benifical for limiting lithium plating.
Fast charhing at 30C is not going to sabe the battery.
Research about fast charging in general (no preheating) on panasonic nca show that they get hurt quite bad.
Read this report, The chapter about charhing/fast charging and lithium plating to se how much the battery life is shortened https://mediatum.ub.tum.de/doc/1355829/document.pdf#page85
I think its needed to read about how bad lithium plating is when supercharging, before starting to read about the mitigating of it.
From this we see that supercharging will cause a very much shorter battery life if we do nothing to prevent lithium plating.
(Theres more than one report of this, I just took this as an example, remembering the name of the report and knowing it is good. It actually covers the most thibgs we need to know, except not how to mitigate lithium plating.)

This is a saved pic in the phone, you can see that 1C only ( = 80kW ob a 80kWh batt) would need 25C to charge without issues.

 

[drtimhill]

I was talking about specifically about SMT that doesn't break out preconditioning energy...last I checked anyway.

If you look on the energy screen (Drive tab) you will see it shows the energy used by preconditioning (as well as other energy usage).

 

[derotam]

If you look on the energy screen (Drive tab) you will see it shows the energy used by preconditioning (as well as other energy usage).

Yes, and again, I was specifically ONLY referencing to the OBDII data via Scan My Tesla data.

 

[drtimhill]

Yes, and again, I was specifically ONLY referencing to the OBDII data via Scan My Tesla data.

Ah missed that my apologies