Battery imbalance cell reading changes

[5JWX]

I have a 2014 Model S with a refurbished 2020 85 battery pack. Car won’t drive or charge, shows numerous errors
Using scan my Tesla I have a big imbalance issue, ranges in seconds from 1000 to 2245mV
The bit I don’t understand is the voltage in modules 8 & 10 fluctuate between 3.6v to 1.6. Guessing I have a coolant leak to have effected 2 modules?
Can anyone confirm the module numbering system?
Anyone had this before, everything I’ve read online has more or less static low voltage not fluctuating like mine.
Thanks

 

[5JWX]

Update
I have a cell difference 2.5v but the reading are very erratic & does this on modules 8 & 10. It will show all cells approx 3.6v then 2 cells from module 8 & 2 from 10 will suddenly drop to 1.6v then back up to 3.6v the faulty cells are in the same position on scan my Tesla; 1st row top right & 2nd row bottom left. They change at t tree exact same time.
Car is a 2014 model S with reconditioned pack fitted 2.5 years ago by Tesla.
I’ve stripped down the pack expecting to find water/ coolant but cannot visually see anything.
I have isolated each module & checked the voltages. From memory 12 are all 22.76, the lowest reading was 22.74, some 22.75. Which were all much closer than I had expected. My knowledge suggests all my modules are good?

I now assume I have a fault with the BMS circuit boards but how can 2 go down at the same time? Or could it be a BMS wiring fault within the pack, or fault the pack circuit board situated at the rear of the pack?
Has anyone come across this?

If I don’t get any response my next plan is to move the modules around in the pack, & the BMS circuit boards, refit the pack to the car & see if the fault has moved.
Unless someone has a better idea.
All input welcome

Thanks

 

[danielk04]

Okay 5J,
It sounds like you're moving in the right direction.
Pull modules 9 and 11 out of the pack.
Measure voltage on 6 individual bricks on each module.
Do some bricks measure 1 or 2 volts?
Yes? Module requires replacement.
No? 3 volts or more? Lucky you. Most likely BMS board on the module failure, requires board replacement.

 

[5JWX]

Okay 5J,
It sounds like you're moving in the right direction.
Pull modules 9 and 11 out of the pack.
Measure voltage on 6 individual bricks on each module.
Do some bricks measure 1 or 2 volts?
Yes? Module requires replacement.
No? 3 volts or more? Lucky you. Most likely BMS board on the module failure, requires board replacement.

Thanks.
Does the fact the whole module voltage is a consistent 22.76v not mean the module is ok then?

Okay 5J,
It sounds like you're moving in the right direction.
Pull modules 9 and 11 out of the pack.
Measure voltage on 6 individual bricks on each module.
Do some bricks measure 1 or 2 volts?
Yes? Module requires replacement.
No? 3 volts or more? Lucky you. Most likely BMS board on the module failure, requires board replacement.

Thanks for the info, I assumed if the module are all equal the batteries were ok but your right if one or two cells are out it wouldn’t show across the entire pack.
I’ll check the individual cells over the next few days.

Scan my Tesla shows issues with 8 & 10, you say check modules 9 & 11, is that correct, if so I didn’t realise that, I’m working on the modules being numbered as per attachment but not 100% sure it’s correct.
Thanks

 

[danielk04]

I was trying to make out the numbering on your somewhat fuzzy scan my Tesla attachment photo. You are probably correct with modules 8 and 10. The important thing is that you get what I'm driving at.
And the layout attachment above shows correctly. You know what you're doing. You'll be fine.
Let us know what you find, battery trouble or circuit card trouble.

 

[Electric700]

Thanks.
Does the fact the whole module voltage is a consistent 22.76v not mean the module is ok then?

Thanks for the info, I assumed if the module are all equal the batteries were ok but your right if one or two cells are out it wouldn’t show across the entire pack.
I’ll check the individual cells over the next few days.

Scan my Tesla shows issues with 8 & 10, you say check modules 9 & 11, is that correct, if so I didn’t realise that, I’m working on the modules being numbered as per attachment but not 100% sure it’s correct.
Thanks

If you still have the issue, I think you should ask for a low cost Tesla repair option. I understand that they are offering at least a 4-year warranty with pack refurbishments / replacements. These third-party companies can help also with main battery repairs:

www.057tech.com (Offers a main battery warranty)

www.recell-ev.com

 

[cloneguru]

Was the root cause of the "suddenly drop to 1.6v then back up to 3.6v" every determined?

 

[brainhouston]

Most likely sense wire or bmb failure
See here:

Condensation Failure Inside Battery Pack (BMS_F107 code)

Sharing some failure mode analysis with others who has been inside the battery pack (NOTE : HV can be deadly dangerous so absolutely need to be familiar with all HV protection precautions) This is @mr_hyde car with his original post here and here. Car is 2014 MS P85D 67k miles. In Seattle...

teslamotorsclub.com

 

[cloneguru]

I'm having very similar f107 issues on a 2012 P85. I verified corrosion on C27 (BMB failure) and replaced the BMB. The delta between cell#3 and cell#4 dropped from ~80mv to ~60mv after the BMB swap however the periodic drop from ~3v to ~1v persisted on the same module even after the full BMB replacement.

I'll be replacing the entire module next week and will continue my testing on the current/failing module outside of the car. I'm guessing I'll find a bad sense wire on the bottom side (we checked the top side pretty well last time the module was out but do not recall checking the bottom side.) If replacing the battery module (including the BMB) does not resolve the issue, I'll swap the BMS board.

This image is from the old BMB that was replaced, however the issue persisted after replacing the pictured BMB

 

[cloneguru]

I'm having very similar f107 issues on a 2012 P85. I verified corrosion on C27 (BMB failure) and replaced the BMB. The delta between cell#3 and cell#4 dropped from ~80mv to ~60mv after the BMB swap however the periodic drop from ~3v to ~1v persisted on the same module even after the full BMB replacement.

I'll be replacing the entire module next week and will continue my testing on the current/failing module outside of the car. I'm guessing I'll find a bad sense wire on the bottom side (we checked the top side pretty well last time the module was out but do not recall checking the bottom side.) If replacing the battery module (including the BMB) does not resolve the issue, I'll swap the BMS board.

This image is from the old BMB that was replaced, however the issue persisted after replacing the pictured BMB

Not only was the BMB bad but there was also a sense wire that had come off. It seems oddly coincidental the C27 corrosion issue to be on the same module as the loose/detached sense wire. Have others seen corrosion and loose/detached sense wires? Does one cause the other?

FYI: The entire module was replaced and the vehicle is back on the road. It happily charges, supercharges, drives, etc.

 

[brainhouston]

Not only was the BMB bad but there was also a sense wire that had come off. It seems oddly coincidental the C27 corrosion issue to be on the same module as the loose/detached sense wire. Have others seen corrosion and loose/detached sense wires? Does one cause the other?

FYI: The entire module was replaced and the vehicle is back on the road. It happily charges, supercharges, drives, etc.

Keep us updated how it works out long term, some ppl claim module swap doesn't work long term...
Did u match the module in any way? ie CAC, donor year or mileage...

 

[cloneguru]

Keep us updated how it works out long term, some ppl claim module swap doesn't work long term...
Did u match the module in any way? ie CAC, donor year or mileage...

Can you reference the "some ppl" claims? It's not that I don't believe you... I'm curious myself. There's so much misinformation out there, it's hard to determine what's real vs bad assumptions, etc.

Matched BMB revision (physically looked the same). Both original and donor BMB were the soldered-on non-modular BMB with wire not ribbon sense wires 444 count 250Ah 5.2 kWh modules.

Both were within a year manufacturer date.

Brought the donor voltage to within ~8mv ( thought I was dead on however I did not account for temperature delta between the pack readings and where I was preparing the donor module )

I have no idea what mileage the donor module had.

After sitting for ~24hrs... my highest delta module is 7mv and pack delta (highest cell - lowest cell) is 20mv. The module I replaced has a 3mv delta and blends in pretty well.

The car charges, supercharges, and drives without issue.

 

[brainhouston]

Can you reference the "some ppl" claims? It's not that I don't believe you... I'm curious myself. There's so much misinformation out there, it's hard to determine what's real vs bad assumptions, etc.

I don't believe those claims either, thats why i ask...
"some ppl" are all over the TMC/fb 029 group...
most advisors of that were wk057/recell...

I'm all up for module replacements working out!

For CAC matching, i meant like measuring CAC of urs n new one, full charge/discharge on bench..

 

[wk057]

Module swaps never work. At one point half the cars in the 057 shop with battery issues were there because some other company/person tried a module swap "repair." I eventually just stopped accepting them because people who don't know what they're doing when opening up the battery make the work infinitely more difficult to follow after, since they'll use improper sealants, strip bolts, invite moisture for additional corrosion, etc. My "favorite" was a battery that had been resealed with bathroom silicone that had some color additive that was conductive enough to cause a HV isolation issue.

Wasn't worth it, so started refusing to service such vehicles.

If you swap a module, sure, it might drive now. Getting that far is usually as simple as matching voltage to another module and popping it in.

But it will leave you stranded at some point. I'm not saying it's likely. It's a 100% chance. Might not be today, maybe not tomorrow. But every mile driven is a dice roll and eventually you're going to crap out. (The only way it won't is if you just don't drive it or charge it. lol... even then eventually the BMS will still probably sniff out your attempt to deceive it eventually.)

There's more than just CAC and voltage that need to match for the BMS to be happy with a pack, and it's so improbable as to be effectively impossible to get everything from a module swap to match well enough to be even a few month repair, let alone a life-of-the-car duration solution. Combine that with poor workmanship of most anyone who's going to try and reseal a pack DIY, and you've beyond guaranteed a failure. (Did you know a lot of the packs that have had corrosion related failures are early packs where Tesla did field service opening and resealing them to replace contactors? And that's Tesla service doing it with proper tools and instructions.)

Module swaps and/or the error reset nonsense seem to be only useful for scammers before flipping cars with bad batteries. Should be illegal.

 

[cloneguru]

Module swaps never work. At one point half the cars in the 057 shop with battery issues were there because some other company/person tried a module swap "repair." I eventually just stopped accepting them because people who don't know what they're doing when opening up the battery make the work infinitely more difficult to follow after, since they'll use improper sealants, strip bolts, invite moisture for additional corrosion, etc. My "favorite" was a battery that had been resealed with bathroom silicone that had some color additive that was conductive enough to cause a HV isolation issue.

Wasn't worth it, so started refusing to service such vehicles.

If you swap a module, sure, it might drive now. Getting that far is usually as simple as matching voltage to another module and popping it in.

But it will leave you stranded at some point. I'm not saying it's likely. It's a 100% chance. Might not be today, maybe not tomorrow. But every mile driven is a dice roll and eventually you're going to crap out. (The only way it won't is if you just don't drive it or charge it. lol... even then eventually the BMS will still probably sniff out your attempt to deceive it eventually.)

There's more than just CAC and voltage that need to match for the BMS to be happy with a pack, and it's so improbable as to be effectively impossible to get everything from a module swap to match well enough to be even a few month repair, let alone a life-of-the-car duration solution. Combine that with poor workmanship of most anyone who's going to try and reseal a pack DIY, and you've beyond guaranteed a failure. (Did you know a lot of the packs that have had corrosion related failures are early packs where Tesla did field service opening and resealing them to replace contactors? And that's Tesla service doing it with proper tools and instructions.)

Module swaps and/or the error reset nonsense seem to be only useful for scammers before flipping cars with bad batteries. Should be illegal.

LOL @ Bathroom Silicone.... that'd be a fun surprise when investigating a pack.

You seem pretty confident with your "100% chance".

What separates a module swap with "100%" chance of failure from a Tesla/057/ReCell remanufactured pack if proper sealant is used?

I very much appreciate hearing the "what not to do" posts however if it comes off as a scare tactic when the post has no mention of the "correct way" or any constructive guidance.

 

[wk057]

I've directly sourced data on nearly 50 vehicles where third parties have attempted module swaps to fix Model S's, and data from a... source... on an additional ~110. The long term success rate of those is exactly as expected: 0%. There was a contender for a little while that kept that number at about 0.6% success, and I was pulling for it... but it was clearly trending towards failure from day 1 and eventually failed after 9 months of light use (about 4000 miles over 9 months). Oh, and that was my own best effort at module matching knowing everything I know about the BMS and what it expects.

"You did a module swap? OMG!" Yes, of course I've tried all sorts of ways to do module swaps over the years (not on customer vehicles, of course).... being able to actually swap modules for repairs would be tons of profit in repairs. But it's sadly not possible. Even Tesla attempted this with early refurb packs and couldn't make it work.

That leaves two and only two correct ways to repair a pack with a bad module:

• Replace the entire pack with a good pack; or
• Downgrade the pack by removing the bad module(s) and doing the other necessary work to make the pack into a standard configuration with the lower module count.

Those are your only legitimate options for a long term repair. This is what Tesla does, and to the best of my knowledge what @Recell does as well. No one with any sense is mixing modules. And sadly, the complexity of the latter option puts it outside the reach of a DIY project anyway.

There's also a lot more to getting the pack ready for re-use after doing an internal change than just using the correct sealant. The correct sealant isn't going to do you any good if you seal moisture inside, damage an o-ring on a coolant line, leave any debris in the pack, or any of dozens of other ways a problem could be caused. Opening and reusing the top cover of a pack is more of an art, too. You definitely don't want to be relying solely on sealant vs a tight fit. There's also o-rings around some of the bolts that tend to get damaged during removal/install that need inspection/replacement, tons of seals and pieces that tend to get pinched and useless when re-closing a pack if you don't know what you're doing, etc etc etc. Far too extensive of a list to realistically get into.

Suffice it to say there's no DIYer that's going to catch and mitigate all of the potential issues. Even Tesla fails at some of the edge cases in their refurbished packs occasionally, especially anything done at the service center level. I'm not perfect either, and have had to do a couple of redos over the years.

 

[cloneguru]

T

I've directly sourced data on nearly 50 vehicles where third parties have attempted module swaps to fix Model S's, and data from a... source... on an additional ~110. The long term success rate of those is exactly as expected: 0%. There was a contender for a little while that kept that number at about 0.6% success, and I was pulling for it... but it was clearly trending towards failure from day 1 and eventually failed after 9 months of light use (about 4000 miles over 9 months). Oh, and that was my own best effort at module matching knowing everything I know about the BMS and what it expects.

"You did a module swap? OMG!" Yes, of course I've tried all sorts of ways to do module swaps over the years (not on customer vehicles, of course).... being able to actually swap modules for repairs would be tons of profit in repairs. But it's sadly not possible. Even Tesla attempted this with early refurb packs and couldn't make it work.

That leaves two and only two correct ways to repair a pack with a bad module:

• Replace the entire pack with a good pack; or
• Downgrade the pack by removing the bad module(s) and doing the other necessary work to make the pack into a standard configuration with the lower module count.

Those are your only legitimate options for a long term repair. This is what Tesla does, and to the best of my knowledge what @Recell does as well. No one with any sense is mixing modules. And sadly, the complexity of the latter option puts it outside the reach of a DIY project anyway.

There's also a lot more to getting the pack ready for re-use after doing an internal change than just using the correct sealant. The correct sealant isn't going to do you any good if you seal moisture inside, damage an o-ring on a coolant line, leave any debris in the pack, or any of dozens of other ways a problem could be caused. Opening and reusing the top cover of a pack is more of an art, too. You definitely don't want to be relying solely on sealant vs a tight fit. There's also o-rings around some of the bolts that tend to get damaged during removal/install that need inspection/replacement, tons of seals and pieces that tend to get pinched and useless when re-closing a pack if you don't know what you're doing, etc etc etc. Far too extensive of a list to realistically get into.

Suffice it to say there's no DIYer that's going to catch and mitigate all of the potential issues. Even Tesla fails at some of the edge cases in their refurbished packs occasionally, especially anything done at the service center level. I'm not perfect either, and have had to do a couple of redos over the years.

thank you. This response was significantly more constructive than the last.

 

[Recell]

we never swap modules in any of our certified packs. it's not a judgement on swapping methodology per se (see note below), and certainly not a scare or sales tactic, we just don't have to with our remanufactured packs. all our certified packs contain factory-original modules, sitting side-by-side, in the same battery case that they left the factory in. and if they don't all meet spec, the pack doesn't get certified. simple.

fwiw: our data shows that when done at scale, swapping can last up to 24-30 months. but as @wk057 notes, sooner or later they all end up back on a repair rack... we see lots of the first-gen Tesla remans come through that used module swapping and capacity matching to do the job and it's the same story, back on the rack...

we often say: the problem isn't getting a pack back on the road, it's keeping it back on the road.

 

[nicog]

Hello.

first of all marry Christmas and happy new year to you.

@wk057
I have red many threads you wrote about Tesla HV batteries and I know you are expert in this segment.
because of this I have a question.

My 2015 70D model S 135k KM began for about a month to increase imbalance. the problem according to SMT has module 4's last brick. The imbalance varies between 18-30 mv.
I already charged it fully and discharged to 15-20% 2 times, but no decrease in imbalance.
I'm now charging it with 5 amp on 220v socket to help to balance it. but until now I see no improvement.
The 100% range dropped from 353km to 345km in this one month period.
Now dais it is colder here: 1 to 10 Celsius.
How can I balance it or check if the pack begins to failure?

Thank you in advance.

Regards

Nico

 

[Electric700]

Tesla told me that the vehicles are supposed to automatically be balancing the batteries, just charge to 80% - 90%. They said that they don’t see vehicles in the service center with imbalanced batteries. May be your case is so minor that the BMS is ignoring it for now?