diy solar

diy solar

My Tesla batteries are dying...Saved them!

sed6

Knows just enough to be dangerous...
Joined
Apr 17, 2020
Messages
128
After some 15 months of being completely ignored my Tesla batteries gave me a big middle finger last night. Cells 1 & 2 are 3.5v and cells 4-6 are 4.0v. After discharging overnight to 21.4v Cells 1 & 2 are 3.1v and cells 4-6 are 3.8v an even bigger spread. Ugh. So I'm scrambling to figure out a solution and determine why those two cells went so far out.

Some solutions I'm considering.
a) Charging cells 1 & 2 with a hobby charger to get them up to voltage and then Installing a 5 amp 6s active balancer to keep them even.
b) Skip charging and see if the 5A balancer can bring them back to balance in a reasonable amount of time (anyone know the math on that?).
c) Consider the SimpBMS or other BMS solution. I'll only try this if #1 or #2 fails.

Some causes I'm thinking about.
a) My 6 Cell Cell Monitor I have plugged into each somehow drains cells 1 & 2?
b) I'm charging to 24.5v and discharging to 22.5v so that's about 85%-25%. Wonder if going to 100% would help?
c) These were used cells from a wrecked 70k mile car, I guess the cells could be reaching their end of life?

I'm open to questions, suggestions, solutions and I told you so's...
Thanks all!
 
What BMS do you currently use?

Was this abrupt, or did you just happen to check it, and it could have been going on for some time?

Charging single "cells" is a viable approach.

A single bad cell in a parallel group can kill the whole group. If you have a FLIR type thermal camera, the failed cells will typically be warmer as they are draining their "cellmates" and have higher current flowing into them. You might be able to detect this with a laser guided IR gun as well. If you find cells at notably higher temps, you can cut their fuse wire and take them out of the group. That will at least keep them from draining their "cellmates."

NCA cycle life is pretty low compared to NMC and LFP, but with the 3.75-4.08 cycle range AND the reduced current compared to rated, I wouldn't expect you're adversely impacting life. Charging to higher voltages is more stressful than discharging, so a reduction of the peak voltage for normal operation might help.

a) Doubt it. They typically pull from the whole pack.
b) A top balance at 4.20 is warranted as a troubleshooting measure.
 
What BMS do you currently use?

Was this abrupt, or did you just happen to check it, and it could have been going on for some time?

Charging single "cells" is a viable approach.

A single bad cell in a parallel group can kill the whole group. If you have a FLIR type thermal camera, the failed cells will typically be warmer as they are draining their "cellmates" and have higher current flowing into them. You might be able to detect this with a laser guided IR gun as well. If you find cells at notably higher temps, you can cut their fuse wire and take them out of the group. That will at least keep them from draining their "cellmates."

NCA cycle life is pretty low compared to NMC and LFP, but with the 3.75-4.08 cycle range AND the reduced current compared to rated, I wouldn't expect you're adversely impacting life. Charging to higher voltages is more stressful than discharging, so a reduction of the peak voltage for normal operation might help.

a) Doubt it. They typically pull from the whole pack.
b) A top balance at 4.20 is warranted as a troubleshooting measure.
Thanks. No BMS, hence the problem. I'll try a top balance.
 
After some 15 months of being completely ignored my Tesla batteries gave me a big middle finger last night. Cells 1 & 2 are 3.5v and cells 4-6 are 4.0v. After discharging overnight to 21.4v Cells 1 & 2 are 3.1v and cells 4-6 are 3.8v an even bigger spread. Ugh. So I'm scrambling to figure out a solution and determine why those two cells went so far out.

Some solutions I'm considering.
a) Charging cells 1 & 2 with a hobby charger to get them up to voltage and then Installing a 5 amp 6s active balancer to keep them even.
b) Skip charging and see if the 5A balancer can bring them back to balance in a reasonable amount of time (anyone know the math on that?).
c) Consider the SimpBMS or other BMS solution. I'll only try this if #1 or #2 fails.

Some causes I'm thinking about.
a) My 6 Cell Cell Monitor I have plugged into each somehow drains cells 1 & 2?
b) I'm charging to 24.5v and discharging to 22.5v so that's about 85%-25%. Wonder if going to 100% would help?
c) These were used cells from a wrecked 70k mile car, I guess the cells could be reaching their end of life?

I'm open to questions, suggestions, solutions and I told you so's...
Thanks all!
So… the Tesla batteries are not on Warranty?
 
Well good news, this cheap active balancer seems to be doing the trick. After 48 hours the cells have gone from about .5v difference between highest and lowest to about .2v difference. I figure another couple of days and they'll be perfect.

20211208_092928.jpg

As to the cause? I was right to suspect the CellMeter8. The instructions say to remove after balancing and that the meter does indeed draw from cells one and two.

"Please pull out of the CellMeter8 port in time to complete the balance of the lithium battery.If thelithium battery is balanced after the completion of the long connection on the CellMeter8 port,Becausethe system power source is from the 1S and the 2S core of the lithium battery,So the standby powerconsumption of the system will lead to the 1S and the 2S of the lithium battery,The power of thelithium battery will not be in balance."

I'll do a couple of charge discharge cycles on the batts to make sure they stay in balance but I'm pretty confident this will fix it. Going forward I'll leave the cell meters unplugged and use them only occasionally to check for balance. As a bonus I have a cool cell balancer I can use on my dozens of hobby lipos. ?
 
Well good news, this cheap active balancer seems to be doing the trick. After 48 hours the cells have gone from about .5v difference between highest and lowest to about .2v difference. I figure another couple of days and they'll be perfect.

As to the cause? I was right to suspect the CellMeter8. The instructions say to remove after balancing and that the meter does indeed draw from cells one and two.

"Please pull out of the CellMeter8 port in time to complete the balance of the lithium battery.If thelithium battery is balanced after the completion of the long connection on the CellMeter8 port,Becausethe system power source is from the 1S and the 2S core of the lithium battery,So the standby powerconsumption of the system will lead to the 1S and the 2S of the lithium battery,The power of thelithium battery will not be in balance."

I'll do a couple of charge discharge cycles on the batts to make sure they stay in balance but I'm pretty confident this will fix it. Going forward I'll leave the cell meters unplugged and use them only occasionally to check for balance. As a bonus I have a cool cell balancer I can use on my dozens of hobby lipos. ?

Yikes. I guess if it doesn't do balancing, it kinda makes sense as it's only needed if you have 2 or more cells, and if you only have two, you can't power from the 1st and 8th cells...

I had it in my head that you were using something like:


It has a balancer function.

At least one other fellow used it on a 6S Tesla module.

Good plan on the cycle tests. I'd really scrutinize cells 1 and 2. If they've been operated at the low end of the range, they may have lost capacity vs 3-6.

You're still going to get a BMS, right?
 
Yikes. I guess if it doesn't do balancing, it kinda makes sense as it's only needed if you have 2 or more cells, and if you only have two, you can't power from the 1st and 8th cells...

I had it in my head that you were using something like:


It has a balancer function.

At least one other fellow used it on a 6S Tesla module.

Good plan on the cycle tests. I'd really scrutinize cells 1 and 2. If they've been operated at the low end of the range, they may have lost capacity vs 3-6.

You're still going to get a BMS, right?
The CellMeter8 has a balance function. But like the one you linked to you need to go tell it to balance. It doesn't do it all the time or automatically. It like mine will probably suck down cells 1&2.

I'll keep an eye on 1&2 but they are probably fine. My charge discharge current is low enough that the cells were never really stressed. It became unbalanced from a small constant draw, not from a stressful charge discharge schedule There's a big difference between the two, so I think these cells will be just fine.

As for a BMS, probably not. Again they just don't see enough cycling to go out of balance. Cells 3-6 were all within .003v (4.017-4.020) after 15 months of use! That makes me continue to believe a BMS is unnecessary for my situation.
 
Last edited:
As for a BMS, probably not. Again they just don't see enough cycling to go out of balance. Cells 3-6 were all within .003v (4.017-4.020) after 15 months of use! That makes me continue to believe a BMS is unnecessary for my situation.

This is NEVER true. What happens when your charge controller fails and feeds unrestricted full panel voltage and current to your battery? This has happened. What will be the outcome?

If you're just risking your life, that's your choice. If there are other living things in the dwelling, ask yourself if you're okay risking those lives. Better yet, ask them.
 
running this chemistry battery without a safety disconnect is asking for trouble.

when used in a vehicle part of safety disconnect is literally an explosive device to separate the connection instantaneously

overcharging this chemistry is much more dangerous to human life than lifepo4, warranting a BMS in my humble opinion

no disrespect, just, hearing "NCA battery bank with issues and no BMS" makes me stare at ceiling in dark room late at night unable to sleep 0_0
 
Well good news, this cheap active balancer seems to be doing the trick. After 48 hours the cells have gone from about .5v difference between highest and lowest to about .2v difference. I figure another couple of days and they'll be perfect.

View attachment 75012

As to the cause? I was right to suspect the CellMeter8. The instructions say to remove after balancing and that the meter does indeed draw from cells one and two.

"Please pull out of the CellMeter8 port in time to complete the balance of the lithium battery.If thelithium battery is balanced after the completion of the long connection on the CellMeter8 port,Becausethe system power source is from the 1S and the 2S core of the lithium battery,So the standby powerconsumption of the system will lead to the 1S and the 2S of the lithium battery,The power of thelithium battery will not be in balance."

I'll do a couple of charge discharge cycles on the batts to make sure they stay in balance but I'm pretty confident this will fix it. Going forward I'll leave the cell meters unplugged and use them only occasionally to check for balance. As a bonus I have a cool cell balancer I can use on my dozens of hobby lipos. ?
thank you for documenting this successful case of the capacitive active balancer operating ? redistributing the energy is much more appealing than simply burning it off as heat (to me)

gently to encourage you to add BMS proactively before additional issues arise. admittedly i am inexperienced with NCA DIY builds, and also not familiar with the system you're working with, so my advice might come from a place of ignorance.
 
I hear ya'll but what happens when the BMS fails? What happens if the fire department doesn't show up? Life is full of risk. Show me an example of someone burning down their house with a Tesla battery and I'll show you someone who still doesn't care. Just because Will is afraid of LiPo's doesn't mean we all have to be.
 
I hear ya'll but what happens when the BMS fails? What happens if the fire department doesn't show up?

You don't "year ya'll." You don't care what we're saying. You're just providing gross logical fallacies to support your feelings/bias.

Life is full of risk. Show me an example of someone burning down their house with a Tesla battery and I'll show you someone who still doesn't care.

The difference is the first someone learned a hard lesson. The second someone is a moron because they see what can happen and choose otherwise.

Just because Will is afraid of LiPo's doesn't mean we all have to be.

You don't have LiPo cells, so what does that have to do with anything?

@Will Prowse isn't relevant to your position. Every lithium cell manufacturer, including the ones in your Tesla pack, insists a BMS is needed regardless of Lithium chemistry, but you know better? You're deluded.

Your user Custom Title is wrong. "Knows just enough to be dangerous..." should read, "Knows better, but still chooses to be dangerous."

I pray you live alone and are not placing others in jeopardy.
 
I hear ya'll but what happens when the BMS fails? What happens if the fire department doesn't show up? Life is full of risk. Show me an example of someone burning down their house with a Tesla battery and I'll show you someone who still doesn't care. Just because Will is afraid of LiPo's doesn't mean we all have to be.
JBD BMS is ~130 usd and Military and Medical Grade Components meets my personal definition of what Medical and Military Grade means, as compared to Hobby Grade. there’s 12v 24v and 48v versions. I use these because I want to survive, live, thrive, when my mistake is revealed to me late in the game.

Installing a BMS and respecting Lithium are fundamental physical realities. Will has nothing to do with this.

You could kill someone if you’re not the only person in that building. Hopefully you don’t need anyone to tell you that’s worth changing.

No shame, no disrespect, just.. You are Off Base about safety and in the gentlest way, education. It’s your decision to play fire man. Please only play that game with yourself?

To the degree it’s unsupervised, a BMS is Even More Justified In My Humble Opinion.
 
Last edited:
Citation please.
heresay from claims about other devices it is integrated into at an OEM level, not DIY jobs.

it may not meet everyone’s definition of Medical Grade or Military Grade. i don’t really care about those arbitrary labels, because personal individuals have Military and Medical departments. ultimately any plain noun is meaningless; Specific Standards such as Specific ISO standards, NEC, etc is what I consider to be more helpful.

after having done maintenance on my JBD BMS, I looked up the MOSFET. according to the specifications available online, the individual FETs are rated quite generously to give margin in my opinion as an amateur.

capability to tolerate large surge current is specifically the aspect i am thinking about. since OP casually suggested the BMS may fail.

for convenience, here’s the link to how to learn more about the Real Components

Post in thread 'JBD / Overkill BMS 120A 4S Lifepo4 in series / parallel'
https://diysolarforum.com/threads/j...-lifepo4-in-series-parallel.19228/post-332656

TL;DR military and medical are unqualified subjective terms that anyone may use when they feel like it. no claims of specific standard compliance should be inferred from this claim.

JBD 120A 12V model meets my personal definition of what it means to functionally be military and medical grade.

DALY is neither.
 
a MOSFET’s ability to continue to safely pass Amps and still Be Able To Disconnect, is Temperature. whatever BMS, Dissipation of Heat is Critical.

using a BMS at its rated Amp rating can cause Heat to Build. cheap ones are known for this. some will not heat up much when passing their Rated Amps. this is desirable for safety and reliability.

MOSFET when very hot and many Amps, can CLOSE! like a short. bad.

this is why heat sink heat spreader is so important in MOSFET BMS
 
MOSFET often have a max pulse amps at all rating. also a constant use amps rating. usually the max one is like >5x the constant one. and often, the max one heats up the MOSFET body so fast that it can only be sustained for less than a blink of an eye in some cases. if the MOSFET overheats it can Fail Short Closed. undesired. result: design goal dissipate MOSFET heat keep temperature stable and low to prevent Fail Short Closed.

40V 354A is the individual rating i could find for the actual MOSFET on the JBD 12V 120A BMS here. without a specific label, i have to assume that 354A is the pulse/max rating and definitely Not the Constant Amps Rating.

there are seven of them side by side on my BMS. if the current were perfectly distributed between all the components, this suggests a possible 354*7 Amperes or 2,478 Amps. Usually perfect current distribution is not the case so 2,478 * 0.6 for up to 40% prevention headroom. that’s 1,486 Amps pulse tolerance implied. Is any of this tested by me? No. If anyone is aware of a JBD BMS that has failed shorted I really want to know so as to learn about the conditions in which it occurred.

tesla packs can often do Juicy Pulse Amps, so that’s kind of the motivation in this rambling

good luck, hope this helps somewhat
 
Back
Top