Best way to rescue over-discharged cells?

How do you store and manage your electricity?
Post Reply
HenryF
Noobie
Posts: 3
Joined: Wed, 02 Jun 2021, 19:29
Real Name: Henry Freiburg

Best way to rescue over-discharged cells?

Post by HenryF »

Hello, I am the new owner of "Ohmboy", a 1990 Suzuki Carry with 450 Headway 40160 LiFePO4 Batteries in packs of 10 and Batrium BMS. Not having a SOC% meter and foolishly testing the maximum range, I discharged the LiFePo batteries to the point that 1/3 of the 45 packs are under 2.5volts. The BMS modules are in error and it will no longer allow charging. I understand the potential damage I have done to the batteries... Please learn from my mistake!

I found a "Manual Charge Utility" on the BMS, but I get error: "UMonHub Host, the specified channel is not open". I have not been able to get through to Batrium for support yet...

I was able to bring three packs back to life by connecting them to an old motorcycle battery. 120A max from the battery, across 30 cells @ 80A max charge each = 2400A max charge / 120A = 8% charge "rate" which I read is about right for a very slow charge. Connected across three packs in series splits the 12V charging battery into 4v charge, about right... After 30 min the cell voltages went from about 2.2V to 2.6V and the BMS errors cleared (Yeah!). Problem is, the motorcycle battery voltage dropped to less than 8V and it would not charge a second string of 3, so it will be a long slow process...

Any ideas on how to do this better/quicker? I'll try a bigger/fresher battery tomorrow.

Should I dare to use my welder as a DC power supply? IF it does about 50A at 10V at its lowest setting, it would theoretically be safe across 3 packs (similar calculation as above)...

Does anybody have a manual/adjustable 240v charger I can borrow?

Thanks for your ideas/advice.

(of course, the batteries are 9 years old, this could be my excuse to get a fresh set... I only got 80km from 22Kwh in a 1000kg van)

-Henry
Neutral Bay/Sydney
0423 298 841
User avatar
jonescg
Senior Member
Posts: 3854
Joined: Thu, 21 Jan 2010, 23:05
Real Name: Chris Jones
Location: Perth, WA.

Re: Best way to rescue over-discharged cells?

Post by jonescg »

Hi Henry, glad to see the most bad-ass bongo van in Sydney has new home. I hope Mark is doing well - we've not hear from him in ages!

As for the batteries, yeah that's not good for them hey. But so long as the LFP cells were not taken below about 1.5 volts they might be recoverable. Slowly slowly is the key. You might need to just set up a few power supplies to bring them up to 3.5 V. Go down to Jaycar and buy a couple of 30 V, 5 amp constant current supplies and set them to top up the low cells.

Of course, if they went negative, there's no bring them back.
AEVA National Secretary, WA branch chair.
User avatar
weber
Site Admin
Posts: 2826
Joined: Fri, 23 Jan 2009, 17:27
Real Name: Dave Keenan
Location: Brisbane
Contact:

Re: Best way to rescue over-discharged cells?

Post by weber »

Chris is right. The problem is that if they sit for any length of time below 2.0 volts, the copper current collectors start to dissolve in the electrolyte. Then when you recharge, it electroplates back out again. But it doesn't necessarily go back where it came from. It tends to grow spikes (called dendrites) that can grow across from one plate to the other and cause an internal short and a fire, unless the current is very low. So until you get the rested voltage back above 3.0 volts you shouldn't charge at more than about 0.01C. The lower the better. And do it in a place where you can deal with it (e.g. hose it with water) if a cell starts to smoke. And don't leave it unattended.
One of the fathers of MeXy the electric MX-5, along with Coulomb and Newton (Jeff Owen).
T1 Terry
Senior Member
Posts: 1287
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Best way to rescue over-discharged cells?

Post by T1 Terry »

If you get in touch with Trev or Marcy @ https://www.drivebynature.com/ and buy a few single cell chargers. These will slow charge one cell at a time. When you get each cell to hold 3v after a few hrs the normal charger will take over. Do not attempt to fast charge the cells until they are above 3v at this will turn them into paper weights. A cell that has been dragged below 0v will remain at 0v, but if it even holds 0.? whatever, very slow charging will bring it back to life.
If they start to smoke, plastic bags full of sand is the safest method to avoid burning the vehicle to the ground. If it gets hot enough, the plastic will melt and smother the fire. LiFeP04 chemistry doesn't make oxygen within the cell so the sand seals off any path for oxygen to feed a fire and the cells will just cool down after a while and stop gassing.
Slow recharge until the cell will hold 3v is the key to LFP cell recovery, attempt to fast charge will destroy them in minutes/hrs, I have $10, 000 worth of paper weights here to verify just what happens if they are fast charged.

T1 Terry
Green but want to learn
HenryF
Noobie
Posts: 3
Joined: Wed, 02 Jun 2021, 19:29
Real Name: Henry Freiburg

Re: Best way to rescue over-discharged cells?

Post by HenryF »

Thanks for the advice... I attached the fully-charged (but old) motorcycle battery this morning. Big spark, wires getting hot... I disconnected quick-smart. I think it was OK yesterday due to the battery being low on charge. I have ordered a variable PS to get control over the process. I have also asked my daughter to beg some lab power supplies from school, they look about right to do the job.
When I sampled the pack voltages, the lowest was 1.7V, so I think I will be okay... just a slow process to fix.
HenryF
Noobie
Posts: 3
Joined: Wed, 02 Jun 2021, 19:29
Real Name: Henry Freiburg

Re: Best way to rescue over-discharged cells?

Post by HenryF »

Update: My daughter borrowed 4 lab power supplies from her science teacher (North Sydney Girls). Where three batteries in a row were under voltage, I set 12v. 2 batteries got 8v, 1 battery, 4v. I blew the fuse on the first attempt, then added a 1-ohm, 10w resistor. It went up to 250-degrees, so I attached a big heat sink (plus fan), which held the temp at 50-degrees. I also tried using two 2-ohm resistors in parallel (20w capacity), which ran much cooler. I ran three power supplies at once, moving them between batteries every 10 - 20 minutes, when voltage got above 2.5. Took about 2 hours. After troubleshooting a loose connection on the Batrium cell monitors, automatic charging worked (though it strangely cycled between 1 and 5 amps measured at the outlet, stabilizing and going to the predicted 10 amps after a few hours. After manual charging, the batteries were then all over the place (2.5 - 3.2). I don't know how (no bypass was happening AFAIK), but by the time the packs were getting close to 3.3v they were all very close in voltage... Ohmboy is back on the road!
Attachments
charging.jpg
charging.jpg (144.91 KiB) Viewed 79 times
User avatar
coulomb
Site Admin
Posts: 4749
Joined: Thu, 22 Jan 2009, 20:32
Real Name: Mike Van Emmerik
Location: Brisbane
Contact:

Re: Best way to rescue over-discharged cells?

Post by coulomb »

HenryF wrote: Wed, 09 Jun 2021, 12:01 I blew the fuse on the first attempt, then added a 1-ohm, 10w resistor.
What you really needed was lab power supplies with two knobs (voltage and current limits), but beggars can't be choosers.
After manual charging, the batteries were then all over the place (2.5 - 3.2).
So they are in dire need of balancing.
I don't know how (no bypass was happening AFAIK), but by the time the packs were getting close to 3.3v they were all very close in voltage...
They're probably still in dire need of balancing, but they're at one of the plateaus in the LFP voltage versus SOC curve. One is at 3.29 VPC (45-60% SOC), another is at 3.33 VPC (85-90%). Hopefully your BMS will balance them over time. If not, you may need to get some more power supplies and do more manual balancing.

Edit: another possibility is that the capacities vary widely, but the BMS has been doing a pretty good job of top balancing. The way to tell if that's the case is to measure the voltages when the pack is above the 3.33 V plateau (i.e. above 90% SOC), so that the cell voltage once again indicates SOC fairly well.
MG ZS EV 2021 April 2021.
Nissan Leaf 2012 with new battery May 2019.
5650 W solar, 2xPIP-4048MS inverters, 16 kWh battery.
1.4 kW solar with 1.2 kW Latronics inverter and FIT.
Patching PIP-4048/5048 inverter-chargers.
T1 Terry
Senior Member
Posts: 1287
Joined: Thu, 30 Sep 2010, 20:11
Real Name: Terry Covill
Location: Mannum SA

Re: Best way to rescue over-discharged cells?

Post by T1 Terry »

HenryF wrote: Wed, 09 Jun 2021, 12:01 Update: My daughter borrowed 4 lab power supplies from her science teacher (North Sydney Girls). Where three batteries in a row were under voltage, I set 12v. 2 batteries got 8v, 1 battery, 4v. I blew the fuse on the first attempt, then added a 1-ohm, 10w resistor. It went up to 250-degrees, so I attached a big heat sink (plus fan), which held the temp at 50-degrees. I also tried using two 2-ohm resistors in parallel (20w capacity), which ran much cooler. I ran three power supplies at once, moving them between batteries every 10 - 20 minutes, when voltage got above 2.5. Took about 2 hours. After troubleshooting a loose connection on the Batrium cell monitors, automatic charging worked (though it strangely cycled between 1 and 5 amps measured at the outlet, stabilizing and going to the predicted 10 amps after a few hours. After manual charging, the batteries were then all over the place (2.5 - 3.2). I don't know how (no bypass was happening AFAIK), but by the time the packs were getting close to 3.3v they were all very close in voltage... Ohmboy is back on the road!
You would have been better to use 4 cells inseries and control the current by adjusting the voltage. Hopefully the cells will survive the 10 amp charge rate, a max of 4 amps would have been far better over a longer period.
I would still recommend you buy a few single cell chargers from Marcy or Trev @ https://www.drivebynature.com/ so you can balance the cells manually to assist the BMS balancing.

T1 Terry
Green but want to learn
Post Reply