I accidentally discharged 1 LiFePo4 cell to 1 V, can I make a 7s or 15s pack, or do I need to find a replacement for the dead cell.

[coplate]

I recently got my order of 16 90AH cells, and I was excited to start testing with them.

Summary: I made some very stupid mistakes.
Capacity testing an 8s pack of raw cells, one of the cells dropped to below 1 volt, and is now swollen like a balloon, say 2-4mm on each side.
Is it reasonable to use 7s or 15s bms with a charge controller, or do I need to try to find a replacement that matches?



Longer version

I have some experience building batteries from used 18650 cells, to make portable power supplies for things like my boombox, laptop, etc. But I have never worked with LiFePo4 before.
I have been preparing to get a workshop/home-office built in my yard, and wanted to see if I could set up solar panels on the roof, and try to run it off solar with battery backup. I ordered 16 90AH cells, and they arrived recently.

I picked 8 out to make a 24v pack, and they were all at 3.2v, so I went ahead and put them in series, and started to charge them up at 28.8v.

I had completely forgotten that with lifepo, even though they are the same voltage, they can be at drastically different SoC. ( This could be the reason, or one cell could have just been bad )

After some time, my charger was in CV mode and the current had dropped, so I took the batteries off, and got ready to capacity test them, assuming the batteries would be around 90% SoC.

When I started the capacity test, everything started out well, but after 10 hours, I noticed the voltage had dropped lower than I expected. I started checking each cell, and one of them was below 1v, and the reset were all around the same voltage, 2.8 or similar.

I started a slow charge of that cell, and I didn't notice if it happened during discharge or recharging but that cell swelled up dramatically. To the point where I would not trust it at my level of experience with these cells.

I have read plenty of posts that discuss compression, and possibly draining it to see if it will debloat and so-on, but most likely I simply do not want to deal with this cell.

I wasn't able to find any posts talking about using 7s or 15s as the source for a 48v or 24v inverter or all-in-one. I know that plenty of solar charge controllers have programmable voltage cuttofs, so I wanted to ask about anyone's experience in this situation.

There are probably a thousand questions I don't even know to ask, so any other thoughts on this would be appreciated.

 

[A.Justice]

It's definitely possible to use a 7 or 15s pack, I'm pretty sure a JBD 8s BMS will take 7 or 8s, and the 16s will take 15s. Someone feel free to correct me if I'm wrong, but I'm pretty sure I read that in the spec sheet on my 120 amp unit.

Most charge controllers have a programmable voltage, so I don't see charging it with solar as much of an issue.

The biggest issue I see is your inverter. A lot of them take very specific voltage ranges, and you're likely to fall outside of the range of most of them. That being said, I'm sure there's some available that do meet the voltage specs you need, but you're going to severely limit your choices.

So, it's totally possible, but it's definitely going to add some build time sifting through specifications of different equipment to try to find one that can match your voltage.

 

[coplate]

@A.Justice
Thanks for the comment. Right now I have a cheapo 48v inverter I bought during a power outage so I could run my fridge with my e-bike battery, which is 13s 18650s, so I know it cuts off around 40 volts ( 3..v per 18650 cell ).
That would be 2.67v for a 15s LiFePO4 pack, instead of the typical 2.5v for a 16s pack that I have seen.

If I'm math is right and not backwards, that probably better for me in the long run, right; the more cells in the pack, the lower each cell has to be to hit the inverter cutoff voltage? So in theory a 20s battery would drain each cell to 2v to hit that cutoff voltage, which is too low.

I'll lose a little capacity, but on lifepo4 the difference between 2.5 and 2.67 is pretty small, right? That's the equivalent of cutting off a 12v pack at 10.7 instead of 10.

But I don't have any experience with the big-boy inverters that I would need to power a whole home-office, the one I have has only ever been used for a single appliance at one time, and even then, beyond the startup rush there was not a ton of drain from my fridge.
I dont have much insight into what my expected voltage drop from the load would be, but I know it isn't always negligible.

 

[rmaddy]

If I'm math is right and not backwards, that probably better for me in the long run, right; the more cells in the pack, the lower each cell has to be to hit the inverter cutoff voltage? So in theory a 20s battery would drain each cell to 2v to hit that cutoff voltage, which is too low.

This is not really the proper way to look at it. If there are more cells then you would need to raise the inverter's cutoff voltage to avoid draining the cells too far. In your case with one missing cell, you would need to lower the inverter's cutoff voltage to match the lower total voltage of the battery.

In either of those cases you want the cutoff to be appropriate such that each cell ends up at the same safe low voltage.

If the inverter's cutoff voltage can't be changed then it would be bad to use it with a battery made up of extra cells. And in your case the higher voltage would limit how much of your battery that you can use because it would cutout too soon.

 

[coplate]

@rmaddy
Thanks for the response.

I was looking at it the same way, but it looks like I just phrased things poorly. What I was trying to say about the extra cells would be that it is very bad for the cells, and that fewer cells would lose capacity, but would stay safe. 10s was chosen as exaggeration, to just show a extreme low voltage on the cell.

This is assuming that an inverter that has a cutoff at 40v, would leave the balanced LiFePO4 cells cutoff at 2.67, instead of 2.5, and that that would be safe, assuming 2.5 is safe. And having a cutoff of a higher safe voltage would lead to a very slightly longer lifetime of the cells.

Beyond the loss of capacity for cutting off a little early, I can't think of anything dangerous about using an inverter that is safe for a 16s LiFePO4 battery pack, that would become dangerous with a 15s, but I don't know if I'm missing something fundamental.

 

[MisterSandals]

Capacity testing an 8s pack of raw cells, one of the cells dropped to below 1 volt, and is now swollen like a balloon, say 2-4mm on each side.

Have you tried to save this cell? I would put a little compression on it (boards and clamps in a pinch) and get it over 2.5V asap. It may regain its proper shape and be ok. Maybe not.

 

[coplate]

@MisterSandals
I charged it up to 3 volts as soon as I noticed it. I was draining the 8s pack at about 150w, so it was not very many amps coming through it at the time, which I assume is a good thing.

I am letting it set for a few days, to see if it self-drains below what I set it at, and then I will try another charging cycle of just this cell to get a capacity, and see how close it is to the other cells. I have clamps and boards in my garage, I can also set it under a flat board and control the weight on top of it with weights.

Would you say I should try to compress it while doing the charge/discharge cycles, or just try to store it compressed for a while, or anything else?

I mentioned in my first post that I had seen various posts on trusting or not cells that have bloated up like this, but right now with the lack of experience I figured my best bet is to assume it is dead.

 

[MisterSandals]

Would you say I should try to compress it while doing the charge/discharge cycles, or just try to store it compressed for a while, or anything else?

My understanding is that cells naturally expand and contract while charging and discharging, which is why most recommend compressing them. I consider what i do as light compression but very firm constraint. So to answer your question, i suspect your best chance of reforming the original shape would be compressing while charging and discharging a couple times.

I had seen various posts on trusting or not cells that have bloated up like this,

There have been quite a few cells bloated what looked like a half inch or more on each side. Those are dead and should not be trusted. Yours bloated a couple millimeters so maybe salvageable. I would really keep any eye on it until you figure out its condition. Anything you learn will help the collective knowledge here!

Good luck, keep us posted!

 

[coplate]

@MisterSandals

Thank you and everyone for your comments.


I've set it up clamped in some plywood, just finger tight, that's the best compression method I have available to me today.

I took a couple pictures, Currently discharging it at 9 Amps, ( 0.1c ). Don't mind all the extra wires hooked up, That's to help make up the voltage drop since all my charging wires are sized for my 18650 stuff.

On that first image, the total bulge is about 1/4", but I didn't have enough hands to hold the ruler up while taking the picture.

 

[coplate]

@MisterSandals
I've done a few charge/discharges on it, with it clamped up, but when I remove the clamps, it looks like it springs out nearly as far as before.
The capacity is testing close enough to all of the other cells in the lot I bought though, so I am trying to decide if I just use it swollen, or ditch it.


Any thoughts?

 

[MisterSandals]

Any thoughts?

Its beyond anything i have personal experience with so i really do not know. My thought is if it behaves normally, with no heat on charge or discharge, and you keep it constrained (so many ifs!) it may be good to go. I would definitely watch it closely until it proves itself stable.

Maybe start a treat asking if anybody has experience with "recovered bloated cell".