Overdischarge of Lithium Iron Phosphate

llile

1 kW
Joined
Dec 18, 2010
Messages
457
I'm helping a guy who has been building a little electric truck out of a small Toyota pickup. He's using a 120VDC Li-Fe-PO4 pack made out of prismatic cells very similar to this one in appareance, probably 60 AH or 100 AH.

Well, it's bad news, his project has been sitting around for a few years and it looks like the cells have been overdischarged. I'm getting 200 mv across each cell. If this were LiPOs I would have already junked them, but it's Lithium Iron Phosphate for certain.

So my question is:

1. Is it safe to try to charge them at all?
2. We might just pull one cell and attempt to charge it (in a metal box? outside?). If so, any recommendations on a charge protocol? we could set up basically any CC and voltage profile needed.
3. What are the odds that the cells could be revived, even if they have severely damaged cycle capacity?
 
llile said:
1. Is it safe to try to charge them at all?
2. We might just pull one cell and attempt to charge it (in a metal box? outside?). If so, any recommendations on a charge protocol? we could set up basically any CC and voltage profile needed.
3. What are the odds that the cells could be revived, even if they have severely damaged cycle capacity?

1. Yes, LFP is generally recognized as safe even after many overdischarge cycles.
2. No special precautions needed, but I would charge with standard lithium battery precharge, which is 1/2 standard charge current back up to the nominal max discharge voltage of 2.5V (or 2.8V to be safe), then standard charge profile.
3. 100% can be revived, but expect 25% capacity loss.
 
THe cells *are* damaged, and will not perform like they used to. How safe it is to recharge them will depend on the damage that has occured to them. If they are swollen up, they may be damaged in unrecoverable ways.

Not every cell behaves the same when this happens, but since the cells were probably discharged slowly over time rather than rapidly under a high load, there is probably less severe damage than otherwise.

Slowly recharging them (personally I'd go much slower than the previous post recommends, more like a trickle charge of a few dozen to few hundred mA) until they reach the typical minimum safe discharge voltage, and then recharging them as normal from there, may bring them back up to a full charge.

They probably will no longer hold a voltage above 3.2-3.3v (new cells usually hold a voltage up to 3.6-3.65v if charged to that for balancing) once the charger is removed, and they will probably not all be in the same state of charge, meaning they would become unbalanced more easily under higher loads or deep discharges. Their capacity will be degraded significantly, but you will have to test them to find out by how much. Their capability (how much current they can deliver without significant voltage sag) will also be degraded, and you'll have to test that, too.

It is possible they will create significantly more waste heat under their originally normal max C-rates, along with voltage sag to match. It is not likely that this will be sufficient to cause fires or similar thermal damage, but it is something to keep an eye on during initial testing.
 
fatty said:
2. No special precautions needed, but I would charge with standard lithium battery precharge, which is 1/2 standard charge current back up to the nominal max discharge voltage of 2.5V (or 2.8V to be safe), then standard charge profile.


noooooo. dont charge that fast if they are under 1V. charge at 10mAh per Ah of capacity until you reach 3V. so 60 Ah blocks need 600mA of charging current.
charging too fast at such low voltages utterly destroys the anodes, something you are trying to avoid.

at 3V crank up the charging current to 1/10th the capacity, so 6A for 60Ah cells and hold it at 3.7V until the current has dropped to less then 600mA and then let the pack rest for 72 hours to check self discharge. and bad blocks will be shown clearly when you measure voltage of each block.

if the pack never dips below 600mA you need to continue charging and see what cells are puffing up or just trash the whole pack. because it means cells are really busted.

but if he has those crap chinese blocks i will assume the pack is simply FUBAR. those cells you linked to are junk.
 
The literature I've read indicates charge at 0.1C, which isn't a bad idea, but enough to buy a special programmable charger?

The attached study ascribes the capacity loss specifically to overdischarge, though the effect of charge rate was not investigated.
 

Attachments

  • c6ra01677d.pdf
    3.4 MB · Views: 11
if somone has a high voltage DIY lifepo pack in a EV one can assume he has a lab power supply or a way to control/limit the charger he already has.

no programmable "charger" needed.

and please stop injecting your own ignorance every time charging is mentioned. you lost the conversation every time. grow up and accept you dont know enough about the subject. just let i go and use the time to learn the basics on electronics first.
 
flippy said:
if somone has a high voltage DIY lifepo pack in a EV one can assume he has a lab power supply or a way to control/limit the charger he already has.
Also someone who let the cells overdischarge to 0.2V. Your assumption here isn't supported.

flippy said:
and please stop injecting your own ignorance every time charging is mentioned. you lost the conversation every time. grow up and accept you dont know enough about the subject. just let i go and use the time to learn the basics on electronics first.
I don't know what this is referring to. I agreed that a lower charge rate is used in the literature. What published effect does it have on capacity loss? Post up a study.

I think I've only ever disagreed with you on your inappropriate advice to noobs to naked charge with a power supply. We get it -- you're an EE. I don't know why you incorrectly assume everybody else has your level of expertise and risk tolerance. You should know better than anyone to give fail-safe advice.
 
For reference, we're talking about the difference between 0.25C and 0.1C. The cells are already down 25% capacity. How much additional capacity loss does the 0.15C difference induce?

When the OPs written alternative is junking the cells, what difference does it make?
I'm not saying don't charge at 0.1C. I'm just saying I wouldn't go out of my way to worry about it.
 
fatty said:
The literature I've read indicates charge at 0.1C, which isn't a bad idea, but enough to buy a special programmable charger?

The attached study ascribes the capacity loss specifically to overdischarge, though the effect of charge rate was not investigated.

As I indicated we have plenty of kit around here to create ANY charge current and voltage profile needed. Just a matter of cobbling it up. There's about six Mean-Well CC-CV supplies with adjustable parameters, a few adjustable voltage current-limiting power supplies, and a whole electronics bench, between those and a few jumpers we can produce basically any CC-CV profile that's needed.

We'll use an old-school programmable power supply - a voltmeter and an eyeball. My old professor called this "Airman control", you assign an Airman (basically a Private in the air force) to watching a voltmeter with his hand on the switch. It'll be a few long boring days babysitting this thing. Although people have said it's probably not a fire hazard, we're still moving it out of his garage into the driveway, 30 feet from any structure. Just don't want to take any chances.

I'll investigate these cells further and try to determine what the original specs were. They could be CALB cells, not sure yet.

0.1 C for a 100AH cell would be 10 amps! I won't hit it with near that much juice at first. One of the guys on this thread recommended 10 MAh per AH, that's more like 1 amp, we can easily cobble up a system that will push that. Once I'm confident I know what the original specs are, we'll start charging the whole pack at 0.010 C while carefully monitoring voltage, up to 2.5V per cell. We can monitor individual cells and look for any that seem suspicious.

Even if we've lost 25% or 50% of our capacity, that's better than 100% loss and we could at least test the whole kit to find out if his truck can go!

Thanks all for the help!
 
So far results are great.

As you can see from the pics below, we cobbled up a controlled voltage/controlled current power supply out of some lab power supplies, MeanWell power supplies, car batteries and whatnot. Current was carefully monitored and no more than 500 milliamps for these 100 AH CALB cells. That's 5 ma per amphour rated capacity. Voltage rose slowly over the span of about four hours, and we achieved 2.5 volts per cell just as we ran out of more wierd voltage sources to put in series with the whole mess. All cell voltages were within about 50mv of each other, well balanced and no cells appearing any worse off than the others. There's a charger built into the vehicle but it is high current and I don't trust it.

We've let the pack sit for several days, and will go back to check them. We'll bring back a charger capable of bringing the pack up to full charge, but still work very slowly - no more than an amp this time which would take four days if we left it. I'm not charging this pack without constant observation, so that'll be a lot of time tinkering in the shop while keeping an eye on it.


http://lile.biz/wp-content/uploads/2021/05/IMG_20210501_160552577_HDR-scaled.jpg

http://lile.biz/wp-content/uploads/2021/05/IMG_20210501_160701158_HDR-scaled.jpg

http://lile.biz/wp-content/uploads/2021/05/IMG_20210501_160603186_HDR-scaled.jpg

http://lile.biz/wp-content/uploads/2021/05/IMG_20210501_160631656_HDR-scaled.jpg
 
Back
Top