Swollen cells, how much is acceptable and what are the failure modes.

[Luk88]

So I've found myself in unexpected possession of some more or less bloated cells. As I'm going through with discussion to get my money back I'm also wondering if, with some acceptance of risk, these can be used, but also I'm interested in the topic in general.

I think quite a few people have bloated cells in operation, judging by what I found said in some old threads. So I would like to ask. How much bloating is definitely too much?

And more importantly, what is the mode of failure of these bloated cells? Let's say one was monitoring things like AC IR, and temperature (also fixture pressure if they are compressed). Is it possible to intercept such a cell before it decides to blow its valve? I'd love to at least have an opportunity to measure capacity before they are sent back.

Let's say a cell is taking a charge and is not heating up. Is it likely it may suddenly develop an internal short and "run away"?

If anyone is interested. The cells I got are mostly bloated by few mm. 2~4, the worst is about 6mm, but that one is also concave slightly on the opposite end. As if it drooped...

The voltages are 3.26V and very equal which makes me think they were equalised/charged before. (they came in 2 days after ordering).

One of my cells has also a crack on its side. Spreading it with a plastic spudger I could see pouches inside that don't look damaged, but they may well be. Surprisingly that cell is not swollen much. Maybe 1.5mm. It doesn't smell more than when I open my ammo box full of my fpv li-ion packs.

I have these cells in a worshop with a CO2 fire extinguisher nearby and not much flammable stuff around them. Just in case.

 

[Luk88]

I can actually share a pretty good tip here which I discovered (probably nothing new for many of you, but I didn't know DC IR issues are a good proxy for SOH).

There is a very good way to test swollen, or otherwise bad cells without a risk of charging them. If they arrive at 3.2V or higher their DC internal resistance shouldn't be far from what it would be when full.

What I did was to connect then with a BMS and try pulling 0.5C and then 1C current from them briefly (for 10s or so). The voltage meter has to be connected to the battery cells to avoid wiring/bms influence.


For example, my cells have dropped 0.4V each under 0.5C load and couldn't be tested at 1C as bms cutoff disconnected at 2.5V (these were at 3.26V).

Not everyone has equipment to test individual cells at currents of 100A or 200A, but by setting up a pack and using an inverter this DC internal resistance test is available to anyone. Also a Bluetooth enabled BMS s a big plus as it can show you if all the batteries are crap or just some.

 

[MisterSandals]

Pics would help others understand what you are seeing. 2-4mm swelling for a cell (1-2mm per side) is nothing that i ever worried about or noticed any problems as a result.

 

[Luk88]

Pics would help others understand what you are seeing. 2-4mm swelling for a cell (1-2mm per side) is nothing that i ever worried about or noticed any problems as a result.

So you're saying this swelling 2-4mm (not 1-2mm per side, they are not equally swollen, often one side is swollen the other isn't) has nothing to do with the DC resistance these cells are exhibiting?

Here is what around 4mm looks like on a flat surface:


Id love to find out what others think. Is inability to deliver 1C current linked to this swelling or is it a separate matter?

I read somewhere these (blue plastic) cells have not been made since many years ago. Can it be simply storage and the swelling (simply a result) was caused when they were recharged too quickly after very long storage?

Finally, the seller will say something like "it will improve with cycling" (about the DC resistance). This is simply not true from the papers I read. I'm very interested if anyone has information to the contrary. Specifically, if your cells have improved their DC IR significantly after cycling.

 

[Higher_wire]

Swelling isn't abnormal and is generally caused by both thermal expansion and gassing during operation, especially at higher temperature and/or C rates. As the cell ages, this accompanies a whole host of things going on inside the cell, including electrolyte loss, SEI film formation, and loss of active lithium ions. The result is impedance rise along with capacity and power fade. Your cells exhibit exactly that, having massive voltage sag under moderate load.

From the studies I've read, mechanical strain can be used as an early detection for TR; however, we're talking a couple minutes' warning at best.

If these were "new" then they were likely manufacturing rejects and sold to a reseller as B-grade or lower. If they've been sitting then they've likely lost a bit of electrolyte and lithium inventory. Are they fine for stationary with a max of C/2 continuous? More than likely, though I'd imagine their cycle life is going to be limited.

 

[MisterSandals]

Id love to find out what others think. Is inability to deliver 1C current linked to this swelling or is it a separate matter?

I am not familiar with the plastic cased cells. The video looks like the cell compresses easily and might just be a bowed case? Does it feel hollow at the bowed areas?

For example, my cells have dropped 0.4V each under 0.5C load and couldn't be tested at 1C as bms cutoff disconnected at 2.5V (these were at 3.26V).

First off, 3.26V is pretty low state of charge. What is the cell size and what loads?
I've not tried 1C discharge but perhaps trying with a fully charged battery?

Sorry i'm not much help.

 

[Luk88]

Swelling isn't abnormal and is generally caused by both thermal expansion and gassing during operation, especially at higher temperature and/or C rates. As the cell ages, this accompanies a whole host of things going on inside the cell, including electrolyte loss, SEI film formation, and loss of active lithium ions. The result is impedance rise along with capacity and power fade. Your cells exhibit exactly that, having massive voltage sag under moderate load.

These are supposedly new cells... Also they measure AC IR within specs. I think you're right at SEI growth and active lithium loss. I wonder if they are truthful in that these cells were not used. I found threadlocker in one of the threads so there is that... But in principle could such deterioration happen just from long storage? Perhaps?

From the studies I've read, mechanical strain can be used as an early detection for TR; however, we're talking a couple minutes' warning at best.

If these were "new" then they were likely manufacturing rejects and sold to a reseller as B-grade or lower. If they've been sitting then they've likely lost a bit of electrolyte and lithium inventory. Are they fine for stationary with a max of C/2 continuous? More than likely, though I'd imagine their cycle life is going to be limited.

I disagree. How can they be fine for C/2 when they drop 0.5V per cell at such load? No way, unless you meant to write C/10. They very much remind me of the crappy battery I had before. It too had horrible voltage sag, but capacity measured correct and there was no swelling in that battery.

I have a theory how this swelling occurs. I'd love to hear what others think about it. I think these cells end up sitting for ages and slowly loosing their charge. When finally someone checks them 4 years later they sit at something like 2.4V (they've already gown thick SEI film and lost active lithium so they have lots of internal DC resistance). Then the seller wants to charge them, so they slap them together quickly into "batteries" possibly with no BMS and they are charged as quickly as possible. This is what causes electrolyte breakdown when they inevitably exceed voltages and heat caused swelling on others. It's just an idea I have.

I imagine bunch of guys in an old warehouse putting together a 100S pack and giving it rectified 240V AC (that's about 340V DC) to charge.

I am not familiar with the plastic cased cells. The video looks like the cell compresses easily and might just be a bowed case? Does it feel hollow at the bowed areas?

If so the video was misleading... This is really very hard plastic. They are much tougher than aluminum cells. I can bend slightly swollen aluminium cell in my hand. There is no way in hell I can do it with that.

Just to illustrate how hard they are. I've built the 4 threaded rods compression thing to see if they'd compress and I cranked it to 400kg(as per disc spring deformation) .

There was only about 3mm of movement. In a pack of 8 cells. At the same time I had gaps ~4mm still. These are very different than aluminium encased cells.

First off, 3.26V is pretty low state of charge.

3.26 is in the region of 50%+. DC IR is pretty flat for lifepo through its entire SOC (going maybe 10% up at both ends). This is one of the reasons everyone loves them so much.

The seller asked me for a "video of my measurements". I did that and after I recorded a 8s pack made of these cells dropping over 3V with 0.5C load I also recorded the same test with a cheap "Kepworth" brand 24v 100ah lifepo battery. You know what the voltage drop was? 0.6V.

What is the cell size and what loads?

200ah, the load was an inverter running a heater that pull about 2300W from the battery. At this state of charge this translates to about 100A. That was the 0.5C load.

The 1C was two such heaters.

However I've since built a "special" resistor that has two legs 0.032 ohms each. Which let's me pull exactly 50A, 100A or 200A from a single cell. (legs connected in series, only one leg, in parallel). Now I have to figure out how to switch it on and off (probably with a massive scr, or a bank of mosfets) and I'll have a single cell high current tester for DC IR.

I've not tried 1C discharge but perhaps trying with a fully charged battery?

It doesn't really work like that. You're supposed to be able to pull 1C till the very end loosing maybe 10% of capacity because of the voltage drop. When the drop is that high you loose 80%+

Sorry i'm not much help.

No worries, we're all learning here

 

[Luk88]

Has anyone heard about cells deteriorated by long storage improving their performance after first proper charge? I certainly didn't, but of course the seller is going to say anything for a shred of chance I agree to keep them.

All studies I read claim once SEI film thickened there is no way of getting it thinner again.

Anyone has any experience to the contrary? We are talking about old, and "second life" lifepo cells of course.

 

[Daddy Tanuki]

Has anyone heard about cells deteriorated by long storage improving their performance after first proper charge? I certainly didn't, but of course the seller is going to say anything for a shred of chance I agree to keep them.

All studies I read claim once SEI film thickened there is no way of getting it thinner again.

Anyone has any experience to the contrary? We are talking about old, and "second life" lifepo cells of course.

Long explanation, but read all of it.

I have a bunch of the CALB's that are blue cased like yours. I bought mine starting five years ago and my research at the time indicated they had been out of production for about 2~3 years when I bought my first batch.

The first order of 32 were pretty solid and 8 months later I then ordered another 20. 8 months after that the last 16 were ordered. I quickly had balancing issues & capacity issues when i started running 4 separate packs. I then tried putting them together into a 4p16s pack and still had issues.

it was not until the last order that I managed to set up a trustworthy and repeatable test rig to test capacity that I found out that 8 from the second batch were sub optimal as well as 8 from the third batch.

as such I now have 48 out of the 68 that I bought that are in use and show close to full capacity. I am getting about 95% of the capacity depending upon how high of a C rate i test them. The remainder test from 80% all the way down to 55%

of these "reject cells" about half of them have some form of minor bloating, very similiar in appearance to the video you posted.

I would send them back and claw your cash back. you got 5-8 year old cells that were used and the pulled from an electric bus, or forklift and cleaned up.

I have since found a reputable Manufacturer that cells their own cells online. Thundersky Winston.

long story short, Winston, Sinopoly, CALB were all part of the same company that later split up into factions. Thundersky came along later and merged with Winston.

I cannot speak for any of the aluminum cased cells, but if you want nylon cased cells, which do have their benefits over the aluminum cells, Winston is the way to go.

 

[Luk88]

Long explanation, but read all of it.

I have a bunch of the CALB's that are blue cased like yours. I bought mine starting five years ago and my research at the time indicated they had been out of production for about 2~3 years when I bought my first batch.

The first order of 32 were pretty solid and 8 months later I then ordered another 20. 8 months after that the last 16 were ordered. I quickly had balancing issues & capacity issues when i started running 4 separate packs. I then tried putting them together into a 4p16s pack and still had issues.

it was not until the last order that I managed to set up a trustworthy and repeatable test rig to test capacity that I found out that 8 from the second batch were sub optimal as well as 8 from the third batch.

as such I now have 48 out of the 68 that I bought that are in use and show close to full capacity. I am getting about 95% of the capacity depending upon how high of a C rate i test them. The remainder test from 80% all the way down to 55%

of these "reject cells" about half of them have some form of minor bloating, very similiar in appearance to the video you posted.

I would send them back and claw your cash back. you got 5-8 year old cells that were used and the pulled from an electric bus, or forklift and cleaned up.

I have since found a reputable Manufacturer that cells their own cells online. Thundersky Winston.

long story short, Winston, Sinopoly, CALB were all part of the same company that later split up into factions. Thundersky came along later and merged with Winston.

I cannot speak for any of the aluminum cased cells, but if you want nylon cased cells, which do have their benefits over the aluminum cells, Winston is the way to go.

Thanks. Yes I'm definitely returning them, but there is a bit of a problem with shipping. You see on Ali where I bought them they are both a "free return" item as well as in the 14 days after the internet purchase here in EU one can return anything for any reason or even without a reason. So I've logged a case with Ali and it was quickly resolved that I need to return these and I'll get a full refund. There was even a shipping label generated. So what is the problem you may ask?

Well, the shipping label is a standard courier service that accepts up to 30kg, while these are two boxes 23kg each. I'm on third "support case" with Ali now, two previous ones both being closed in about a day or so with an email to me like this (paraphrasing) "We regret you had problems, we're very sorry, here is your solution: please put all items in one box". I explained in words, I explained with pictures like to a 3 year old. Literaly, here are the boxes.... each weights 23kg, total weight is 46kg, and your label is with a courier company that only does standard service which is limited to 30 kg. I even offered to pay for the bloody shipping myself with another courier (the only courier that can take this as one package is Fedex). Nope, I have to use their label or the system will not register it. They are still "escalating" so I hope eventually after lots of back and forth they will let me post the bloody thing myself. They even had a system like this before where you'd supply your own tracking number. However people complained ab out not getting the shipping costs back so they "improved it".

Oh well, live and learn.

So in the meantime I'm seeing if I can use'em for anything if this prooves a huge hassle and they propose something like 80% discount. At this point, seriously if they don't let me post them I should charge them for disposal in addition.

 

[Luk88]

In case anyone's interested. I've measured capacity in addition to DC resistance. It is 110Ah.

 

[Daddy Tanuki]

In case anyone's interested. I've measured capacity in addition to DC resistance. It is 110Ah.

yep you got a bunch of the pull offs. when I bought mine I spoke with the vendor repeatedly and even called him. I could never get him to admit they were pulloffs. but it is obvious when you measure the capacity and its barely half.

 

[Higher_wire]

I disagree. How can they be fine for C/2 when they drop 0.5V per cell at such load?

Yes brain fart on my end.

Good job catching some second life cells, and I’m sorry for your experience. Those are truly at the end of their life!

 

[Warpspeed]

These cells are made of very tough plastic material (?) but internal gas pressure will cause some normal slight swelling if the cells are not rigidly clamped as recommended by the supplier.
Once swollen, they will not return back to shape all by themselves, without assistance from clamping.
If pressed back into shape, they will then stay that way.

I have successfully returned swollen cells back to flat by clamping after a terrifying undervoltage accident.
That has been my own personal experience.

Now there is not a lot of information about all this out there, but the only real problem I can see with swelling, is that the internal cell volume will have obviously increased.
That, if extreme, can lower the electrolyte level excessively. If the swelling is very slight, only a very few mm, forget about it being a problem.
If it blows up like a football with TENS of mm of swelling, then the electrolyte level will have dropped dangerously to the point of the tops of the pouches will then be running high and dry.

Even that might not be a problem if the swelling is immediately corrected (within a few hours) with some serious clamping. That has been my own personal experience. That happened several years ago, and those very briefly swollen cells are still working just fine.
If the cells are hugely bloated and have been like that and sitting on a shelf for years, they are very likely stuffed.

People become paranoid about 1mm of swelling. Fair enough, these cells are not cheap.
But the reality is, that unless the swelling is gross enough for the internal pouches to dry out at the top, a bit of swelling is not a problem.

 

[Luk88]

These cells are made of very tough plastic material (?) but internal gas pressure will cause some normal slight swelling if the cells are not rigidly clamped as recommended by the supplier.
Once swollen, they will not return back to shape all by themselves, without assistance from clamping.
If pressed back into shape, they will then stay that way.

I have successfully returned swollen cells back to flat by clamping after a terrifying undervoltage accident.
That has been my own personal experience.

Now there is not a lot of information about all this out there, but the only real problem I can see with swelling, is that the internal cell volume will have obviously increased.
That, if extreme, can lower the electrolyte level excessively. If the swelling is very slight, only a very few mm, forget about it being a problem.
If it blows up like a football with TENS of mm of swelling, then the electrolyte level will have dropped dangerously to the point of the tops of the pouches will then be running high and dry.

Even that might not be a problem if the swelling is immediately corrected (within a few hours) with some serious clamping. That has been my own personal experience. That happened several years ago, and those very briefly swollen cells are still working just fine.
If the cells are hugely bloated and have been like that and sitting on a shelf for years, they are very likely stuffed.

People become paranoid about 1mm of swelling. Fair enough, these cells are not cheap.
But the reality is, that unless the swelling is gross enough for the internal pouches to dry out at the top, a bit of swelling is not a problem.

Thanks for your reply. I agree these are fine to swell during use, but not when new and unused. Swelling is in a way a proof they are not new cells.

Also, does anyone have any piece of official paperwork (datasheet) from CALB or any official chinese grading of automotive batteries that says the batteries can't be deformed initially? All makers other than CALB say it outright, but CALB just says the appearance needs to be "good". What does "Good" mean?

 

[Warpspeed]

Luke, there are no "new" cells, they are all charged and then tested during manufacture, then may be stored in a warehouse at high ambient temperatures, or even shipped overseas in a container that has been sitting in direct sun for weeks on a container ship sailing through the tropics.
There is nothing in the factory packing to resist a slight internal pressure buildup.

When the final user clamps the cells between two rigid plates as recommended, any slight swelling will be corrected.
The cells are completely sealed, nothing has changed, and very slight swelling is NOT a problem.
Likewise a slight scratch or blemish is not a total catastrophe either.

 

[Luk88]

Luke, there are no "new" cells, they are all charged and then tested during manufacture

This is so wrong when we're talking about 7~8 year old cells that measure slightly half of their rated capacity and an internal DC resistance worse than an old tiered lead acid battery.

Yes new cells exist. You want to see them? Order from Basen or Quishou, see the QR code and it will be less than 3 months old. With not a single charge on it other than the test at the factory.

You know what else can pass as new? My Higee cells I bought recently. They are a year old (confirmed by qr code) and out of 24 I bought not one came swollen. All measure at rated capacity.

Yes, new cells exist. This is why in fact we're talking about these end of life cells in a special subforum. Not to clutter threads of people that use real new cells with our crappy old cell problems.

, then may be stored in a warehouse at high ambient temperatures, or even shipped overseas in a container that has been sitting in direct sun for weeks on a container ship sailing through the tropics.
There is nothing in the factory packing to resist a slight internal pressure buildup.

Do you actually believe this or are you repeating what the sellers of these swollen cells say? I never heard of a shipment of new, unused cells that 80%, of would swell during shipping including to the point of rupturing the enclosure. It is insane to suggest this is acceptable and it isn't.

Manufacturers themselves know it very well. For example an excerpt of Higee datasheet:

Can you see where it says "free from deformation"? Deformation is not acceptable in grade A new cells.

Furthermore here in Europe there is a plethora of national standards that forbid use of supposedly new cells that are swollen. Don't believe me? Here is an official EU overview of current legislation: https://publications.jrc.ec.europa.eu/repository/handle/JRC135870

Cells that are swollen as new are not considered safe. This is completely different for used cells which when they aren't clamped will swell slightly in use. Especially at higher temperatures.

When the final user clamps the cells between two rigid plates as recommended, any slight swelling will be corrected.

The swelling during storage as opposed to swelling in use is a sign that something very bad is happening inside the battery. The amount of units that swelled in a given batch matters too. If you have a batch of 100 and one swells, perhaps that one had some manufacturing issue that slipped through the QA. But when 80% swell including to the point of breaking the rigid plastic container? These are not new cells, or they are manufacturing rejects.

The cells are completely sealed,

And the level of electrolyte drops exposing parts of the cell possibly causing much more intense heating and a thermal runway. This is why these are not acceptable to use and if your house burns down because of them the insurance will decline to pay because you used cells that don't meet the specs for stationary use.

nothing has changed, and very slight swelling is NOT a problem.
Likewise a slight scratch or blemish is not a total catastrophe either.

If you're buying a new car and they send you one that was rear ended do you also say it's a very slight deformation and is not a total catastrophe either?

If you then discover the engine is shot and had 100k miles on it will you say, it is a completely different issue and has nothing to do with the fact it was rear ended? IMO, both things point to the fact this is a used car/cell.

If it can be safely used at substantially derated capacity and discharge rates that is a different matter, but the fact is you have to now consider if any part of that system causes a fire it is your fault when using such cells. The manufacturer disclaims all liabilities for deformed cells and ones that are under 70% rated capacity.

This is a "second life" cells forum so we may consider second life for such cells, but please don't say things like "there are no new cells", because there are.

 

[Warpspeed]

If you're buying a new car and they send you one that was rear ended do you also say it's a very slight deformation and is not a total catastrophe either?

If I bought a brand new car and one of the tires had low air pressure and was slightly deformed, I would not insist the car dealer replaced the whole car under warranty. I would simply fix it myself by adding air and say nothing.
Hysterics at the car dealer will very likely NOT get you a replacement vehicle over a complete non issue.

The fact is, cells do deform quite easily.
Its why the manufactures insist we clamp the cells between flat rigid surfaces in the approved manner.
A very slight swelling can be pushed back into shape by clamping without any ill effects.
That is what I suggest you do.

 

[Luk88]

If I bought a brand new car and one of the tires had low air pressure and was slightly deformed, I would not insist the car dealer replaced the whole car under warranty. I would simply fix it myself by adding air and say nothing.
Hysterics at the car dealer will very likely NOT get you a replacement vehicle over a complete non issue.

The fact is, cells do deform quite easily.
Its why the manufactures insist we clamp the cells between flat rigid surfaces in the approved manner.
A very slight swelling can be pushed back into shape by clamping without any ill effects.
That is what I suggest you do.

Very funny. Do you by any chance sell these swollen cells too?

 

[Warpspeed]

Nope.