Possibly overclamped EVE LifePo4s prismatics; what went wrong, and have I broken them?

[Tobster]

Hi All, I'm new here, and am in the process of assembling a house battery pack for daytime use of cheap off-peak EV electricity.
The pack consists of two rows of eight EVE 280Ah 3.2V prismatic LifePO4s, clamped together with six M5 bolts. There are 1mm plastic sheets between each cell, and then the end plates are 1mm plastic sheet, followed by 12mm OSB board, followed by 3mm angle iron, followed by coin washer and nyloc M5 nut.
The spec says to clamp them with 300kg of force.

This was my methodology:
Using this website's calculator.
Two rows = 600kg. 600kg * 9.81ms^2 = 5886N.
Per bolt, /6 = 981N per bolt.
Website, on default settings, gives 5.89Nm per bolt.
To be on the safe side I decided to go with the minimum of what my torque wrench will do; 2.5Nm. In theory, this is going to give under half of the required load; ~100-150Kg, which "should" be alright; still a decent amount of clamping force and better than nothing.

Not spotting any issues while repeated sequence-tightening, and everything seeming okay; I eventually get to 2.5Nm on each bolt and have a really close inspection of everything.
The end cells are slightly deformed ?
The short sides "bulge up" against the end plates. It's not as obvious from the photos (although you can see what's happening in the reflection of the M5 rods in the cell casings to some extend). In real life it's a bit cringier!
There's no visible deformity in the other cells.

Is this deformity to be expected or has it gone wrong; and if so where did my method fail?
Should I reject these cells?

 

[TomC4306]

Should I reject these cells?

Reject?
You did it.
We out here in DIY land are not equipped to compress these cells the way the manufacturer did. Computer controlled hydraulic press.
I mean OSB? Really? You couldn't come up with something stiffer?
I built boxes from ¾" spruce plywood to contain and fixate the cells so they would not move relative to one another. With a 1.5 mm paper spacer between each cell. At full charge I can almost not get a cell out of the box when I tried to break it down to top balance.
That is plenty. All this hoo-ha about compressing and springs and etc. Silly.
Edit... In my opinion.

 

[TomC4306]

Two rows = 600kg.

This was your failure.

 

[SeaGal]

...I eventually get to 2.5Nm on each bolt and have a really close inspection of everything.
The end cells are slightly deformed ?

Really? With only 2.5Nm torque?

 

[MisterSandals]

Should I reject these cells?

Geesh, they're fine. Aim more towards very firm constraint than hundreds of kg of pressure (IMHO).

 

[OffGridInTheCity]

Is this deformity to be expected or has it gone wrong; and if so where did my method fail?
Should I reject these cells?

Seems like a case of focusing on an individual tree (tightening things) rather than the forrest - e.g. gentle pressure to keep them from expanding at full charge. I'm not sure they're ruined.

Take a look a this discussion on fixture concept (as apposed to compression) and examples of how several commercial batteries were assembled.

These LifePo4 blue cells have a characteristic of 1-2mm expansion / contraction based on SOC but I think the idea is that if you charge them to 100% often AND do not have fixture they will tend to bulge out against the flat metal case (which is not very strong) bending it out beyond the 1-2mm. Over time, this can lead to a shorter life as they can continue to bulge out beyond the 1-2mm as the case allows and you charge them fullly. Thus, a certain amount of fixture will keep it from bulging too far and increase life span.

Another unspoken 'thing' is that you insulate the metal cases from each other so they don't short. The blue wrapper alone may not be sufficient to stop shorts as I understand it. This is why many have 'foam' inbetween and this is kind of mixed up with the fixture/compression discussion. **Please someone correct me if I'm wrong on this***

 

[Symbioquine]

Two rows = 600kg.

This was your failure.

My gut feeling is that this is the right answer, but not necessarily because the math is wrong. Just that it assumes perfectly inflexible clamping surfaces. Are the middle cell edges also deformed?

The other aspect that hasn't been pointed out is the question of whether the torque wrench is accurate and being used correctly. There are cheap ones, broken ones, and even very high quality working ones which are a little hard to use correctly (multiple unit scales, unintuitive adjustable stops, etc).

Should I reject these cells?

I hope you mean reject them from your battery build process, not try and get a refund from the seller. If they're damaged, there's a good chance it's not the fault of the cells here.

In terms of whether they're still good, I don't know, but it's probably worthwhile to be cautious. Maybe if you decide to continue with them, plan for additional fire-proofing around the battery and use the battery away from your living space and other valuable system components to reduce the risk in the event of a failure.

 

[Lt.Dan]

I'm pretty sure the OSB is just deforming and putting all the pressure on the very edges of the cells, and not in the center, where actual clamping is required.

Wood is not suitable for this application, and im not sure why people keep using it?

 

[S Davis]

How can you set the compression by the torq on the rods? If one rod has more or less oil on the threads the amount of torque and pressure applied to the cell will be different.

I have seven 12volt 280ah batteries using EVE LF280N cells all clamped to just under the recommended 12 psi at 640psi on the cell face, they are used in a 4x4 work truck with lots of off road. They are not looking like that.

Are your rods spaced evenly they look like they are both towards the bottom of the cells?

 

[curiouscarbon]

I'm pretty sure the OSB is just deforming and putting all the pressure on the very edges of the cells, and not in the center, where actual clamping is required.

Wood is not suitable for this application, and im not sure why people keep using it?

due to the flexture of wood, i reject it for my own use when compressing cells.

stainless steel ~4mm or aluminum ~8mm are what i design with.

of course with electrical insulation between plate and cells.

oshcut laser cuts these materials for a pretty reasonable price.

no offense whatsoever to anyone using wood for compression, i just do not trust it to hold over time and not flex and distribute pressure in a suboptimal way.

kind regards

 

[~Yuri~]

I'm pretty sure the OSB is just deforming and putting all the pressure on the very edges of the cells, and not in the center, where actual clamping is required.

Wood is not suitable for this application, and im not sure why people keep using it?

Well, it's probably not quite right to assume that the wood is not suitable.
I have 21 mm thick plywood side plates. And I think it's very suitable
But OSB 12 mm, of course, is clearly not suitable for this purpose.

And I have 5 studs per package - two on one side of the package, two on the other, and one in the middle.
Here's a description with a photo of my intermediate creation

Pack / Cell compression Optimized By Using Springs.

I’m just curious, I read through the whole thread and noticed everyone seems to be using the springs on the outside of the batteries over the rods and dealing with flex of the end plate, or building heavier to mitigate. My my first question is why not make the fixture rigid with compression...

diysolarforum.com

 

[Tobster]

Really? With only 2.5Nm torque?

Yes; small torque wrench, thumb and third finger light wrist action, *click* - 100kg. To be honest, I didn't believe the calculator either, but I do now!

 

[Tobster]

Reject?
You did it.
We out here in DIY land are not equipped to compress these cells the way the manufacturer did. Computer controlled hydraulic press.
I mean OSB? Really? You couldn't come up with something stiffer?
I built boxes from ¾" spruce plywood to contain and fixate the cells so they would not move relative to one another. With a 1.5 mm paper spacer between each cell. At full charge I can almost not get a cell out of the box when I tried to break it down to top balance.
That is plenty. All this hoo-ha about compressing and springs and etc. Silly.
Edit... In my opinion.

OSB and 3mm angle iron.

 

[Tobster]

Seems like a case of focusing on an individual tree (tightening things) rather than the forrest - e.g. gentle pressure to keep them from expanding at full charge. I'm not sure they're ruined.

Take a look a this discussion on fixture concept (as apposed to compression) and examples of how several commercial batteries were assembled.

These LifePo4 blue cells have a characteristic of 1-2mm expansion / contraction based on SOC but I think the idea is that if you charge them to 100% often AND do not have fixture they will tend to bulge out against the flat metal case (which is not very strong) bending it out beyond the 1-2mm. Over time, this can lead to a shorter life as they can continue to bulge out beyond the 1-2mm as the case allows and you charge them fullly. Thus, a certain amount of fixture will keep it from bulging too far and increase life span.

Another unspoken 'thing' is that you insulate the metal cases from each other so they don't short. The blue wrapper alone may not be sufficient to stop shorts as I understand it. This is why many have 'foam' inbetween and this is kind of mixed up with the fixture/compression discussion. **Please someone correct me if I'm wrong on this***

Yeah I think basically it's going to be just holding tightness on the six bolts rather than any kind of force.
I'm using 1mm plastic sheets (sold far more cheaply as chopping boards!) in between each cell.
Any recommendation what state of charge to apply the gentle pressure?

 

[Tobster]

My gut feeling is that this is the right answer, but not necessarily because the math is wrong. Just that it assumes perfectly inflexible clamping surfaces. Are the middle cell edges also deformed?

The other aspect that hasn't been pointed out is the question of whether the torque wrench is accurate and being used correctly. There are cheap ones, broken ones, and even very high quality working ones which are a little hard to use correctly (multiple unit scales, unintuitive adjustable stops, etc).



I hope you mean reject them from your battery build process, not try and get a refund from the seller. If they're damaged, there's a good chance it's not the fault of the cells here.

In terms of whether they're still good, I don't know, but it's probably worthwhile to be cautious. Maybe if you decide to continue with them, plan for additional fire-proofing around the battery and use the battery away from your living space and other valuable system components to reduce the risk in the event of a failure.

Just those outer ones, right on the edge. Flexible end plates does make sense as the prevailing failure theory now thinking about it; the two middle bolts are spreading their load over both rows of batteries, whereas the end pairs are only applied to the side of each row.

"Reject" from build; not seek a refund from the vendor.

 

[OffGridInTheCity]

Any recommendation what state of charge to apply the gentle pressure?

The specs shown in the youtube for 280ah EVE(s) (at 2:53in youtube) show

An expansion from 71.5mm at 30% SOC -> 72.0mm at 100% SOC +/- 1.0mm. Perhaps you're situation is similar / maybe you have specs for you're cells? Perhaps apply gentle pressure at near 100% SOC.

 

[Tobster]

due to the flexture of wood, i reject it for my own use when compressing cells.

stainless steel ~4mm or aluminum ~8mm are what i design with.

of course with electrical insulation between plate and cells.

oshcut laser cuts these materials for a pretty reasonable price.

no offense whatsoever to anyone using wood for compression, i just do not trust it to hold over time and not flex and distribute pressure in a suboptimal way.

kind regards

How thick is the insulation between your end plates and the cells? I only put the OSB there so the cells wouldn't be any where near the 3mm angle iron clamping the plates.

 

[Tobster]

The specs shown in the youtube for 280ah EVE(s) (at 2:53in youtube) show
View attachment 123102
An expansion from 71.5mm at 30% SOC -> 72.0mm at 100% SOC +/- 1.0mm. Perhaps you're situation is similar / maybe you have specs for you're cells? Perhaps apply gentle pressure at near 100% SOC.

I found a couple of datasheets for the LF280K which had similar but less helpful information. 71.5 -> 72mm for eight cells is 4mm! ?

 

[TomC4306]

For me, the primary purpose of restraining the cells is to prevent the natural swelling from putting strain on the terminals at the busbars.

 

[Dynoman]

Hi All, I'm new here, and am in the process of assembling a house battery pack for daytime use of cheap off-peak EV electricity.
The pack consists of two rows of eight EVE 280Ah 3.2V prismatic LifePO4s, clamped together with six M5 bolts. There are 1mm plastic sheets between each cell, and then the end plates are 1mm plastic sheet, followed by 12mm OSB board, followed by 3mm angle iron, followed by coin washer and nyloc M5 nut.
The spec says to clamp them with 300kg of force.

This was my methodology:
Using this website's calculator.
Two rows = 600kg. 600kg * 9.81ms^2 = 5886N.
Per bolt, /6 = 981N per bolt.
Website, on default settings, gives 5.89Nm per bolt.
To be on the safe side I decided to go with the minimum of what my torque wrench will do; 2.5Nm. In theory, this is going to give under half of the required load; ~100-150Kg, which "should" be alright; still a decent amount of clamping force and better than nothing.

Not spotting any issues while repeated sequence-tightening, and everything seeming okay; I eventually get to 2.5Nm on each bolt and have a really close inspection of everything.
The end cells are slightly deformed ?
The short sides "bulge up" against the end plates. It's not as obvious from the photos (although you can see what's happening in the reflection of the M5 rods in the cell casings to some extend). In real life it's a bit cringier!
There's no visible deformity in the other cells.

Is this deformity to be expected or has it gone wrong; and if so where did my method fail?
Should I reject these cells?

Although not an expert from what I understand yours cells were significantly over compressed. Suggest cells be removed from compression and compress less.

5.89Nm per bolt = 52.13 Inch Pounds.

This is Info I used and torqued a little less at 5 Inch Pounds with a torque wrench at about 3.2 to 3.3 volts charge in each cell.

The spec from EVE was 300 KG force which rounds off to 660lbs. Battery face is approx 6.85"x 7.874" = 53.94 sq inches
660lbs/53.94sqin=12.23 lbs per sq inch
Divide 660 by 4 bolts that's 165 lbs Axial (clamping) force per bolt.
Using 4 course 1/4 in threaded rods that should equate to roughly 8 INCH pounds torque per bolt. Realistically, that's a snug twist of the wrist on a regular nut driver for the average build mechanic.