EVE-280 cells should these be clamped tight or spaced for expansion?

[ArthurEld]

And another thing, if you stack the cells flat, the lower cells already have compression.
Stick that in the calculation

 

[Dzl]

Keep in mind that the spring pressure will not change unless it’s spring bound or loose. If you set the pressure to 12psi with compression springs and they stay in their range you will have 12psi across the SOC range.

This is good to know, and seems like it will simplify implementation.

 

[toms]

As the cells expand the springs will apply more pressure and as they contract they will apply less pressure.

A linear spring will exert the same pressure as it is compressed slightly.

ie, if you get a 5kg spring and press on it against a scale the scale will read 5kg all the way until it is fully compressed.

 

[Deleted member 9967]

A linear spring will exert the same pressure as it is compressed slightly.

ie, if you get a 5kg spring and press on it against a scale the scale will read 5kg all the way until it is fully compressed.

Sounds expensive. But ok.

 

[BiduleOhm]

A linear spring will exert the same pressure as it is compressed slightly.

ie, if you get a 5kg spring and press on it against a scale the scale will read 5kg all the way until it is fully compressed.

No, springs don't work like that. A 10 N/mm spring will exert 10 N of force when compressed by 1 mm, 20 N for 2 mm, etc...

There's some special ones who have a really flat curve and have an almost constant force but that's very rare.

If you want a more flat curve you can use a less stiff and longer spring, and compress it over a longer distance.

For example compressing a 10 N/mm spring by 5 mm will give you a force of 50 N; if you now use a 2 N/mm spring you'll need to compress it by 25 mm to have the same 50 N force, but you also now have only 2 N of variation with each mm of change due to the cells expansion instead of 10 N/mm

 

[Colonel.lp]

A bit of real world observation. My new 4s pack I'm still testing was clamped with 4 m6 threaded rods and 10mm aluminium end plates. They were clamped pretty tight while fully charged, to the point I was scared about deforming them.

I've just done a capacity test on each cell, and at the end noticed the the bottom rods were loose by 1mm or more, but the tops were still tight.

So expect that the pressure will vary by 1 to 1.5mm of spring force per spring you have.

So for example I'm going to be having 16 springs (one each side of my 8 rods) so the force change will be:

16 X 1.5mm X 5kgf/mm X 0.5(as the 1.5mm is spread between the two springs on each rod)
=60kgf.

So the maths should work out the same no matter how many springs or rods as long as you are starting with the same 300kgf per side.

If you clamp at 50% charge (assuming linear expansion) that's 270kgf to 330kgf.

 

[John Frum]

I've just done a capacity test on each cell, and at the end noticed the the bottom rods were loose by 1mm or more, but the tops were still tight.

Anyway got a hypothesis on why their is a difference between the top and bottom?

 

[sremick]

Anyway got a hypothesis on why their is a difference between the top and bottom?

Total shot in the dark:

Gravity pulls more of the electrolyte to the bottom of the cell, so the bottom experiences slightly more expansion/contraction as the cell is cycled.

 

[BiduleOhm]

My guess would be top ones were tightened first and cells were already a bit swollen so they got closer at the top.

My recommendation would be to clamp them before even the first cycle and to tighten the bolts evenly (for those who know: follow a star pattern).

Gravity pulls more of the electrolyte to the bottom of the cell, so the bottom experiences slightly more expansion/contraction as the cell is cycled.

Forces at play are far far greater than gravity, so we can rule that out for sure.

 

[Colonel.lp]

Unfortunately my cells arrived charged, so had no choice with first clamping.

I would say the bottom swelling is to do with the cell sides not being flat. I can see daylight through the gaps in my clamped cells, and the top plates are a little wider than the area of cell under it. Thinking about it, the fact that many people have cells with undulating sides and concave areas, this will massively affect the usefulness of clamping.


It also makes me wonder if this is anything to do with why these cells are cheaper.

 

[BiduleOhm]

Charged to 100 %?

Concave sides will not matter much because as long as the cell isn't swollen the force will mainly be exerted on the cell periphery as it's the only part really rigid. Then when the internal pressure rises the side will unconcave and make contact with your clamping plate.

 

[Colonel.lp]

Charged to 100 %?

Concave sides will not matter much because as long as the cell isn't swollen the force will mainly be exerted on the cell periphery as it's the only part really rigid. Then when the internal pressure rises the side will unconcave and make contact with your clamping plate.

Yeah, well, 99% took about 6ah of charge while parallel top balancing 4 cells to get to 3.65V and 0amps. Ongoing discussions with Amy's replacement at Xuba.

Would damage not be occurring in the cells swelling to fill the gaps? Some are substantial.

 

[John Frum]

Unfortunately my cells arrived charged, so had no choice with first clamping.

I would say the bottom swelling is to do with the cell sides not being flat. I can see daylight through the gaps in my clamped cells, and the top plates are a little wider than the area of cell under it. Thinking about it, the fact that many people have cells with undulating sides and concave areas, this will massively affect the usefulness of clamping.


It also makes me wonder if this is anything to do with why these cells are cheaper.

WILD SPECULATION FOLLOWS:
Maybe the fancy test clamp in the production facility is part of the cell qualification process.
They measure the expansion and contraction during the first cycle.
Possibly the uniformity of the expansion and contraction also.
Just a guess, but I was wondering why they would go to the trouble of engineering a constant 12psi compression rig.

 

[BiduleOhm]

Yeah, well, 99% took about 6ah of charge while parallel top balancing 4 cells to get to 3.65V and 0amps. Ongoing discussions with Amy's replacement at Xuba.

Would damage not be occurring in the cells swelling to fill the gaps? Some are substantial.

Wow, that's weird. I never heard of any Li batteries being shipped at 100 % SoC.

What gap sizes are we talking here?

 

[Colonel.lp]

WILD SPECULATION FOLLOWS:
Maybe the fancy test clamp in the production facility is part of the cell qualification process.
They measure the expansion and contraction during the first cycle.
Possibly the uniformity of the expansion and contraction also.
Just a guess, but I was wondering why they would go to the trouble of engineering a constant 12psi compression rig.

Possibly, sounds an expensive process. More likely that engineers like to play with ideas, and if they have big contracts to cutting edge purchasers, that would be valuable in house specifications to be able to supply. As much as we see lots of sub standard articles from China, these battery manufacturing facilities are state of the art, and probably involve a lot of in house dev and testing.

 

[Colonel.lp]

Wow, that's weird. I never heard of any Li batteries being shipped at 100 % SoC.

What gap sizes are we talking here?

I know. Not happy. But we will see.

I can't measure at the moment as they are clamped and charging, but if I remember, if you put a straight edge over the sides, possibly 2mm concave just at the tops, might be 1mm. Will look when I next disassemble. I'm not the only one with this.

 

[BiduleOhm]

Should be ok but given you're in a weird case photos would confirm if it's fine or not

Mine came with concave faces as well, at most 2 mm, I just checked with a straight edge. I consider it fine, more than that would be worrying.

 

[Colonel.lp]

Should be ok but given you're in a weird case photos would confirm if it's fine or not

Mine came with concave faces as well, at most 2 mm, I just checked with a straight edge. I consider it fine, more than that would be worrying.

It's fine, I'm not concerned about the sides. Was just curious that if the whole point is to stop expansion, surely if there are significant areas that can expand, will that not have caused the damage already.

 

[Deleted member 9967]

It's fine, I'm not concerned about the sides. Was just curious that if the whole point is to stop expansion, surely if there are significant areas that can expand, will that not have caused the damage already.

Not a few times no. It is accumulative over a long period of time. Like months on an almost daily charge and discharge cycling.

 

[Sverige]

I asked a vendor I’m in contact with about this issue of whether or not to clamp cells today. They answered :

“If discharging at 1C or more then yes, it would be better to compress them. Because high current charge/discharge will cause the cell inflating. Otherwise, it is ok to make them without compression. Because if amps is large, much heat will generated, and the cells will inflate a bit.”

And they supplied the attached photo to illustrate a typical clamping arrangement. Note, this wasn’t specifically in connection with Eve 280Ah cells, just general info.