Another Cell Compression Thread, this time about foam

[Ron-ski]

I should hopefully be assembling my 16S pack this weekend, can someone check I've worked this out correctly please?

The cells are in two rows of 8.
I'm using 1/4" Poron foam (86375K234) each end, so two pieces of foam.
I have 9 sheets of 1mm thick G10 for between cells/end plates.
The end plates are 200 x 386 x 10mm aluminium

8 cells measure 573mm (I'll recheck this)
The 9 sheets of G10 add 9mm
Two pieces of foam is 12.7mm, 25% compression is 3.175mm, so the foam adds 9.525mm

So the gap between the two end plates should be 573+9+9.525 = 591.525mm (not that I can measure that accurately!)

Does the above seem correct?

Next question, with 10mm aluminium end plates do you think I only need 6 rods to hold it all together or go with 9 rods?

Center two rods are 6mm stainless threaded bar, outer rods are 12.7 x 3.2 aluminium tubes, tapped to take an M8 bolt.

 

[justgary]

@Ron-ski - Your calculations look good, except you might possibly consider a few things:

1) I'm not sure I see the benefit of using G10 separators next to the Poron foam, so you could drop it down to seven sheets for a total of 7mm thickness.

2) I found through testing that it might be better to compress the Poron 40% instead of 25% (as shown in Post #77 of this thread).

3) I built my batteries using aluminum bar rather than rod so that I could put two of them next to each other an still make a 16S battery with the same flexible bus bars. I did it mainly so I could lift each 8S part if I needed to.

Other than that, I think you are on the right track. I squeezed eight cells in Erwin (Pony would do) clamps (with end plates) to measure the total distance (taking an average of my measurements), then added the thickness of the separators, then subtracted the calculated thickness of the Poron. I cut the side bars to length and drilled them, then squeezed the whole thing together with the clamps again to install the screws.

 

[Zwy]

I should hopefully be assembling my 16S pack this weekend, can someone check I've worked this out correctly please?

The cells are in two rows of 8.
I'm using 1/4" Poron foam (86375K234) each end, so two pieces of foam.
I have 9 sheets of 1mm thick G10 for between cells/end plates.
The end plates are 200 x 386 x 10mm aluminium

8 cells measure 573mm (I'll recheck this)
The 9 sheets of G10 add 9mm
Two pieces of foam is 12.7mm, 25% compression is 3.175mm, so the foam adds 9.525mm

So the gap between the two end plates should be 573+9+9.525 = 591.525mm (not that I can measure that accurately!)

Does the above seem correct?

That is correct; total cell length + the separators + compressed foam thickness= length between plates.

Next question, with 10mm aluminium end plates do you think I only need 6 rods to hold it all together or go with 9 rods?

I use 2 rods on each side of 5/16" rod of my 16S packs.

Center two rods are 6mm stainless threaded bar, outer rods are 12.7 x 3.2 aluminium tubes, tapped to take an M8 bolt.

M8 on all would be plenty. As for using an aluminum tube, I'd reconsider unless you have a way of insulating the tubes from the cell walls. On my packs, I use PVC 1/2" conduit slipped over the 5/16" rod. https://diysolarforum.com/threads/compressed-pack-length.51671/


View attachment 131209

 

[Ron-ski]

Thank you for the replies, as for the tubes and studding, I have heat shrink to go over them.

 

[Honuz]

I should hopefully be assembling my 16S pack this weekend, can someone check I've worked this out correctly please?

The cells are in two rows of 8.
I'm using 1/4" Poron foam (86375K234) each end, so two pieces of foam.
I have 9 sheets of 1mm thick G10 for between cells/end plates.
The end plates are 200 x 386 x 10mm aluminium

8 cells measure 573mm (I'll recheck this)
The 9 sheets of G10 add 9mm
Two pieces of foam is 12.7mm, 25% compression is 3.175mm, so the foam adds 9.525mm

So the gap between the two end plates should be 573+9+9.525 = 591.525mm (not that I can measure that accurately!)

Does the above seem correct?

Next question, with 10mm aluminium end plates do you think I only need 6 rods to hold it all together or go with 9 rods?

Center two rods are 6mm stainless threaded bar, outer rods are 12.7 x 3.2 aluminium tubes, tapped to take an M8 bolt.

View attachment 131209

I did a 16S pack and did not used rods in the middle, seems too dangerous to me ... just 2 rods, one on each side (10mm rods) and with a 30x40mm steel tube on each side.
Left and right, those panels can move freely. The others, 2 sides and under can't move.
The boards on each sides, where the steel tube is, are rigid enough ..
I find it better to concentrate "strenght" in the middle, cause it's where the cells bloat.

 

[Ron-ski]

@Honuz I put two layers of heat shrink on the 6mm studding, the G10 also sticks out past the cells, so this will also keep the studding away from the cells. My battery is in a pull out drawer/rack, in my garage, so very easy to check periodically.

I used 10mm aluminium plate end plates, I tightened the outside tubes first, putting a straight edge across the plates revealed they were bowed, measuring confirmed this, tightening the 6mm studding pulled it straight again. Adding a cross tube like you have would have alleviated the problem, but I do wonder if the top and bottom would still be bowed, and I don't have enough space to do that.

I'll post some pictures later.

 

[Zwy]

@Honuz I put two layers of heat shrink on the 6mm studding, the G10 also sticks out past the cells, so this will also keep the studding away from the cells. My battery is in a pull out drawer/rack, in my garage, so very easy to check periodically.

I used 10mm aluminium plate end plates, I tightened the outside tubes first, putting a straight edge across the plates revealed they were bowed, measuring confirmed this, tightening the 6mm studding pulled it straight again. Adding a cross tube like you have would have alleviated the problem, but I do wonder if the top and bottom would still be bowed, and I don't have enough space to do that.

I'll post some pictures later.

On my 24V pack with two rows of cells, I did have center rods. I made the fixture tall enough in height to allow for a rod at the top and bottom. The bottom rods have an aluminum tray over the top. Upper rods have 1/2" electrical conduit for insulation in case a cell did make contact.

 

[Alkaline]

DALY? Really??

 

[Zwy]

DALY? Really??

Works flawlessly too.

That is a 24V 250A BMS and I've never loaded it more than 51A. If you derate the Chinese stuff, it works good. One of these days I'll get to replacing the ANL fuses with Class T's.

I bought the display also, mounted it by the remote inverter power switch on the right. I really like it, perfect for the application. Keeps cells well balanced too. It's been in there for over 2 years, close to 3. Only problem I had just recently was I went to check on battery charge as I never shut off power to the buck converter. Truck and camper parked in the back shed. Flipped on light switch for a draw and nothing. Thought battery was dead. Put the unit in the heated shop overnight, next morning it showed around 80% charge. The Discharge Cutoff Temp was factory default at -15C. It's been way colder than that here. So I know the temp sensor works good.

 

[Alkaline]

Every thread that has some type of Kaboom always seems to have Daly, I hope they work for you, but look in the learn my mistakes section...DALY is there almost every time.

 

[Zwy]

Every thread that has some type of Kaboom always seems to have Daly, I hope they work for you, but look in the learn my mistakes section...DALY is there almost every time.

Many are a case of not derating the Chinese stuff.

I use a Chinese buck converter too.

Works great, many said it would have high idle draw, fail, etc. Not an issue.

And even a Growatt inverter.

 

[Checkthisout]

Can you just stick the cells in a tight-fitting box? The cells will experience compression if they start to bulge.

Or, do you need to actually preload the cells with the given force?

 

[SilverbackMP]

Can you just stick the cells in a tight-fitting box? The cells will experience compression if they start to bulge.

Or, do you need to actually preload the cells with the given force?

That’s what I am doing but plan on lining in with 1/4 hard rubber roll floor mats picked up at the local farm supply. One side of the box will have visible screws and no glue so I can easily remove them if they swell. Each cell will be wrapped in Kaptom tape.

No idea if this is a good plan or not so YMMV.

 

[Checkthisout]

That’s what I am doing but plan on lining in with 1/4 hard rubber roll floor mats picked up at the local farm supply. One side of the box will have visible screws and no glue so I can easily remove them if they swell. Each cell will be wrapped in Kaptom tape.

No idea if this is a good plan or not so YMMV.

Thank you.

Just curious if the objective is to "pre-load" or "maintain shape".

 

[SilverbackMP]

Thank you.

Just curious if the objective is to "pre-load" or "maintain shape".

My objective is to maintain shape while allowing minute movement but not affect bus bars. I don’t know if this is right approach but it’s what I’m doing. Part of the reason is for siize of the boxes. I need to fit up to 10 complete batteries in a metal wall cabinet that is 46w x 72h x 18d (inches). So I can just squeeze two on a shelf (with small space at the end of each box for BMS, like the Digital Mermaids). Using threaded rod would add at least three inches.

 

[Checkthisout]

My objective is to maintain shape while allowing minute movement but not affect bus bars. I don’t know if this is right approach but it’s what I’m doing. Part of the reason is for siize of the boxes. I need to fit up to 10 complete batteries in a metal wall cabinet that is 46w x 72h x 18d (inches). So I can just squeeze two on a shelf (with small space at the end of each box for BMS, like the Digital Mermaids). Using threaded rod would add at least three inches.

Is retention or compression only necessary along 1 axis or does it also need to happen on the sides?

 

[SilverbackMP]

Is retention or compression only necessary along 1 axis or does it also need to happen on the sides?

From my understanding it’s one axis. I’m lining both ways with that rubber (even though it’s a sacrifice of space) because electricity scares me. Lining that cabinet too (mesh side areas will be flex sealed. And I’ve used the same material for the floors of my solar shed.

 

[justgary]

Thank you.

Just curious if the objective is to "pre-load" or "maintain shape".

The datasheet on most cells indicates 300kgf preload at 30%-40% SOC.

 

[Ron-ski]

Still need to connect the balance leads, but its getting there.

Edit @justgary I went with 7 sheets of G10 and 40% compression, only thing I'm not sure on is the state of charge, but voltage is just over 52v

 

[Checkthisout]

Still need to connect the balance leads, but its getting there.

Excellent craftsmanship. Thanks for sharing

 

[Hedges]

Still need to connect the balance leads, but its getting there.

Edit @justgary I went with 7 sheets of G10 and 40% compression, only thing I'm not sure on is the state of charge, but voltage is just over 52v

I like that style of welded terminal. Seems like the only one we've seen that is welded the way intended for busbars in packs.
And, with two studs, prevents cable from loosening the bolted connection.

Is it still just plain aluminum with native oxide on the face terminals bolts to? I wish it was tin, chromate, or some other electrically conductive plating.

 

[Checkthisout]

I like that style of welded terminal. Seems like the only one we've seen that is welded the way intended for busbars in packs.
And, with two studs, prevents cable from loosening the bolted connection.

Is it still just plain aluminum with native oxide on the face terminals bolts to? I wish it was tin, chromate, or some other electrically conductive plating.

I got a quote for those double terminal cells for 167.00 each with shipping.

Kinda spendy but they look so much more robust.

 

[Ron-ski]

I like that style of welded terminal. Seems like the only one we've seen that is welded the way intended for busbars in packs.
And, with two studs, prevents cable from loosening the bolted connection.

Is it still just plain aluminum with native oxide on the face terminals bolts to? I wish it was tin, chromate, or some other electrically conductive plating.

I'm not sure what it is, its a thin coating of something, I cleaned it off with a scotch pad and applied a very thin layer of silver conducting grease. These are my first cells, but the twin stud design makes so much more sense, more surface contact area and two studs are better than one.

I got a quote for those double terminal cells for 167.00 each with shipping.

Kinda spendy but they look so much more robust.

I purchased my cells from Fogstar in the UK back in August for £167 each (genuine grade A), they are actually cheaper now, I do need to order another 16, but not just yet.

 

[Zwy]

Every thread that has some type of Kaboom always seems to have Daly, I hope they work for you, but look in the learn my mistakes section...DALY is there almost every time.

False

You don't own one but you consider yourself and expert on them.

If you are referring to this thread, it would happen with any FET based BMS. https://diysolarforum.com/threads/why-not-to-use-daly-bms-with-mppt-controllers.24922/

 

[Hedges]

the twin stud design makes so much more sense, more surface contact area and two studs are better than one.

I wouldn't worry at all about area. Contact area of surfaces clamped by bolts is vanishing small. But heat escapes in three dimensions.
Cross section area only matters for a long run. For instance, conduit fill reduces ampacity of wires by as much as 66%, but does not apply for 2' or shorter run.

To me the biggest issues are contact resistance, especially due to native oxide. And motion which can cause loosening.
I see you have two separate cables, each with a lug having one hole. Just make sure those can't get moved, rotating and wearing loose.
I've seen some lugs with two holes.
The plate on top does mean the nuts won't get turned by the cable, which is good.