Compress or not, flexible busbar or not

[Cheap 4-life]

The negative electrode graphite, which is about 20% of cell, expands by 9-11% of its volume between zero and 100% state of charge.

That amounts to less than 2 mm thickness change on a 280 AH cell. Barely noticeable.

You seem very knowledgeable regarding this issue. More knowledgeable than I am. but from numerous sources, I have seen how much cells expand. That’s from normal/low c-rates and not over charging. Sometimes it’s a lot more than 2mm or at least it looks like more

 

[SparkyJJO]

To answer your question, not really. However as @cinergi and some others are saying, compression without flexible busbars can cause issues like busbar nuts/screws loosening or busbars sliding and or cell resistance/voltages being read incorrectly by the bms due to the loosening. Without flexible busbars it seems to me (from everything I’ve watched and read) that compression shouldn’t be used. Instead the snake pattern @Will Prowse is using or the space between cells that offgrid garage implemented should be done when solid busbars are used

That's like backwards of what I read/understood a couple months ago. What I understood was a fixture should be used to prevent cell movement when using rigid bus bars (and the cells do better with some compression). So I designed mine using rigid bars, with two rows of 8 cells, each cell separated by a plastic sheet for an insulator, and 3/4" ply end plates, using 6 threaded rods (2 on one side, 2 down the middle, 2 on the other side), and the nuts just snugged down at middle of SoC but not going too crazy.

Too much conflicting info out there on this

Screw it, I'm sticking with this config I guess ?

 

[Cheap 4-life]

That's like backwards of what I read/understood a couple months ago. What I understood was a fixture should be used to prevent cell movement when using rigid bus bars (and the cells do better with some compression). So I designed mine using rigid bars, with two rows of 8 cells, each cell separated by a plastic sheet for an insulator, and 3/4" ply end plates, using 6 threaded rods (2 on one side, 2 down the middle, 2 on the other side), and the nuts just snugged down at middle of SoC but not going too crazy.

Too much conflicting info out there on this

Screw it, I'm sticking with this config I guess ?

I here ya, my heads spinning trying to figure out what’s best. Preventing all movement can’t happen. It’s how much they move and if that’s enough to cause the busbars to loosen. To me it seems that If the busbars are welded (instead of held by screws or nuts) then compression is fine and should be used. Then the terminals will simply withstand the movement and flex. The busbars couldn’t loosen/slide etc due to being welded. I’ve seen some of the better made (more expensive) batteries using the welded busbars and compression of some sort.
Maybe with only 8 cells in a row the cells using mechanically held solid busbars wouldn’t move enough under compression to make re torquing of the busbars needed.

 

[Cheap 4-life]

Post in thread 'Eve lfp280 w/ overkill + victron settings'
https://diysolarforum.com/threads/eve-lfp280-w-overkill-victron-settings.11346/post-198413
Post in thread 'Cinergi's 28 kWh / 4 kW Solar / 10 kW inverter RV build'
https://diysolarforum.com/threads/c...lar-10-kw-inverter-rv-build.13786/post-457384

Both posts where solid busbars with compression caused the need for re torquing due to loosening.
There is a post where re torquing of flexible busbars was needed but they were the braided type not the laminated type.

 

[RCinFLA]

Do you see any harm in putting 8 cells between two pieces of plywood, separated by some sort of padding, and then severed between four pieces of threaded rod?

Honestly, coming from a world of technical data, procedures, and quality assurance, a lot of what we do can be worded as “best practices” by some and “cowboying” by others.

I would not use wood, but it is a good idea to have some amount of thermal barrier and if you are compressing cells have some spacing compliance 'flex' between cells. That way if one cell shorts out it does not create a cascading meltdown to adjacent cells, popping off the whole string.

Main reasons I am not an advocate for compression is:

1) You should not be using thick electrode cells regularly above 0.5C(A) current due to onset of electrode ion migration starvation, therefore little internal electrode heating, therefore little benefit of compression.

2) If you don't have some compliance in compression pressure between cells you can create a high-pressure point on cell surface causing a separator punch through causing cell to short out. Cell wraps do not have a perfectly uniform thickness.

 

[Nobodybusiness]

Like Andy’s meaning that you are using solid busbars and have roughly a 1/4” of space between your cells?

Yes..

 

[cinergi]

The negative electrode graphite, which is about 20% of cell, expands by 9-11% of its volume between zero and 100% state of charge.

That amounts to less than 2 mm thickness change on a 280 AH cell. Barely noticeable.

But amounts to 1.25 inches for a pack of 16. That's a LOT of pressure in a rigid fixture.
EDIT: If the fixture was snugly assembled at a low SoC

 

[RCinFLA]

But amounts to 1.25 inches for a pack of 16. That's a LOT of pressure in a rigid fixture.
EDIT: If the fixture was snugly assembled at a low SoC

Absolutely agree.

Also goes to point that compression can get you in a lot of trouble if you don't know all the in's and out's of what you are doing.

 

[brb58]

When I topped balance all my cells, they were in a fixture. None have mine have been without being in a fixture being charged or discharged.

My battery case has the end plates snug just to prevent any expansion. No springs!

 

[Cheap 4-life]

....

 

[RCinFLA]

My battery case has the end plates snug just to prevent any expansion. No springs!

Not a good idea to have hard fixed constraint.

Sum of each cell's small SOC width expansion from all the parallel stacked together cells can drive compression pressure up exponentially. Can result in crushed cell internal perforated plastic polyethene electrodes separator wrap, shorting out cell. Issue gets worse the more cells you parallel together.

You can also crush electrode material loose, increasing resistance to rest of electrode.

LFP cell construction is like two blocks of Rice Krispy treats (block represents positive LFP and negative graphite electrodes) with thin piece of plastic, with a lot of small holes punched in it, separating the two blocks to allow marshmallow to ooze through the small holes in plastic sheet (marshmallow represents electrolyte). If the block on one side of plastic makes contact to block on other side the cell is shorted.

Charging causes the granules of graphite in negative electrode to expand by as much as 11% in size due to the fully charged stuffing of lithium ions within the graphite lattice structure.

The structure of the positive LFP electrode is such that it does not shrink in size as lithium ions leave it during charging. The 'iron' in the positive LFP electrode provides a support pillar to maintain the LFP positive electrode lattice size after lithium ions have left for charging transfer to graphite negative electrode side. The iron in positive electrode lattice support makes LFP cells' cathode much more rugged than other lithium-ion chemistries. That is the primary reason LFP cells have much greater discharge-recharge cycles lifetime.

 

[Horsefly]

I also am not subscribed to the whole compression fixture thing. I've built four batteries, and all four have had something close to finger-tight threaded rods / bolts just to keep them from moving around and putting pressure on the bus bar connections. I think calendar time will kill my cells before the compression fixture would matter.

I hadn't noticed this thread earlier, but hopefully everyone here knows there are several fairly long and detailed threads here (circa 2020 and 2021) about building compression fixtures. Many folks have posted pics of what they have done and given links to what they bought to build them. Many others (like me) think that all the effort to compression the cells is more likely to damage the cells than extend the life of the cells. But that's just me.

Carry on.

 

[RCinFLA]

Unless you run your battery hard with high discharge and charge current (>0.5 C(A) current) there is little benefit to compression.

Compression helps reduce electrode material delamination from metal foil which is dependent on electrode heating/cooling cycles.
Cell internal heating gets worse the more cell current demand and the more aged the cells (2x-3x heating with age at given current).


Operating at cold temps (<5 degs C) increases internal impedance of cell which increases internal heating for given current and results in more temp cycle shifts.

Delamination shows up as impedance increase on a 1kHz battery impedance meter which primarily measures conduction resistance of cell.

 

[Wet1]

Given the cells significantly expand and contract, and EVE tests cell failure with a rigid fixture, it seems crazy to me that folks are using completely rigid fixtures. To each their own I guess.

EVE recommends compression for maximum longevity, and they basically give us the pressure guidelines. If the user chooses to use compression, it's illogical to use anything other than some form of fairly constant rate of compression (within their guidelines) across the SoC range (basically using springs, foam, or some other media which achieves this). Seems pretty simple, yet there's still ongoing discussion about using rigid compression... I'm not sure why this is even remotely considered a good, or even safe, practice.

 

[SparkyJJO]

I thought their most recent paper on that showed a rigid box, not some spring loaded thing.

 

[RCinFLA]

Given the cells significantly expand and contract, and EVE tests cell failure with a rigid fixture, it seems crazy to me that folks are using completely rigid fixtures. To each their own I guess.

EVE recommends compression for maximum longevity, and they basically give us the pressure guidelines. If the user chooses to use compression, it's illogical to use anything other than some form of fairly constant rate of compression (within their guidelines) across the SoC range (basically using springs, foam, or some other media which achieves this). Seems pretty simple, yet there's still ongoing discussion about using rigid compression... I'm not sure why this is even remotely considered a good, or even safe, practice.

Eve is doing high current discharge/charge tests. They also do not tell you the fixture provides cell cooling.

 

[Wet1]

I thought their most recent paper on that showed a rigid box, not some spring loaded thing.

Only in section 3 (IIRC), where they discuss testing failure modes.

 

[Ampster]

I suppose bloating is different than swelling.

Then there is expansion. I agree with previous posters it is a personal risk management decision. I have seen enough instances of swelling, expansion and bloating that I chose to use thread rod on my pack that is five feet long. My bus bars are ¼ inch solid copper 14:inches long.(3P16S)

 

[brb58]

Not a good idea to have hard fixed constraint.

So, I assembled my fixture while the SOC of all cells is at 100%. I also have 3D printed plastic spacers than are around 6mm thick that are created at 30% infill so it's a honeycomb structure. You still feel that is a big issue? This is for an RV so most of my discharge is less than 0.1C but occasionally will be at 0.3C but not for more than 20 minutes.

 

[noenegdod]

Not a good idea to have hard fixed constraint.

I agree with this completely.

Unless you run your battery hard with high discharge and charge current (>0.5 C(A) current) there is little benefit to compression.

Some will argue that eve 280 cells and the like have no business in a mobile application. Lets disregard that argument at this point.

Given that we both agree that a rigidly constructed fixture is not a good idea, how would you recommend accomplishing mounting eve 280 cells in a mobile application without compression? I built my spring fixture solely for this purpose. 1120ah battery seeing peak loads of 500ish amps for a few minutes at a time. I sized the battery to keep loads low to prevent heating and any of the discharge related issues at bay, but I still had to mount it.

In hind sight, I hate these cells for my intended application and for the time and effort I have into building the jig I should have just worked more and spent triple on cylindrical cells.