Custom Battery Housing, & Connecting Prismatic LiFePO4 Cells

[teaux]

Hey Endless Sphere,

I'm working on my first bike, and I've done something different with my battery that I'd like your opinions on! I disassembled a 20AH 48V battery, composed of 32 10AH prismatic LiFePO4 cells. They are now reconnected in my case in the same configuration (16 parallel groups).

The difference is that I've used all-wired connections (crimp-on eyelets) instead of busbars, in order to position all of the cells in my custom-built housing. The BMS circuits are connected this way too, with all of the circuits connected in the same order as in the factory battery configuration.

Can you see any problems with wiring the cells this way, instead of using busbars? I have used 10 gauge copper wire throughout. I have an eyelet bolted onto the each of the four cell posts in each group. Those four wires are then connected together using very strong press-in style blocks. My blocks have only four slots each, so I have connected the BMS wire in each group to one of the eyelets on the cell posts.

Pictures Attached. In the pictures, the orange wires are to the BMS. Red and blue are connections between the cells. They are colour coded by length, and not for any electrical reason. Before I installed the cells in the case, I also stretched cut sections of inner tube over the tops of the cells to prevent any shorting between cells, which are in close proximity to one another.

My number one concern in asking for your advice is safety of course!

This pack will power two Golden Motors MP4 hub motors.

Thanks!!

 

[amberwolf]

What mounting do you have securing the packs into their locations?

Or do they just sit bouncing against each other inside the case? :?

 

[teaux]

It's a bit tough to see, but they're padded in selected places with strips of adhesive foam board. There's actually a second crossbar running through the middle of the case that restricts motion as well. They aren't individually restrained, but I can tell you that there's no room for them to move in there. The packing is very tight. I can tip the frame over with the open side facing down without anything falling out.

 

[amberwolf]

FWIW, if it's likely to see any shock or vibration (normal even on smooth streets) you might want to hard mount them away from each other, and/or ensure no wires are routed between any of them.

Padding and foam (even closed-cell) can squish quite a lot, letting things move way more than you'd expect, and if it pinches a wire in there, or worse pinches two together hard enough to break their insulation and short them, bad stuff could happen.



I have some blue closed cell foam in my packs that is strong enough I can't squish it to less than 3/4 of it's volume by hand, but over time and shocks the stuff at the bottom and side of my ammocan packs has been compressed to less than 1/4, maybe even 1/5 of it's volume, in places.

I also use mousepads and other similar closed cell foam, and that is better, but still gives quite a bit with shocks on hard sharp edges.

 

[teaux]

Thanks, I'll look for some of the closed cell foam you mentioned. Maybe the safest bet would be to completely fill all of the void space. A short is definitely my main concern, after once accidentally seeing what these batteries can do when the wrong wires are crossed.

I've installed my motors and tested it out briefly today and it seems to be running beautifully. It's currently charging outside under close supervision (first charge since the re-wiring).

 

[amberwolf]

Oh, also, if these cells were in compression in their orignal pack, you should keep them in compression the same way in their new groupings.

Otherwise it's possible for them to swell and have issues of various kinds. Not all flat (prismatic) cells have that problem at all discharge/charge rates, but those that do can end up with serious problems over time, and the higher the charge or discharge rate is, the worse they can be. There's a number of threads discussing it if you look up compress or compression.