Adding parallel cell group into existing battery pack

[chronosgarage]

I’m building a 20S9P battery pack from recycled laptop battery. The battery have a cell level fuse.

The battery worked fine until there is a short in the votol em100 controller. When the controller shorted out there are 4 cell fuse that is blown, the bms is also blown. The 4 battery have a black burned marks on it. When I check the voltage, the battery was in normal voltage. So i Replace the bms and put another cell fuse to the 4 battery. (The photo is just as reference).

The problem is when I use the battery, the bms keep shutting off. When I’ve check the cell group voltage, that group that has the cell fuse replaced is causing the low voltage cut off on the bms. The group do not have a high cell Self discharge.

I’m thinking of making a 1S4P battery pack and connect the battery to the damaged cell group in parallel. Because to replace the 4 pcs of 18650 battery means that I have to dismantle the battery pack.

What do you think about this solution?
Any help is appreciated.

 

[goatman]

so youre going to disconnect the fuses on the 4 bad cells, build a 4p group and connect it to that bus bar to match the mah capacity of the other parallel groups.
im not a stranger to mickey mousing things together in a pinch but i would cut the pack in half where the bad parallel group is and just fix it

how many amps is your controller?

 

[john61ct]

Putting a **weaker** pack - higher ESIR, lower capacity - in parallel with a better one

will do no harm.

But if one truly fails, cells develop internal shorts, go negative voltage

that can damage the other connected sub-packs.

Just one string of health groups is really the best way to go.

This demonstrates how useful it would be, to just be able to easily atomize the pack, cap test maybe ESIR each cell, sort Go/No-Go and reassemble back into a pak composed of only good cells.

 

[chronosgarage]

so youre going to disconnect the fuses on the 4 bad cells, build a 4p group and connect it to that bus bar to match the mah capacity of the other parallel groups.
im not a stranger to mickey mousing things together in a pinch but i would cut the pack in half where the bad parallel group is and just fix it

how many amps is your controller?

I think cutting the pack in half is much hassle. Thank you so much for your reply.

My controller is 100A 72V controller. But I limit the current to 30A. I’m building second battery right now 20S12P with bms that will be paralleled with the main battery, so that i can set the current limit to 60A.

 

[chronosgarage]

Putting a **weaker** pack - higher ESIR, lower capacity - in parallel with a better one

will do no harm.

But if one truly fails, cells develop internal shorts, go negative voltage

that can damage the other connected sub-packs.

Just one string of health groups is really the best way to go.

This demonstrates how useful it would be, to just be able to easily atomize the pack, cap test maybe ESIR each cell, sort Go/No-Go and reassemble back into a pak composed of only good cells.

Thank you for your reply

 

[goatman]

youre only max bursting at 3.5amps/cell so 4p is 14 amps from that 4p.

if youre building all your packs from recycled cells id build the packs in 4p and parallel the the packs together.

i actually thought of building cubes of 4s4p and just connecting them in series and parallel, when i was using recycled cells so that when one cell goes bad you dont wind up in this situation

 

[chronosgarage]

youre only max bursting at 3.5amps/cell so 4p is 14 amps from that 4p.

if youre building all your packs from recycled cells id build the packs in 4p and parallel the the packs together.

i actually thought of building cubes of 4s4p and just connecting them in series and parallel, when i was using recycled cells so that when one cell goes bad you dont wind up in this situation

That’s a great idea. Thank u

 

[miro13car]

you limit the current to 30A as you write.
so those jumpers are way too thick , soldering them put real thermal stress on those poor cells.

 

[chronosgarage]

you limit the current to 30A as you write.
so those jumpers are way too thick , soldering them put real thermal stress on those poor cells.

I see. Thank you for your reply.