Mooch covered this, but I think it’s important to say the TLDR (or maybe too detailed didn’t fully comprehend) here which is that if you do discharge bypass you can’t actually cut off at a minimum voltage for a specific group. You’ve said you are going to do bypass so for 12s all you can do is set the ESC to shut off at 36-38.4 V and then check where the groups actually end up unless your BMS and ESC can communicate which the BMSes you’re looking at don’t do. If you stop at 36 V you might end up with all the groups between 2.9 and 3.1 V or you might end up with a better or worse spread. If you bypass discharge you can’t have a BMS cutoff.
3 Likes
Im about to add a few cells to my existing pack and I’ve got a question.
Its a 12S8P pack, should I rearange it into 8s12p and build 4s12p or should I just add 4cells to every pgroup?
Using differently aged cell in a pack is always risky. But I’d make new p-groups from the new cells, that way its easier to switch p-groups out if one of the old ones fails earlier than one of the new ones.
1 Like
I am doing it on my own risk. And Ive got llt bms so I am monitoring cells pretty often (its a habit I wold say)
Thanks for confirming voices in my head
1 Like
Agree with replacing the wrap and gentle Dremel to remove leftover metal. Removing the wrap will also let you inspect the outside of the cell. I bought older cells and some had some pretty good dents and one was punctured, but the wraps on them concealed it.
id mix the old and new in the same p group then stronger cells in the group keep the weaker parallel cells from going too low. also wont need as much balancing current and many bms balance with very low current.
But if you parallel aged and new cells wouldn’t that be a real waste of recources if just one of the old ones goes bad?
the weaker cells are there reducing your capacity one way or the other. if theyre in parallel with good cells its safer
1 Like
Yeah it would just decrease the life of the pack overall but at least it would be equal
Sharing load more effectively, but in his case, it’s not like he’s gonna disassemble and re-build the entire thing at the cell level, I would just buy new cells at that point
How I understand it @spei wants to do excatly that, rebulding the 12s8p to a 12p12s pack. That would requiere cell level disassembly, and if the common consensus is to mix the cells, then I’m not the one objecting to that. I just feel like separate new p-groups would be easier to deal with if one shits the bed.
Well technically he said 8s12p which made me think he would keep part of it assembled
But ya if he did cell level and wanted to keep the old cells (assuming they aren’t super bad) then mixing evenly is probably good.
Might even be worth doing a capacity test and repackr but that would be time consuming
i think one wont “shit the bed” if that crap cell is in parallel with strong cells keeping it in a safe voltage range. i dont know his particular case but generally thats how i understand it
how would that work, right now there are only 8p-groups. 
Lol idk what his pack construction looked like - but good point
I’ve got 12S8P pack. What I had in mind is:
From 12S8P to 8S8P, than split 4x 1S8P into 8x 1S4P, and lastly add them for a total of 8S12P
Its a brick pack, easy to reconstruct. Even easier to add new-reclaimed cells to old pack
1 Like
Now I’m confused. Do you want to add new cells or not? You make it sound like you want to run a 8S pack now.
Fuck me.
and add new-reclaimed 4s12p for a total of 12s12p. ta-daa
1 Like
why not just add the new weaker cells on the end of the strong p groups? that makes the most sense from a safety point.
1 Like
So its either add new to old and the old will keep new in shape, or reconstruct the pack, and the bms will have more work between old8s and new4s. Correct?
either add the weak to the strong, and the strong keep the weak from deviating too much, or its separate them and the bms has to do a lot more work.
i wonder on which would result in more capacity exactly but with mixing stuff its safer to take the safer route
2 Likes