Repairing battery pack, mixing a different few cells

05silgto said:

Anyway, I know mixing 1500mah and "3400mah" cells in P isn't ideal if the groups aren't even but I don't think it will matter much if it's just one cell since it will be slightly less mah than the other groups.

Click to expand...

Well, it would be best to test the good parallel groups and build your replacement group with the same aggregate capacity. But there's probably already equivalent variance in the other groups, so I wouldn't sweat it.

05silgto said:

Thoughts? I've heard that it will just pull current from the cells with the highest voltage.

Click to expand...

Cells in parallel just become one bigger battery. They'll all have the same voltage.

05silgto said:

Also, please remind me, do I need to worry when connecting fully charged cells to other cells that are less charged?

Click to expand...

It's best practice to get the cells to the same charge before connecting, but low capacity, low discharge rate cells are probably fine to do once while assembling.

Thanks for the replies. I repaired the battery, originally I made the battery by combining two 48v batteries to make 72v. I didn't have a spot welder so I soldered them as best I could. this time I used a spot welder.
Well the strange thing is, after repairing it, it would only charge for a couple mins. and then my charger was saying it was fully charged. So, today I went for a ride, it was showing 78v instead of 84v. I don't think it fully charged.
Anyway, I went for a 4 mile ride, I rode it very moderately as I was riding with someone that didn't have a ebike. After, I hit 4 miles the bike shut off at 76v. It wouldn't turn back on unless I unplugged the battery and then plugged it back in but then It would just cut off again after little use. Seems like the bms is cutting off the power at 76v for some reason. I just got home after pedaling a 100lb fat bike 4 miles, which isn't fun. I plugged the battery in and I'm hoping it charges to 84v and fixes itself.
Best guess is I have a few bad cells or a bad bms, thoughts? Pictures here.
https://ibb.co/FsmL2nm
https://ibb.co/GFYVwBB
https://ibb.co/P9tZnkw

I note that you haven't listed all the individual group voltages, only the main pack output voltage. Presumably you had tested all of those before the repair process, and afterward?

Then presumably also tested the interconnects between the groups and the new BMS, at the BMS board itself, to be sure they are all working, and the BMS is correctly reading each group?

Also, presumably you balanced the entire battery, and then fully charged it, and then verified it was still balanced, before using it?

amberwolf said:

I note that you haven't listed all the individual group voltages, only the main pack output voltage. Presumably you had tested all of those before the repair process, and afterward?

Then presumably also tested the interconnects between the groups and the new BMS, at the BMS board itself, to be sure they are all working, and the BMS is correctly reading each group?

Also, presumably you balanced the entire battery, and then fully charged it, and then verified it was still balanced, before using it?

Click to expand...

No, no and no. Serves me right I guess for rushing through it. I just took everything apart to get into the battery again. I measured each group of cells and I got 4.2v in some cells,4v in some, and 3.6v in others. Will balancing each group of cells and putting it back together fix it? I don't have much experience with building batteries, especially with mixing cells in a repair job.

Edit: Now that I think about it, this makes me think my bms is not working as it should, I know I wired it right. Looks like I might get a different bms.

Remember your BMS can only do a little bit of balancing at any one moment. You have cells that are more empty than full, and cells that are almost full, and cells that are full. Most typical BMSs will take days to weeks to fix that kind of imbalance, if you leave the pack on the charger full-time. They're not designed to fix this kind of problem, as a pack that is this imbalanced would need to be repaired.

If you manually charge up the low cells, or manually discharge the full ones, or both, so that they are within a tenth of a volt or so of each other, the BMS could do it's job much easier and quicker.

amberwolf said:

Remember your BMS can only do a little bit of balancing at any one moment. You have cells that are more empty than full, and cells that are almost full, and cells that are full. Most typical BMSs will take days to weeks to fix that kind of imbalance, if you leave the pack on the charger full-time. They're not designed to fix this kind of problem, as a pack that is this imbalanced would need to be repaired.

If you manually charge up the low cells, or manually discharge the full ones, or both, so that they are within a tenth of a volt or so of each other, the BMS could do it's job much easier and quicker.

Click to expand...

Thanks, I'm working on charging all the groups and I currently have everything at about 4v or greater but the bms doesn't seem to be doing much if at all. I left it plugged in overnight with no change on any cells. It's a daly 20s 72v 50a bms.
Do you or anyone else know of some kind of 20s balance board I can install in conjunction to my bms?

I don't know of anything like that; you could try to search ebay or aliexpress for such things, if you really want to do it that way.

All the little balancers I have seen wouldn't do any higher balancing currents than the typical BMS would, anyway, so it wouldn't change your balancing time to replace your BMS with just a balancer board.

To use the balancer(s) *and* your BMS you'd have to add a whole new set of balance wires and connector for the balancer(s).

It's typically simpler to either charge up the low cell groups manually (usually before you install them into the pack, using single-cell chargers, but you can use alligator clips from such a charger to do it to the whole group) or discharge the high cell groups down manually with a load (low ohms high power resistor, car light bulbs or headlights, etc). Once all the cells are within a tenth of a volt or so (especially at the high end, 4.1-4.2v), balancing by the BMS goes a lot faster.

There's numerous battery repair threads where those are discussed; I don't have direct links but they are generally in the battery forum; some in the ebike troubleshooting or technical forums, some scattered wherever someone felt like posting.

Thanks, I'm using a old 3.7v rc helicopter to discharge the high volt groups and a smart 3.7v charger to charge the low voltage groups. I have a few cell groups which won't charge to 4.2v. One stopped at 4.03v and another at 4.06v. All my batteries are at at least 4v at this point. I'm guessing when I had a cell group fail, along with my bms suddenly frying shortly afterwards, It damaged some of these cells which is why they are worn out and can hold limited voltage. I haven't used this battery much since I made it and it always charged to 84v before the damage.
I'm going to plug in a 72v to 12v converter hooked up to a 12v 60w led tonight after charging the rest of the groups to test the capacity and to make sure they recharge evenly. If I don't get the capacity I'm looking for or if the cells don't charge evenly, I'll just end up buying 100 tesla model 3 cells and make a new battery.

Edit: I'm starting to think that my 3.7v smart charger wasn't designed with this kind of capacity in mind and it cuts off prematurely thinking the cells aren't taking any more charge.

I discharged the battery overnight. In the morning it was cut off at 79v. All the groups were around 4v except for 1 that was at 3.4v. I have 3 extra cells leftover, I'm gonna try adding them to that group to see if it fixes the problem.

youre going to find the bad cell in that p-group first and disconnect or remove it right?

you could just remove that p-group and have a 19s pack

goatman said:

youre going to find the bad cell in that p-group first and disconnect or remove it right?

you could just remove that p-group and have a 19s pack

Click to expand...

The cells in this specific group charged to 4.2v and stayed there. They only depleted when I discharged the battery, I'm guessing the batteries in this group are worn out and have little capacity vs being completely bad. I was gonna just add the cells as is.

I added 3 cells to the weak battery group, I had to solder wires to connect the cells cause Im out of room in my battery box width wise. I discharged the battery overnight again with my 60w led light. When I woke up, the light was still on so that's a good sign. The battery was at 79v and the weak cell group was at 3.7v. The next weakest group is at 3.9v. It's not perfect but already much better. The weak cell group was only at 4.13v at the time of discharge so there might be room for a slight improvement.