Fixing Reention battery with unevenly discharged cell group

[mikel-]

Hello,

I just wanted to get a better understanding what could have caused the problem with my battery.
I have a new Reention Eel 48v 14Ah battery. I was not able to charge it above 52v (should be 54.5) and BMS was shutting down the output if I pulled more than 200W at 49v.
I opened the case, inspected the wires and measured cell voltages. It is hard to tell if any wires are partially broken. I did not see anything suspicious. I can measure the voltages OK from all cells on BMS connector.
The battery is 13s4p. I measured all 13 groups of 4 cells. Most of them are at 4.15v, but group #8 is at 3.8v.
I am trying to charge that group with RC battery charger, but need more time. So far it looks to be charging OK. Maybe another hour to full capacity. Maybe the cells are OK. After that I will try to discharge and charge the battery with a regular ebike charger and monitor the voltage.

What could have caused this problem?
Maybe that group was already discharged more than the other ones when the battery was assembled?
I am not sure how BMS works, but I have some experience with RC batteries. I am guessing BMS is similar to RC balancing charger. If 1 group is deeply discharged and the other 12 are normal, a balancer would have hard time equalizing the voltage. Too much energy to dissipate.
What would happen if the balancing wire (or pin in the connector) from group #8 was broken ? I can't measure the voltage on BMS PCB because it is covered in some kind of glue/resin and I am afraid of damaging something if I try to cut/pull it off.

 

[markz]

Probably have a weak group bringing the rest down.
Fix uneven voltages by charging individual groups, independently which I think means node charging. What I did was get plug in cell phone chargers and charge up the low ones to the top voltages.

BMS will balance the pack near the final voltage, at what point that is I do not know. I let my charge a long time, say 2hrs its within 0.5V, then it takes another 1-2hrs to complete the balance charge. That all depends on the balancing current, cheap 50-80W lip balance chargers take a very very long time to balance because their balance current is just so low, like 0.5A or something. And it deals with one inbalance at a time, very slow.

 

[mikel-]

Balancing current is probably 0.05A - 0.1A.
It can correct about 0.05 - 0.1Ah per hour of charging.
My group #8 was about 8Ah lower than other groups. So it would take 160 - 80 hours of charging to balance.
And my charger stops when most cells are fully charged. So I would have to discharge the battery many times.
Basically, it is almost impossible to balance my battery with BMS.
I found another unbalanced group. Will charge individually and see how it works.

 

[john61ct]

Yes, as high as half an amp is hardly ever encountered in normal (reasonably priced) BMS or resistance based hobby chargers.

Specialist dedicated balancers sometimes get that high, but usually slow way down as delta closes.

The big advantage is you can balance at any chosen point in the curve

not forced to hold the cells at the stressful top point for hours.

At a delta of more than 0.1V (bad design needs correcting) manual rebalancing is best

the ideal protocol being to atomize the pack balance with all cells in parallel at 1S.

 

[mikel-]

Thanks!

I am just surprised to find that abnormal groups in my battery seem to work fine now. I charged/balanced these groups manually and after some heavy ebike riding (about 20A draw) all the groups discharged to 3.6v and remain balanced. Now need to do a few more cycles to verify.
I used to deal with RC batteries that do not have BMS. By the time I notice that one cell is discharged significantly lower than others, it is usually already damaged (from being discharged below min voltage during normal use). And the fact that it was grossly unbalanced usually means some cells could be problematic from the start. Charging these cell manually does does not help. They will not perform as good as the others after damage.
Having BMS with voltage cut-off for each group is nice.

 

[john61ct]

Are you saying there are 4 BMS each 13S working in parallel?

 

[mikel-]

No, just 1 regular BMS. Maybe my last sentence was misleading.

In RC systems they usually have a battery cut-off (in the controller) that activates when battery voltage drops below certain level. It looks at total output voltage, not at individual groups or cells. It does not prevent discharge of single weak group or cell below min voltage.
If I understand correctly, BMS in ebike battery monitors individual groups and will disable output if any group drops below min voltage.

 

[john61ct]

I use the term cell/group for the 1S level voltage you mean.

Really functionally the same whether each is a single cell at 14Ah, or as in this case a group of four in parallel 3500 mAh each.

I just wanted to check it wasn't four separate strings, modular 13S sub packs paralleled at the 48V level, as I am interested in figuring out how to accomplish just that, either each protected by its own BMS

or wiring the sub pack balance leads together to use just one.

 

[mikel-]

I use the term group the same way. I guess everybody does.

Wiring balance leads together is probably fine.