Battery Repair, BMS?

[Gionnirocket]

I have a LiitoKala 48v/10ah battery that is only charging to 52.6v and shutting down at around 44v+/- when sagging to 43v+/- under load.
Some prior discussion on the pack here for those totally bored.
So after my ride today I decided to cut it open and see what's going on.
On first inspection @ 47.15v all groups were at 3.63v except for group #2 which was slightly lower at 3.59v
After a partial charge to 51v all groups were at 3.93v except for #2 which was now slightly higher at 4.05v
After an attempted full charge it shutdown down at 52.68v and all groups were at 4.04v except for #2 which was higher and at a full charge of 4.20v
This can be a malfunction of the BMS, but I'm thinking group #2 has a lower internal resistance and it's charging/discharging faster then the rest of the pack. Or there's a bad cell in the group basically causing the same effect.
So I'm thinking one work around would be to replace the simple BMS with a balancing BMS as it will cut charging to group #2 when it reaches 4.2v and then allow the others to catch up. This won't solve the faster discharge of that group but the battery as is already provides enough for my daily 20mi ride with at least another 10mi left in the tank.

Not really caring about salvaging it, but it's just another project to keep me out of trouble and keep my wheels turning with problem solving.
So.... am I missing something.. or does anyone have any experience with a similar situation? aaahem @tomjasz @harryS, @George S.
I'd ask Ravi K to chime in but I know he's jumping hoops right now trying to get Zens rolling

Current BMS

 

[George S.]

I have a LiitoKala 48v/10ah battery that is only charging to 52.6v and shutting down at around 44v+/- when sagging to 43v+/- under load.
Some prior discussion on the pack here for those totally bored.
So after my ride today I decided to cut it open and see what's going on.
On first inspection @ 47.15v all groups were at 3.63v except for group #2 which was slightly lower at 3.59v
After a partial charge to 51v all groups were at 3.93v except for #2 which was now slightly higher at 4.05v
After an attempted full charge it shutdown down at 52.68v and all groups were at 4.04v except for #2 which was higher and at a full charge of 4.20v
This can be a malfunction of the BMS, but I'm thinking group #2 has a lower internal resistance and it's charging/discharging faster then the rest of the pack. Or there's a bad cell in the group basically causing the same effect.
So I'm thinking one work around would be to replace the simple BMS with a balancing BMS as it will cut charging to group #2 when it reaches 4.2v and then allow the others to catch up. This won't solve the faster discharge of that group but the battery as is already provides enough for my daily 20mi ride with at least another 10mi left in the tank.

Not really caring about salvaging it, but it's just another project to keep me out of trouble and keep my wheels turning with problem solving.
So.... am I missing something.. or does anyone have any experience with a similar situation? aaahem @tomjasz @harryS, @George S.
I'd ask Ravi K to chime in but I know he's jumping hoops right now trying to get Zens rolling

Current BMS

View attachment 100087

It seems like you have 12/13th of a good pack. The only thing I can think of is to run the whole pack down and then charge that bad set and one good set, to compare capacity. The good sets are all the same so one set is as good as another. If there is a bad cells in a set of 3 it should be pretty obvious if you run it down from 'full'. I've charged through the balance leads, but these are just voltage leads for the cutoff. If you have a bad set of 3 cells maybe LK would send you a few of these cells, if you want to cut and paste. Your idea would work if the weak set is the same capacity as one of the good sets. Hope this helps. I've sure you are on the right track. This is pretty useful information about the pack.

 

[harryS]

This past summer, I did in fact, install two balance BMS in the hope of getting an unbalanced battery to work. The BMS did do their jobs.

On one battery, the weak group would hit the 3.0V LVC when the other groups were at 3.43V, and the BMS would max it at 4.07V when the rest were at 4.22V. Upon usage, the weak group would still hit 3V first. However, I had seen this group self discharge itself down to 2V earlier, so I decided to set that battery aside for safety. I plan to replace those cells someday.

The other one also had one weak group, I installed a balance BMS (from amazon), and after about 3 hours of waiting after the initial charge, the cells all equalized. The BMS has LED's to show which cells are low, but I also checked manually with a meter. On subsequent cycles using the battery on my wife's 500W ebike, the battery stayed equalized. I was happy with that, so we're using this battery. The other fortuitous thing was that the balance connector was the same on both the old and new BMS, so I only had to connect three wires.

 

[Gionnirocket]

Thank you gentlemen for your reply.

@harryS ...my 1 bad group isn't acting as a weak group as yours were. Here it seems to be both charging and discharging faster than the the rest of the pack. In my mind the only thing that I can think of is that it has a lower resistance or has a cell that has shut down. . Perhaps a balance BMS might make it a little more useful than it currently is... Do you mind posting a link to the BMS you bought?
That said, if it is a bad cell... It might be dangerous and I'm better off moving on. I don't think it possible to properly test without fully disassembling the group in question. Is there a way to take individual cell resistance and voltage readings with the group connected? I have to think about that, but I don't believe so.

@George S .. It can also be that I have 38/39 of a pack if only one cell has capped out, but LiitoKala was totally unresponsive to request for assistance both before and after opening a dispute. As far as I am concerned, they can keep their terds. There's really no reason to perform a full capacity test as it won't change anything. In its current configuration it doesn't act as the others and the last thing I'm going to do is try and match new/different cells to this terd. Add that I don't have spot welding capabilities and have no desire to add this to my list of toys. I received a 50% refund via Aliexpress and I'm pretty confident that my CC will make up the rest. I appreciate the input and I like experimenting and learning as I go but there's a limit to what I feel is practical.

I can't think how.... But is there anyway that this is only a BMS problem?

Thanks again gentlemen for your input... very helpful!

 

[George S.]

Thank you gentlemen for your reply.

@harryS ...my 1 bad group isn't acting as a weak group as yours were. Here it seems to be both charging and discharging faster than the the rest of the pack. In my mind the only thing that I can think of is that it has a lower resistance or has a cell that has shut down. . Perhaps a balance BMS might make it a little more useful than it currently is... Do you mind posting a link to the BMS you bought?
That said, if it is a bad cell... It might be dangerous and I'm better off moving on. I don't think it possible to properly test without fully disassembling the group in question. Is there a way to take individual cell resistance and voltage readings with the group connected? I have to think about that, but I don't believe so.

@George S .. It can also be that I have 38/39 of a pack if only one cell has capped out, but LiitoKala was totally unresponsive to request for assistance both before and after opening a dispute. As far as I am concerned, they can keep their terds. There's really no reason to perform a full capacity test as it won't change anything. In its current configuration it doesn't act as the others and the last thing I'm going to do is try and match new/different cells to this terd. Add that I don't have spot welding capabilities and have no desire to add this to my list of toys. I received a 50% refund via Aliexpress and I'm pretty confident that my CC will make up the rest. I appreciate the input and I like experimenting and learning as I go but there's a limit to what I feel is practical.

I can't think how.... But is there anyway that this is only a BMS problem?

Thanks again gentlemen for your input... very helpful!

I guess I would not use it. To check a cell or group you'd have to cut it off, cut the nickel around it and then do a test. Not much fun. You'd need to spot weld the strip if it checked out or you replaced a cell. You could jump that cell group with a wire. Cut the group off and then run a wire between the connection groups. You'd have a 12s pack, which would muck up your motor controller. It would shut down, probably half way down, but maybe usable. If you worked out how to cut the nickel and left room for solder on the strips, it would not be totally unsafe. You'd have to figure out the serial connection.

I would give you 50 bucks for the pack but shipping known defective packs is a big no-no. If you stripped out the cells, I figure you could get some money on Ebay. A lot of Ebay stuff is 'don't ask-don't tell' shipping.

If LK is unresponsive, that's a huge minus. It seems to be their business model. At least you got the 50%. I see a couple of Chinese companies that are clearly aiming for a long term kind of model. It's still bare bones and I don't know how they respond to problems. Reading the reviews, a lot of the LK stuff goes to Russia.

Thanks for sharing a 'bad news' story.

 

[Explorer-1]

I would just try discharging the high group until it is closer to the others an then recharging the pack. Or if you have patience your bus may balance the pack for you, leave it plugged in for 48hrs, measure the different again. The BMS is operating just as it should, no fault. or you can manually change each of the other groups to get them closer then discharge the entire pack to about 60% recharge and the BMS should be able to balance the pack a bit quicker with a smaller difference in the cells. but since you already have it off the bike, the difference seems like it is within the ranch of the bus to balance, so I'd just leave it plugged in for a couple of days.

 

[Gionnirocket]

I would just try discharging the high group until it is closer to the others an then recharging the pack. Or if you have patience your bus may balance the pack for you, leave it plugged in for 48hrs, measure the different again. The BMS is operating just as it should, no fault. or you can manually change each of the other groups to get them closer then discharge the entire pack to about 60% recharge and the BMS should be able to balance the pack a bit quicker with a smaller difference in the cells. but since you already have it off the bike, the difference seems like it is within the ranch of the bus to balance, so I'd just leave it plugged in for a couple of days.

If you look back at the initial post you can deduce that the voltage of the high group did intersect and match the rest of the pack at around 48v.
The problem with leaving it plugged in for a couple of days is that it will continually top off the weak group during this time and you'll basically be cooking it while the rest catch up. Not the best thing for an already weak group. I think a safer approach would be to implement a balance BMS as this will greatly reduce the time it takes to get them all in line. But even then I'll still be left with a group that discharges quicker than the rest but it would probably give me some added drive time. Until I get a replacement or try replacing the BMS with a balance one, I think it better to charge to about 51v which is giving me enough capacity for my daily ride without stressing the weak group further

 

[Gionnirocket]

The other one also had one weak group, I installed a balance BMS (from amazon),

Is this the BMS you installed... Seems to meet your description.
My only concern in my case is that it is rated at 20A... a little low for a BBS02B, even though I have it restricted to 20a at the controller.

 

[harryS]

I'm running it on 20A controller, not the BBS02. For a few days, the LED's took a lng tome to go out after a recharge, and the resistorw would hit about 120Fm but they either burned out or the cells are staying in balalnce.

looks like same bms.

 

[Gionnirocket]

I'm running it on 20A controller, not the BBS02. For a few days, the LED's took a lng tome to go out after a recharge, and the resistorw would hit about 120Fm but they either burned out or the cells are staying in balalnce.

looks like same bms.

Thanks Harry

I like the other parameters of the bms... but I think I'm going to search around for something with a higher continuous current rating.
That said I'm not sure this battery is worth the effort as it's weakness is on both charge and discharge rates, not just a cell sitting at a lower voltage or just out of balance