Cell group down to 2.2V, Toast?

docw009

1 MW
Joined
Aug 2, 2015
Messages
2,387
Location
Illinois, Not far from Wayne's World.
I have an unbalanced pack. Opened it up to find 9 groups at 4.07V, and one group at 2.20 volt. For academic purposes. I brought it up to 3.96 volt and it may go higher, but I stopped charging.

There are various things I can do going forward, like bringing all the cells to max charge and seeing if the cells all hold up, I can also see if the group drains just sitting there, or if the BMS was draining it. I can replace the BMS if that were the case

All moot if the cell group has been damned by its trip to low voltage land.

What say you guys. Is the battery damaged goods now? They are chinese generic cells, so no spec sheet is available.
 
Just revive it and throw on some cell monitoring for the questionable group as well as the group before and after it.
Hopefully you can view the sag live as you ride, then drain it to LVC and see what the voltages are.
If it didnt sit at 2.20V for to long it just might recover well.
Since you have it open, think about replacing the bad group as it all depends on how much work you want to put into it.
 
Add an active balancer board, and then it won't matter if that cell group is a little slow.
 
Threads like this make me want to avoid a BMS altogether and just get a display that shows individual cell voltages. Or maybe use 2 cell checkers and some switches to disable them when not in use.

On a slightly related note I've heard that Daly BMSs do not play well with MPPT solar charge controllers.

https://www.youtube.com/watch?v=JbnFyU79Nf0
 
Yes, it'd be cool if someone came up with something, like a display you can install and look at live as you ride or something you just plug in after a ride to check all the strings out.
Well there was that one product from Hobbyking you could plug into the balance leads and see maybe 6S or 8S, so having 2 or 3 would be the alternative, but I'm just saying be cool to have a discrete unit on the handlebar to look at active voltages as you ride up a hill.

Maybe one day Justin from Grintech will develop a top quality BMS and sell his potting material. There's just to many different BMS' out there to know which ones are good, just have to go on what people say in the forums but people keep their cards close when they do find a good BMS seller.

TrotterBob said:
Threads like this make me want to avoid a BMS altogether and just get a display that shows individual cell voltages. Or maybe use 2 cell checkers and some switches to disable them when not in use.

On a slightly related note I've heard that Daly BMSs do not play well with MPPT solar charge controllers.

https://www.youtube.com/watch?v=JbnFyU79Nf0
 
I got one of these just to play with and use for a future project but its a bit too cumbersome for most ebike builds. Doesnt run off of cell voltages and takes a separate power input for the display.

https://www.aliexpress.com/item/1005001428342027.html

A more compact version of this that would mount to the handlebars would be great. There are BMS that offer individual cell voltages but those are usually bluetooth modules and require a mobile phone.
 
Safety first. I'm scrapping the battery. I put in a balance BMS, and it's able to charge the 10S battery to 41.8V, However the pack now shuts off at 33.8V, with the one suspect cell group at 3.0V and the other nine at 3.44V. I rarely run my packs below 50%, so this is fine, but I don't like it.

A little voice tells me this is how vapers and ebikes catch on fire. A cell group goes below the 2.5V minimum. Dumb owner revives it.

I already got a replacement pack. This one will go into quarantine in my old BBQ and maybe I'll try to replace the bad cell bank as a winter project. The cells are in plastic forms, no glue, so it's possible.
 
Could look at the Speedict range- has a 10-15S and a 20S model that work really really well. Soft On off switch to prevent a connector spark, android phone app shows individual cell voltages, current and speed & temp monitoring. I have 2.

Posts on ES too, search for speedict.


markz said:
Yes, it'd be cool if someone came up with something, like a display you can install and look at live as you ride or something you just plug in after a ride to check all the strings out.
Well there was that one product from Hobbyking you could plug into the balance leads and see maybe 6S or 8S, so having 2 or 3 would be the alternative, but I'm just saying be cool to have a discrete unit on the handlebar to look at active voltages as you ride up a hill.

Maybe one day Justin from Grintech will develop a top quality BMS and sell his potting material. There's just to many different BMS' out there to know which ones are good, just have to go on what people say in the forums but people keep their cards close when they do find a good BMS seller.
 
As for OP, you could try unplugging the BMS balance plug and seek whether the voltages drop over time. Many take power from the #1 cell group so it could just be that.
 
docw009 said:
I have an unbalanced pack. Opened it up to find 9 groups at 4.07V, and one group at 2.20 volt.
You really don't know the condition of each cell in that 2.20 volt p-group until disassembling to test each individual cell in that p-group. It's possible some of the individual cells could still be usable, but unlikely.

Whether using a Cell Checker, Balance Charger or BattGO BG-8S they only show the mV of each p-group. In order to determine the condition of each individual cell in a p-group you to have to disssemble and test each individual cell to determine it's condition (good or bad).

https://www.manualshelf.com/manual/isdt/battgo-bg-8s-smart-battery-checker/user-manual-english.html ... It would take forever and likely never would balance your p-groups in their current condition with that 2.20v p-group whether using an iDST BG-8S, balance charger or a BMS. The p-groups are too out of whack to be properly balanced with that 2.20v p-group. Need to replace it or disassemble and test individual cells to see if any are still usable (good) in that 2.20v p-group and reassemble with good cells.

Have used two of the BG-8S as my 10S3P can be split into two 5S3P although not necessary with two 5S balance leads or pralleled as one 5S6P in which case you'd only need one BG-8S. In a manner of speaking it's a BMS more for either bottom or top balancing p-groups, but doesn't tell you the condition of each cell in each p-group such as unacceptable sefl-discharge more common with a energy dense cells rated at 15A-20A Maximum Continuous Discharge.
 
What kind of pack is it where is this parallel group located in the pack ? What cells are they how big is the pack and how many ampere-hour ?
How old ?
 
The pack is a run-of-the-mill 36V12.5Ah Silverfish, 10S-5P, came with a cheap ebike in April 2019. 5x10 cell array, using 2400maH cells, INR18650S.

IMG_1352.JPEG

One cell group, #7, became unbalanced. It originally read 3.6 at full charge, with a non-balancing BMS, I set the battery aside, and when I checked it later, it had dropped to 2.2V. COuld be self-discharge or bad BMS. I never checked.

When I put in a balance BMS, it would balance at 4.10V. Nonetheless, group 7 is still weak. During a full discharge, it decays to 3.0V when the other nine groups are at 3.4V. As a result, I get about 7-8Ah out of the pack, but it does deliver the 22A asked by my controller. I could be OK with that, but my main concern was that it decayed to 2.2V.

I've read over-discharge causes metal precipitates to form in the electrolyte. Most datasheets say you cannot go below 2.5V, so I have no need to take a chance.

I've got some NCR18650B's, 2500 maH cells, coming in soon. I know how to cut this old group out and replace.
 
I took out the suspect row of cells and replaced with five new NCR 18650B's, which were 2500mah like the original cells. It cost me $38 shipped for the cells from china, and $60 for a spot welder on amazon.

The battery uses a plastic form to hold the cells, so removal consisted of cutting pulling off the welded nickel strips to isolate the row, and then cutting the form.This was not as hard as it sounds. Snapped the new cells in, a little glue to hold the battery together. Re-weld the electrodes.

I doubt a repair service could make any money doing this. It's more of a DIY cost-be-damned project,

Used a drill to remove the plastic between cells. There were all at the same voltage, so I didn't mind if it cut thru the PVC. Better to have set the drill depth. Anyway, with the electrodes peeled off. it cracked apart easily.

PA020258.JPG

The new cell had short nickel tabs already attached, which I re-used, but they had to be folded, so it made the job a bit messier. I tested my spot weld settings on some scrap metal, and the tabs won't pull apart,

PA020268.JPG

One oversight. I didn't check the new cell voltages before assembly. Four were at 3.83 volt and one at 2.82V. Luckily, I caught it before I started spot welding and I had an extra cell to replace the bad one.

Everything checks out. The BMS works, but I will charge the new row manually tomorrow. Old cells will get pitched.
 
I had once a cell group of a battery (10s4p used on a eBike) down to 1.6v due to failure of the balance board, I already notice earlier that the group was out of balance (2.5v) and put it away and not thought about it (stupid). When I wanted to repair it, it was 1.6v. Anyway I brought it back to normal voltage and did a couple of charge/discharge cycle's and it holds up well, also temperature and under load, so I just leave it for now.

Once in a while, after 10-15 cycles I have to discharge that one cell group to get the same voltage again as the other groups, I still need to buy a new BMS ;)
 
Seems to me that BMS most useful function is to stop people from charging a battery that has one or more p groups that are to low.
This is ok for the majority of the population. Doe now fit well for members here. We need access to battery information to be proactive for the care of them. Only balancing when fully charged is not good enough as anyone that has min knowledge would not do this enough. How often it needs to be done depends on things we can not see.

I'm in a dilemma over it all. I'm sure some or more already have over come this issue. Its seems we have a few choices or I do. Can start building my own batteries, which would very few if they lasted. Could open up new batteries and add leads that would allow me to check on p groups and add active balancer to them.

Don't really want to start spot welding batteries. Ripping open a new battery is not something I want to do. Hoping for a serviceable battery. or just go to EV batteries and forget the Cans.
 
ZeroEm said:
Seems to me that BMS most useful function is to stop people from charging a battery that has one or more p groups that are to low.

I am now thinking that active balancing can hide a battery problem.

The above cells would only charge to 4.0V manually, and were weaker, so they would hit LVC at 3.0V when the others were at 3.4V. Without balancing, they recharge to 3.8V on the second charge and shut off early again. They top off at at 3.4V on the third charge and now the pack is dead. Probably best. With a balance BMS, the weak cells will start at 4.0 and still hit LVC earlier, but I would get about 60-70% of the pack's capability and be happy, However, I knew the cells had drained to 2V and been revived. I also knew they were now damaged, not being able to hit 4.2V. So took the pack out of service, and for a lark, I replaced the cells.

Balance is good to handle the 50-100mv variation you might see over a few months of use. It takes a while to do that. You probably want the BMS to err on the conservative side and take the battery down if it gets to 400 mv like mine.
 
Your acting like people should when the battery is dying. While 2v is bad and should cause worry, other's on here want to charge up from 0v. That just would not let me sleep at night. I don't want to hide a bad p-group just want to balance at 80% and not be forced to charge 100% and still don't know if they are balanced. I'm a 80%-40% type person. Do go higher and lower if my trips are that far. Not very often. Only do 100% now maybe once a year now.

by docw009 » Oct 04 2021 11:10am

ZeroEm wrote: ↑Oct 04 2021 7:55am
Seems to me that BMS most useful function is to stop people from charging a battery that has one or more p groups that are to low.
I am now thinking that active balancing can hide a battery problem.

The above cells would only charge to 4.0V manually, and were weaker, so they would hit LVC at 3.0V when the others were at 3.4V. Without balancing, they recharge to 3.8V on the second charge and shut off early again. They top off at at 3.4V on the third charge and now the pack is dead. Probably best. With a balance BMS, the weak cells will start at 4.0 and still hit LVC earlier, but I would get about 60-70% of the pack's capability and be happy, However, I knew the cells had drained to 2V and been revived. I also knew they were now damaged, not being able to hit 4.2V. So took the pack out of service, and for a lark, I replaced the cells.

Balance is good to handle the 50-100mv variation you might see over a few months of use. It takes a while to do that. You probably want the BMS to err on the conservative side and take the battery down if it gets to 400 mv like mine.
 
Back
Top