Series connection Charges/Discharges quicker

[multifrag]

I have a 13s9p battery. One of the series connection get's easily out of balance. Either when charging it reaches 4.2V quicker and goes beyond, or when discharging it goes into LVC and stops the battery from discharging fully. My question is, could this be just the nickel strip having loose connection(broken stop welds) with couple of cells in the parallel array or it's possible that the whole 9 cells degraded?

 

[amberwolf]

Could be anything that reduces the capacity of the group, so either of those is possible. Is also possible that the channel on the BMS has a balancing shunt stuck "on" so it's always draining that group. This can be tested by measuring voltage across teh shunt resistors. Any resistor with a voltae on it when the system is not fully charged or in the process of charging is stuck on. (typically easier to replace the BMS than fix it).

To test for poor connections to cells, test it while it's discharged to empty by measuring the voltage of each cell in the suspect group, without touching the interconnect strips, just the cell cans and tops.

Then retest it after it's charged to full, the same way.

If any of the cells come out different voltages than the rest in either test, there's a connection problem.

If none of them come out different, there's a cell problem (one or more of the cells is lower capacity than it should be).

If the latter is true, then you can find out which cell(s) by dividing the group by half to find out which half it's in. For this, you have to cut the interconnect between that group and the next one (on either positive or negative side, doesn't matter which), and then cut the group in half (on the same side you just cut). Then measure the voltage of each half. If they are the same now, let them sit a few hours to a day. If the group is draining into a defective cell, that section will be lower than the other (they should remain identical).

The same process is repeated on the section that's lower until you find the cell or cells taht are the problem.

If the voltages all drop, then there's more than one cell with a problem.


Sometimes the cells don't leak internally, but are still low capacity; for that you divide the group, then test each half with the same load for the same time (preferably using a wattmeter to monitor, but it has to be a wattmeter that has external power source as it won't likely run on just one cell group) and check the voltage. If they're different, then one half has lower capacity than the other. If the number of cells in the group is not divisible by two, then you have to calculate the capacity of each section and check the SoC chart for that cell type, and see what voltage each should be at for the amount of capacity you've drained.

 

[multifrag]

Could be anything that reduces the capacity of the group, so either of those is possible. Is also possible that the channel on the BMS has a balancing shunt stuck "on" so it's always draining that group. This can be tested by measuring voltage across teh shunt resistors. Any resistor with a voltae on it when the system is not fully charged or in the process of charging is stuck on. (typically easier to replace the BMS than fix it).

To test for poor connections to cells, test it while it's discharged to empty by measuring the voltage of each cell in the suspect group, without touching the interconnect strips, just the cell cans and tops.

Then retest it after it's charged to full, the same way.

If any of the cells come out different voltages than the rest in either test, there's a connection problem.

If none of them come out different, there's a cell problem (one or more of the cells is lower capacity than it should be).

If the latter is true, then you can find out which cell(s) by dividing the group by half to find out which half it's in. For this, you have to cut the interconnect between that group and the next one (on either positive or negative side, doesn't matter which), and then cut the group in half (on the same side you just cut). Then measure the voltage of each half. If they are the same now, let them sit a few hours to a day. If the group is draining into a defective cell, that section will be lower than the other (they should remain identical).

The same process is repeated on the section that's lower until you find the cell or cells taht are the problem.

If the voltages all drop, then there's more than one cell with a problem.


Sometimes the cells don't leak internally, but are still low capacity; for that you divide the group, then test each half with the same load for the same time (preferably using a wattmeter to monitor, but it has to be a wattmeter that has external power source as it won't likely run on just one cell group) and check the voltage. If they're different, then one half has lower capacity than the other. If the number of cells in the group is not divisible by two, then you have to calculate the capacity of each section and check the SoC chart for that cell type, and see what voltage each should be at for the amount of capacity you've drained.

Thank you for the detailed instructions. I have some nickel strip left over, so will be able to spot weld it back together after operation. DC load will be used if I can't find the culprit easily.

One more question. Let's say I find that one or couple of cells are bad. Replace them with new cells. Wouldn't that create a reverse problem, where while charging it's the lowest voltage group and discharging highest?

 

[amberwolf]

If the other cells in the pack are reasonably new condition, then it won't be that big a difference on the first cycle, and it won't be too much of a deal even later, compared to the ones you're replacing, if they're as bad as it sounds.

Hopefully it's just a connection issue, though, and you won't have to worry about it.