2P8S going to 3P8S

[gregoz]

I currently have a LiFePo system with 16 cells in 2P8S configuration. I would like to get another 8 cells to make it 3P8S using one SBMS0. Is this reasonable or should I be thinking of having two battery banks and two SBMS0?

[Paul Barrette]

Not an issue at all… just add the new batteries to each P and let them balance themselves to the same voltage… then connect them in series and balance the cels using the BMS (will take time if the voltages off

[Jim Fowler]

Not true! You do not want to mix new and old batteries! If they are not matched properly you will have some leeching off others and ruin your pack.

[Dacian Todea]

I assume the cells will be the exact same type if so then it will be no problem to add another set of 8 just connect them as Paul mentioned one of the new cells connected in parallel with the bgroup of two old cells that way the groups of cells will have the same capacity.

[Jim Fowler]

Won't this cause issues if they are different internal resistance? Or capacity loss due to the reduced capacity from cycling of the older batteries?

[Dacian Todea]

Jim,

With LiFePO4 the internal resistance will not change/degrade with age (it will with almost all other type of Lithium cells) but even if internal resistance is different from the new ones and also the capacity since you add two old cells in parallel with a new cell in each of the 8 series connected groups the equivalent internal resistance on all of those 8 groups will be the same as all of them have the same type of cells two older ones and one new.

And say this was a different chemistry and there you can have a large delta in internal resistance for new and old cells then the only problem will be that the new cells will have shared more of the total current when charging and discharging but the 8 series groups will still be equal in all respects of battery will stay balanced.

What will be bad will be to have 4 new and 4 old cells and build an 8s battery as then the old cells will always be the first to fully charge and fully discharge so capacity of the battery will be limited to the old cells capacity but even then it will work even if not ideal.

[gregoz]

Thanks for all the comments. Just to clarify my batteries are only a couple of weeks old (Xuba 280A), and I will get the same, so by the time I get them it will be only a couple of months age between them.

[Paul Barrette]

I can understand the previous comment about age and it makes sense. If you have 2 cells that have 100AH of capacity and you introduce a new cell that has 150 AH. the 150 AH cell will constantly be 'charging' the other two cells.

I have a 128P configuration and my small LiFEPO4 cells vary from 1.4 to 2.0 AH... not an issue even though my higher cells are 50% larger than the smallest ones.

Good luck

Paul

[Casey & Gina]

I currently have 2P8S as well, with thoughts of going to 3P or 4P if needed in the future. Since I bought my cells, I have found better prices on other brand cells which promise more charge cycles (not sure whether to believe it - 4000 cycles versus 2000, same technology?). I also reckon it might save some space to use aluminum-shell cells while the ones I have are plastic.

I was thinking it wouldn’t really matter which cells I added, as long as the addition to each parallel string was identical, but maybe that wouldn’t work out that well.

I’m curious how you manage so many small cells! Without individual monitoring of each cell, is there any concern that a bad cell would cause a problem for the whole string?

[Dacian Todea]

Paul,

That i not true the 150Ah cell will not be charging the other two 100Ah cells (tho a 100Ah cell that initially had 150Ah can be considered at the end of life and no longer used).

Your 1.4Ah if they are the same model as the 2Ah ones where probably abused else could not get to just 1.4Ah

Casey & Gina,

My LiFePO4 has a spec of 5400 cycles and it is real so yes you can have 2000 cycles and 4000 cycles battery with same LiFePO4 chemistry it depends on how the cells were designed and manufactured.

But you can not parallel cells that have different dimensions as you need to compress them and it is not possible to combine different looking cells.

all his cells are in parallel 128 in his case and they will be no different from a single higher capacity cell in any way and all cells will have no choice other than to be at the exact same voltage.