best order for cells with different voltage/ internal resistance

here is the thing...

I've 392 cells, all between 3,53v and 3,63v sorted by voltage.
Now I'll place them in cell holders 14 groups serial of 28 cells in parallel.
Is it without any further tests absolutely irrelevant how I place them (according to voltage) or something to consider ?

Edit: If it isn't any disadvantage I would just because it feels right, or just for fun begin to spread all the 3,53v cells evenly to the 14 groups, then I'll spread all the 3,54 evenly to the groups and so on until all cells are used.

I just want to make sure that if it doesn't matter later for the serial spotweld if I've started with the lowest voltage, the highest voltage ?

Sparfuchs said:

So should I spread all the cells I've sorted absolutely randomly all over the pack or is there something I could do to increase the chances of getting good balance ?

Click to expand...

Balance them before you assemble. Then they'll be identical voltage and it won't matter what order they're in. I'm not the first to suggest this either, it's good practice.

From-A-To-B said:

IMO, when you build, cells shouldn’t be paralleled if they have more than 0.1V difference. So balance, then build.

Click to expand...

If they're all identical voltage when you build the pack. And then the pack falls out of balance after normal use. Then you had bad cells.

Besides. You don't know whether they actually dropped voltage over 8 months:

Sparfuchs said:

So IF! this are all brand new cells, that were all about the same Voltage when I bought them (didn't check back then..just assuming

Click to expand...

If you didn't check, then you don't know for sure that they self-discharged.

You said they all have good IR, and you're not going to bother checking capacity. So, balance all 392 cells that you've picked out to same voltage within .1v of each other, and order them however you want.

harrisonpatm said:

Balance them before you assemble. Then they'll be identical voltage and it won't matter what order they're in. I'm not the first to suggest this either, it's good practice.

If they're all identical voltage when you build the pack. And then the pack falls out of balance after normal use. Then you had bad cells.

Besides. You don't know whether they actually dropped voltage over 8 months:

If you didn't check, then you don't know for sure that they self-discharged.

You said they all have good IR, and you're not going to bother checking capacity. So, balance all 392 cells that you've picked out to same voltage within .1v of each other, and order them however you want.

Click to expand...

hmmm, as far as I know max. 0,1 v difference is ok ? That's why I charged all the cells below 3,53v and checked every 3 minutes so they don't get over about 3,6v.
So I thought, they are now balanced enough
But even if I'd have a charger that can be set to a max. voltage of 3,6V I would not do it. If I'd have the time or equipment to balance my 400 cells to less than 0,1v difference I would go the step further to do the capacity test
So again, I'll use them as they are. If I don't have to consider anything I'll just start to spread the cells starting from lowest voltage

Seems like you are on track. If you aren’t going to measure capacity, then you’ve done all you can do and the rest is just crapshoot guessing.

Be sure to use cell holders and paper rings on the positive terminals.

harrisonpatm said:

You said they all have good IR, and you're not going to bother checking capacity. So, balance all 392 cells that you've picked out to same voltage within .1v of each other, and order them however you want.

Click to expand...

Then the weakest cells might not be evenly distributed.

balancing all the cells doesn’t seem important as they’re going to be put in parallel and balanced there and the bms could handle it and they’re all pretty close.


The goal being weed out the weakest and distribute the semi-weak

Sparfuchs said:

If I'd have the time or equipment to balance my 400 cells to less than 0,1v difference I would

Click to expand...

Could just throw them all in parallel. The voltage difference between cells is small enough or if not u could make all the parallel connections with a high resistance somehow. Maybe one super thin wire of steel on each side to connect them all

Hummina Shadeeba said:

Then the weakest cells might not be evenly distributed.

Click to expand...

I wholeheartedly agree, but OP is choosing not to do capacity testing, so there won't be any way of knowing which are the weakest cells, in order to distribute them. So it doesn't really matter at that point how they're distributed.

There seems a couple options for qualifying as what are “weak” cells:

High Self discharge
Low Capacity
High Internal resistance


How much cross-over is there between these three?
What is the best revealer of “weak”?

Hummina Shadeeba said:

What is the best revealer of “weak”?

Click to expand...

IMO, capacity, then self discharge, then IR.

Sparfuchs said:

I've 392 cells, all between 3,53v and 3,63v sorted by voltage.
Now I'll place them in cell holders 14 groups serial of 28 cells in parallel.
Is it without any further tests absolutely irrelevant how I place them (according to voltage) or something to consider ?

Click to expand...

Even with 28 cells in parallel i'd still take the time to balance all the cells within 0.050V (or even 25mV) of each other before assembling for spotwelding. Having 28 cells in parallel is like extra insurance for getting the most c/d cycles.


What's most important after eliminating 8 possible outliers and cell balancing (say at 3.50V) before assembly is ensuring the best spotwelding possible.