Do you match pack capacities?

Hi All

Bit of a newbie question Iguess...

When your making your packs do you all use therepackr siteor something similar to match pack capacities or is it not important. Strikes me that using something like this on 7s80p packs for examplewould get pretty time consuming so I'm just wondering how necessary it is?

Thanks for any answers.

Cheers

I don't use repacker or any tool. The reason is I am building 14s120p. The time it would take to catalog and sort 1680 cells (days...) is not worth it. I try to build 7 packs at once and spread the cells between them. This helps if you get a batch of unordinary cells like 3000mah when most of yours are 2200mah. Additionally, the capacities on the cells are not 100% accurate. You can test them 10 times and get 10 different capacities.

It's just not worth it with large packs like ours. I'm guessing it would be very helpful if you were building small packs of 5-10 maybe.

Im not using that either. I sort mine in 100mAh ranges and then i just random picks em. Thing is as i earlier wrote on Facebook:

1. Unless you use callibrated and good chargers you wont be able to get numbers worth sorting
2. If you use more than 1 tester you will get different numbers due to #1
3. if you use 2nd hand batts they will change capacity uneven and in 20 runs they differ alot anyways


For instance if i compare 2 of my "cheap" testers that do cost 30eur a piece...

same battery can diff from test to test easily 1-5%. 1 battery tested on several of them can easily diff from 22100->2300mAh. Thats whopping 10%!!

So if you take those 10% you can easily just sort them randomly for instance in 100mAh ranges and then pick them together. If you want to test this after its put together you can do this.


One example to test if they are even:

Top balance the pack to 4.2 all over.

Take the pack down to below 50% soc. Ie around 3.7v. And start meassuring to see what pack that drifts furthest away. That pack will most likely have to little capacity compared to the others and you just add/remove some cells to compensate. Its manual labour but this will in many cases be better to gain the last % that the cheap charger cant.

This is my 5 cents of it. Im personally going to use my BMS system to do the test above and will get reports every xx time on what packs that may have to little capacity compared to others

So mike you don't check capacity at all?

FireFrogSo mike you don't check capacity at all?

You should check the capacity and voltageto get and initial reading. Then wait a couple weeks to check again for any self-draining losses that might occur over time. If the cell drains, toss it out.

Im planning to check capacity once my tester comes in and im doing a week self discharge test after charging

FireFrog said:

So mike you don't check capacity at all?

Click to expand...

Yes, absolutely. I am using cells over 2000mah. I just throw them in a big box so they're nice and mixed up before building packs that way if I have a pile of outliers such as the NCR18650s that are 2900mah, they get spread out randomly across packs, not all thrown in one pack.

YES i made an ebike battery 13s 10p and they were all within 150mah of each other. at 21amph

Oneupthesleeve said:

YES i made an ebike battery 13s 10p and they were all within 150mah of each other. at 21amph

Click to expand...

I use repackr.com for my 24P packs. It helped balance my system so I'm a great fan of it. But -If I was doing larger than 48P I'd probably just pickrandom cells from a bucket of similar capacity cells.

My vid on pack sorting options:

it's like how they mix smarties... Big box and stirFantastic idea don't get me wrong just doesn't scale up

I don't have that many cells yet. About 160 now? Starting with 7s20p. Planning on settings up the packs so that the individual cells are as equal as possible, the best pack averaging at 2800mAh and the worst at 1500mAh. As I harvest more cells I hope this will allow me to hook up evenly crappy packs in parallell and replace one of them with a smaller and greater one. In the end I hope to end up with good capacity packs only.

Perhaps keep the crappy as low charged packs for backup or maybe even in some secondary system that kick in when needed. Haven't really thought about that one too much.

One thing occured to me .... if you know you have some cells with higher capacity , it makes sense to put them at the end of the pack that is nearest to the output terminal ... and put the weakest ones which will deliver less current at the back of the pack where the current has furthest to travel...in this way you will save ever so slightly on power loss in the busbar perhaps an increase in overall capacity of around 0.1% could be expected if using a 7.5 square mm busbar.

I mix my packs totally random in some sense. But i try to keep a certain match in same string
For instance string1 will have 2000-2500mAh. String 2 will have 2500+. String 3 might have 1800-1999....

Only reason i do above is so i know what i have. Other than that the only reason i would not mix is if i would have known how much left a cell actually had... I wouldnt mix new ones with old ones

Hope you got what I was rambling about.

daromer said:

I mix my packs totally random in some sense. But i try to keep a certain match in same string
For instance string1 will have 2000-2500mAh. String 2 will have 2500+. String 3 might have 1800-1999....

Only reason i do above is so i know what i have. Other than that the only reason i would not mix is if i would have known how much left a cell actually had... I wouldnt mix new ones with old ones

Hope you got what I was rambling about.

Click to expand...

Not sure I'm clear what a 'string' is ....I'm thinking about an 80p pack that's 4x20 arrangement ...If the 2,5's are put near the output terminals, and the 1.8's at the other end this will cut down slightly on resistive loses in the busbar ...with your arrangement 7.5mm2 buss and the range of capacities , from 2.5 to 1.8 not much saving , perhaps 0.05% I would guess ...

Daromer...I noticed in one of your videos you had a fuse blow and took out the cell ... do you still have this cell???? ...can it still holda charge???... what's isit's internal resistance ?... what current doesit draws when 4.2V is across it from an 80p pack... all these things will help understand what happens when a cellgoes bad ,...

String as in 14 pieces of 80p packs. Ie 14s80p is a string for me.
Im not going to go into placement of cells. Its to early here and i think that need to be simulated in some software to understand what would happen. Im mixing mine. I rather have 1 good cell and 1 cell that is not as good beside each other and spread all over. The resistance of a busbar is negligible if you ask me. In my case the busbars are so thick atleast
Note that mixing 1800 with 2500 doesnt matter. Because when the smaller one have less to give it will just drop a bit in voltage and the other cell will take over. But what you need to understand is that in the end of a cycle the high value cells need to take some more current load.

No i did not have a fuse blow I simulated it due to that i found 1 cell that had lower voltage than the rest. No issue with that cell. I must have forgotten to charge it fully and have pulled it out of the charger to early.