Cell Matching Guidelines?

Z

zolac

New member
Joined
Aug 15, 2018
Messages
2
Hello everyone,

New to the forum, excited to share and see all the projects all you have going on.

I'm currently in a harvesting/test phases with my cells: ( using those modem packs with the 2200mah LGDAS1865 )


To test I've been using two Zanflare C4's ( little slow, but was the best local option without having to wait on china shipping times )

I've been building a database of my cells, tracking capacity, internal resistance, brand and date codes. With the intention to write a script that would match these cells together in the most effective way.

With that I've been trying to determine what are the MOST importantfactorsof matching these cells, from what I've gathered reading the forums briefly:
1. Capacity
2. Internal Resistance

I'm trying to go a step further with precision with andfactoring in:
3. Date code
4. Brand ( I have both LGDAS3 and LGAAS3 )

Would those last 2 characteristics mean anything to building a matched sets?

For capacity and internal resistance, which one takes priority?

Specifically with capacity, would a cell of 2380mah and a cell with 2460mah be considered an acceptablematch? ( internal resistance being similar or equal ) what is a the generally accepted threshold?

Thanks,


image_tiimhm.jpg
 
First of all those testers can not even come close to correct IR and IR can diff up to 100% easy per test! Therefore that number is totally irrellevant.
Capacity can diff up to 10% easy so in ranges of 100mAh is good enough.

I match all packs on some sets of criterias
* All packs with same amount of high capacity cells
* All packs with same amount of low capacity cells
* All packs with same amount of cells
* All packs with same amount of capacity
* All packs randomized for splitting IR and types of cells.

I try to keep same amount of each type of cell in each pack and so far running system with several 1000s of cells. Total of over 20 000... I have packs over 2.5 year in the powerwall still running and cells in other packs even older still running without bigger issues.

If you need high current and proper IR you also need a proper tester.

Check my videos regarding "arranging cells in packs" and internal resistance testing and you will get some input. I also have videos showing the values and how off they can be.
 
Thanks for the insight!

Upon trying it myself, I noticed completely different numbers for both Capacity and IR upon re-testing a batch of known cells with my two chargers.
(So yes i can see it about +/- 100mah for Capacity and varying for IR. )

I don't actually have any plans for my cell's yet, I'm sort of just testing them and trying to inform myself about the best way I could match them together for a future pack.

What I'm doing is more of an exercise in the process of recording their data and testing methodology.

So I now know that my chargers are great approximation, but not very precise.
What is some good relatively cheap ways I can measure things like Capacity and IR?
I'd like to get capacity precision down to something like 20-50 mah and IR to be a more consistent measurement.
What kind of tools are out there that could do this for a low cost? ( $100 )

All my cells are presumably new ( most often I have tear them from open from sealed plastic bag OEM packages. ) but have been sitting around for some time. So could date code and lot number be a better means of matching cells?

Often cells are group in parallel/series when harvested from these packs, meaning the manufacturer would have already grouped cell's with very similar characteristics? I haven't been keeping these cells together, for the future is it good idea to keep the cells from the same pack in their own "groups"?

I guess I've taken up a unreasonable amount of interest in sorting/grouping very optimally matched cell's, so I realize in most cases this is completely unnecessary.
 
iCharger Duo does it both. But it cost alot more. I also have a video showing IR testing. Check that one out.
You also need to do 4 wire measurement for it to even come close.
They just group them from same batch and nothing else.


Yes sorting in 100mAh is more than enough for most uses. If you have alot of cells in each pack it evens out in the end.
 
daromer said:
First of all those testers can not even come close to correct IR and IR can diff up to 100% easy per test! Therefore that number is totally irrellevant.

- snip -

Check my videos regarding "arranging cells in packs" and internal resistance testing and you will get some input. I also have videos showing the values and how off they can be.
Click to expand...

Daromer, i am in hunt of the IR thingi and am close to write something detailed about it, but the more i look into the site the more i find that many things already have been dealt with and talked about.

I looked for your videos but didn't find them.
Can you please provide a link for me ?
Would you second my idea of dealing (and writing) with it, or do you think its not necessary for the work done here ?
 
The more people sharing the experience and such as long as the data or information is done with good thinking then Im all for the more is better :p We are all here to learn.

Arrange cells:
Test of a tester including comparision to Opus and iCharger:
 
Thanks, Daromer.
I am going to collect some measurement before I will give some thoughts on it, especially on accuracy.

What i would like to ask again is, do you think RI could get a serious parameter taken in account when it could be measured reliably (and easy) ?
 
On low current applications where you are pretty far away of hitting any IR limit i doubt it makes sense to look into it. On high current applications its critical.

So lets say you use max 500mA of current per cell and your IR on the cell would easy supply 5A then you are 10x away from the "limit"

Yes this is not how you should state it but to make it easy for people around here i make this rather dumb example :p
So as you can see as long as you for instance test at 1A and the cell doesnt get hot and the test works out with good amount of capacity you also know that the IR is low enough for a system with max current 500mA as an example.

Thats my opinion and experience in the whole and can be applied to most people where you build a powerwall or battery bank where you aim to be able to sustain running on battery power for 2 hours or more in general. Thats because of the low C rating used.
 
daromer said:
On low current applications where you are pretty far away of hitting any IR limit i doubt it makes sense to look into it. On high current applications its critical.

So lets say you use max 500mA of current per cell and your IR on the cell would easy supply 5A then you are 10x away from the "limit"

Yes this is not how you should state it but to make it easy for people around here i make this rather dumb example :p
So as you can see as long as you for instance test at 1A and the cell doesnt get hot and the test works out with good amount of capacity you also know that the IR is low enough for a system with max current 500mA as an example.

Thats my opinion and experience in the whole and can be applied to most people where you build a powerwall or battery bank where you aim to be able to sustain running on battery power for 2 hours or more in general. Thats because of the low C rating used.
Click to expand...

Understood.
So i suggest there is only need for an accurate measurement, when you have a high -C-Rate discharge unlike the common Joe.

Or, there is no need because there is no simple method of accurate measurement.

Or both :D .
 
I do my test of IR during discharge and charge test. No its not giving an IR number but it shows me that it will work. And as said personally i dont test IR for powerwall but only when I intend to push a battery to its limits :D
 
You must log in or register to reply here.
Back
Top