A question about capacity variation

Dr. Dickie

Member
Joined
Sep 23, 2020
Messages
363
Hey guys, glad I found this forum.
I am in the process of testing the capacity of a bit over 1000 LGM26 cells. I had an Opus tester for about 6 months, and just recently got another (I have 2 MegaCellChargers coming this month--I CAN"T WAIT).
Anyway, when I had only the one Opus, most all cells tested somewhere in the 2680 or 2590 range for mAh capacity.
The new Opus, pretty much shows a higher capacity for cell--the last two slots almost always showing at 2700 to 2720.
Of course I don't know whether the first Opus is low or the second one is high.
What range would be considered to be not worth putting into rePackr to sort out for a 7S160P Powerwall I am making?
I have only done about 200 batteries, and I am thinking of putting things on hold until I get the MegaCell, and then just doing they all in them. If a difference of about 0.050 Ah capacity difference not worth worrying about?--about what I think the new Opus gives compared to the old one.
Thanks for any opinion
 
Opus are notorious for not being accurate. It has a various of about 5% off at times.

you can still use them, but you need to find out what their variations are and calculate the offset for those cells. If one of them reads 5% higher than the other, with the same exact cells, then you know there's an offset.

It doesn't matter if one is more accurate than the other as the numbers are a base line. When you get the Mega charger the offset will be even more. But you'll be able to process more cells.

Range is more a personal preference. Most users here won't use lower than 2000mAh cells. However, this is also based on what the original capacity of the cells were when new. If there were 2800mAh new, and are now 2000mAh, they have lost almost 30% SoH. But if it was 2200mAh new and now 2000mAh, that's only about 10% loss.

You can use as low of capacity as you want, as long as each parallel pack in the string are as close to the same as possible. Just make sure all cells have a similar SoH and you will minimize random failure of cells
 
I have 3 OPUSs. The first 2 were in sync and I created the first 1/4 of my battery bank. Then I got a 3rd one thatconsistently reported 50mah less than the others. In the 2nd 1/4 of my battery build, I religiously added 50mah to each cell tested by that opus.In the last year (last 1/2 of my battery work)I just go 'yawn' and ignore it. My packs are 88-120p (260ah) and at that level,it hasn't made any detectable difference from a battery bank operational standpoint from the early packs to the middle packs to the latest packs:)
 
Well, at a 120p, i can see why :p
 
Korishan said:
Opus are notorious for not being accurate. It has a various of about 5% off at times.

you can still use them, but you need to find out what their variations are and calculate the offset for those cells. If one of them reads 5% higher than the other, with the same exact cells, then you know there's an offset.

It doesn't matter if one is more accurate than the other as the numbers are a base line. When you get the Mega charger the offset will be even more. But you'll be able to process more cells.

Range is more a personal preference. Most users here won't use lower than 2000mAh cells. However, this is also based on what the original capacity of the cells were when new. If there were 2800mAh new, and are now 2000mAh, they have lost almost 30% SoH. But if it was 2200mAh new and now 2000mAh, that's only about 10% loss.

You can use as low of capacity as you want, as long as each parallel pack in the string are as close to the same as possible. Just make sure all cells have a similar SoH and you will minimize random failure of cells

Thanks, for replying I appreciate it.
These are all testing 100% or so (very few test below the 2600 mAh rating).
If I had used only the one Opus, then whether it was low or high, I wouldn't care as they were all measured the same way. With the new one consistently testing higher that will mess up my average capacity when making the packs.
I think I will plug along for now (if I only did this way it will be well into next year before I have them all tested), and then test some when I get the MegaCell.
If a 5% variation is irrelevant to packing for average capacity of the packs, then I won't sweat it.
Yeah, all of these batteries came from the same kind of medical pack, they are all LG M26, and I think similar age. Since they came from medical packs, I assume almost no cycles on them, as they should have been strictly for backup.
Thanks again, as this is my first Powerwall, I am feeling somethings out.
 
The LG M26 cells I am testing all come from ninebot 36v batteries. They test great. 2540-2750mAh, these cells are from tube batteries not the silicone rubber coated ones.
I think the silicone rubber coated ones were damaged when the power connector are cut off. I am having about 50% bad cells(cells 0v, cells that won't charge). The ones that finish testing are in the 2600-2720 mAh range.

I too am waiting for my MegaCellCharger. Friday will be a month since I got the megaCellCharger has shipped notice. Time to email Alex again.Grrr
Later floyd
 
Dr.Dickie said:
Hey guys, glad I found this forum.
I am in the process of testing the capacity of a bit over 1000 LGM26 cells. I had an Opus tester for about 6 months, and just recently got another (I have 2 MegaCellChargers coming this month--I CAN"T WAIT).......................... If a difference of about 0.050 Ah capacity difference not worth worrying about?--about what I think the new Opus gives compared to the old one.
Thanks for any opinion
@Dr.Dickie
You have what I have come to realize as accuracy fatigue.
Let me explain. We all are looking for the perfect mAh results that are consistent every time over and over again. That is not possible due to many reasons. In a perfect world you put a cell in a charger/tester and you get "2590mAh" results and you do it again and you get "2590mAh" results. Unfortunately this does not happen. One time we get 2655mAh then we get 2595mAh and we get on the 3rd try 2602mAhso we are always left to wonder what is the true mAh of the cell.
I have just recently tested 1600 LGM26 cells and if they are coming from the same source as I think they are you are going to be all right.
The only stipulation that i make on that is to make sure you check the IR (1kHz 4 wire AC) and if your cells are in the 35m? to 39m? range they will be good.
I used 10 chargers/testers in my analysys. 1 Vapecell, 1 SKYRC, 3 Opus, 1 XTAR, and 4 Foxnovo.

image_kootun.jpg


The results of theses chargers varied ?5%

image_mtzybt.jpg


And averaging out all the cells by tester resulted in this data.

image_shmhyl.jpg

It obviously shows that at least myOPUS chargers on average report higher than standard results.
So what to do? To be honest nothing. In my case after choosing the lowest IR and best SOH cells out of the 1600 I came up with ?1200 cells from all various chargers.
I ran them through an excel repacker that takes the cells and sorts them into packs by cell # and mAh results. The way this works it takes the highest values first and then distributes all these cells evenly into packs. In the final analysis 100mAhor even 200mAh won't make a big difference since the inaccuracy of the testers will be absorbed into the pack.
After running my 1200 cells for my 14s80p pack the required 1120 cells here extrapolated and I built the packs accorningly. Testing the 14 packs proved to be a success as the difference between the min and max packwas 2.11Ah.

image_qutniu.jpg

So Using 10 chargers of various makes and models in the final analysis didn't make that much of a difference in the whole outcome. Of course using only 1 brand of cell and all the same capacity makes a huge difference. My first pack which was a frankenstein pack had similar results but all cells were very carefully chosen.

image_ogldvd.jpg


I also have 2 Megacell chargers and Have done a test on them seehttps://secondlifestorage.com/showthread.php?tid=9855&pid=68372#pid68372
No need to post it twice.

Have fun

Wolf
 
floydR said:
The LG M26 cells I am testing all come from ninebot 36v batteries. They test great. 2540-2750mAh, these cells are from tube batteries not the silicone rubber coated ones.
I think the silicone rubber coated ones were damaged when the power connector are cut off. I am having about 50% bad cells(cells 0v, cells that won't charge). The ones that finish testing are in the 2600-2720 mAh range.

I too am waiting for my MegaCellCharger. Friday will be a month since I got the megaCellCharger has shipped notice. Time to email Alex again.Grrr
Later floyd
Mmm, mine are coming from the silicone rubber coated packs, and mine are all testing great (i have dug apart about 60+ of those packs so far). Have only tested about 200 batteries, but they were randomly taken from different packs. So far, no duds, I hope my luck holds.
A MONTH since shipping notice.
I bet that is because of the 1 Amp charging change they did, it probably made them hold them back to do the change.
Good luck


Wolf said:
Dr.Dickie said:
Hey guys, glad I found this forum.
I am in the process of testing the capacity of a bit over 1000 LGM26 cells. I had an Opus tester for about 6 months, and just recently got another (I have 2 MegaCellChargers coming this month--I CAN"T WAIT).......................... If a difference of about 0.050 Ah capacity difference not worth worrying about?--about what I think the new Opus gives compared to the old one.
Thanks for any opinion
@Dr.Dickie
You have what I have come to realize as accuracy fatigue.
Let me explain. We all are looking for the perfect mAh results that are consistent every time over and over again. That is not possible due to many reasons. In a perfect world you put a cell in a charger/tester and you get "2590mAh" results and you do it again and you get "2590mAh" results. Unfortunately this does not happen. One time we get 2655mAh then we get 2595mAh and we get on the 3rd try 2602mAhso we are always left to wonder what is the true mAh of the cell.
I have just recently tested 1600 LGM26 cells and if they are coming from the same source as I think they are you are going to be all right.
The only stipulation that i make on that is to make sure you check the IR (1kHz 4 wire AC) and if your cells are in the 35m? to 39m? range they will be good.
I used 10 chargers/testers in my analysys. 1 Vapecell, 1 SKYRC, 3 Opus, 1 XTAR, and 4 Foxnovo.

image_kootun.jpg


The results of theses chargers varied ?5%

image_mtzybt.jpg


And averaging out all the cells by tester resulted in this data.

image_shmhyl.jpg

It obviously shows that at least myOPUS chargers on average report higher than standard results.
So what to do? To be honest nothing. In my case after choosing the lowest IR and best SOH cells out of the 1600 I came up with ?1200 cells from all various chargers.
I ran them through an excel repacker that takes the cells and sorts them into packs by cell # and mAh results. The way this works it takes the highest values first and then distributes all these cells evenly into packs. In the final analysis 100mAhor even 200mAh won't make a big difference since the inaccuracy of the testers will be absorbed into the pack.
After running my 1200 cells for my 14s80p pack the required 1120 cells here extrapolated and I built the packs accorningly. Testing the 14 packs proved to be a success as the difference between the min and max packwas 2.11Ah.

image_qutniu.jpg

So Using 10 chargers of various makes and models in the final analysis didn't make that much of a difference in the whole outcome. Of course using only 1 brand of cell and all the same capacity makes a huge difference. My first pack which was a frankenstein pack had similar results but all cells were very carefully chosen.

image_ogldvd.jpg


I also have 2 Megacell chargers and Have done a test on them seehttps://secondlifestorage.com/showthread.php?tid=9855&pid=68372#pid68372
No need to post it twice.

Have fun

Wolf


No, I am not testing the same battery multiple times, BUT I just noticed that randomly pulling batteries from a box of about 800+, the second Opus shows consistently higher capacity. And the last two slots o that Opus show a very high capacity a very high percentage of the time.
Whether it is high, or the earlier one is low, I don't know. But since 99% of the variation is less than 0.100 Ah, I an not going to sweat it. It just seems that it could mess up the packs. BUT, I do have the same cells, so that should help.
Thanks
 
Dr.Dickie said:
No, I am not testing the same battery multiple times, BUT I just noticed that randomly pulling batteries from a box of about 800+, the second Opus shows consistently higher capacity. And the last two slots o that Opus show a very high capacity a very high percentage of the time.
Whether it is high, or the earlier one is low, I don't know. But since 99% of the variation is less than 0.100 Ah, I an not going to sweat it. It just seems that it could mess up the packs. BUT, I do have the same cells, so that should help.
Thanks
I was not implying that you were testing the same cell multiple times. If you took that as what I said I am sorry for the misunderstanding. I was just showing an example in that statement.

Yes if you look at the graph up above comparing the cumulative average of the testers mAh results you see I have an OPUS that has a slot that is consistently high. OPUS 1 slot 4. OPUS 1 slot 3 and OPUS 2 slot 4 are not far behind. Running ? 45 cells through each slot and averaging the results.
I can tell you that the SKYRC is pretty much dead on and so far the Megacell charger results have been veryencouraging.
As I said before if your deviation is only100mAh you will be just fine. I will look forward to seeing the results of your build.

Wolf
 
Wolf said:
Dr.Dickie said:
No, I am not testing the same battery multiple times, BUT I just noticed that randomly pulling batteries from a box of about 800+, the second Opus shows consistently higher capacity. And the last two slots o that Opus show a very high capacity a very high percentage of the time.
Whether it is high, or the earlier one is low, I don't know. But since 99% of the variation is less than 0.100 Ah, I an not going to sweat it. It just seems that it could mess up the packs. BUT, I do have the same cells, so that should help.
Thanks
I was not implying that you were testing the same cell multiple times. If you took that as what I said I am sorry for the misunderstanding. I was just showing an example in that statement.

Yes if you look at the graph up above comparing the cumulative average of the testers mAh results you see I have an OPUS that has a slot that is consistently high. OPUS 1 slot 4. OPUS 1 slot 3 and OPUS 2 slot 4 are not far behind. Running ? 45 cells through each slot and averaging the results.
I can tell you that the SKYRC is pretty much dead on and so far the Megacell charger results have been veryencouraging.
As I said before if your deviation is only100mAh you will be just fine. I will look forward to seeing the results of your build.

Wolf
Ahh sorry, I misunderstood. Yes, thank you very much for all the data.
That was the answer to my question, I just wanted to make sure that a 50 mAh difference would not mess up my pack.
I love hearing that so far the Megacell charger looks good.
Dang, I wish I had heard about the SKYRC before I got the OPUS, considering what they are asking for the Opus these days, that is not much of a cost differential.
Again, thanks so much for the info. I am sure I will likely have more questions down the road.
 
Back
Top