Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
14s50P capacity variances
#1
I am in the final stages of building my first Power Wall, its a 14s 50p. I have assembled the packs and have been capacity testing them with the icharger X6, and have picked up variances from 91Ah to 102Ah between the packs.

They have been assembled using re-packer, and according to the Litokala's they are all rated at about 107540mAh (107Ah)
I understand that the Lii-500's can be off by as much as 10%, but a pack as low as 91Ah and another as high as 102Ah is quite a variance?

Is it a potential issue? 
Should i be adding cells to them to balance them out ?
Should i be stripping them down and re-testing ?

Thanks any feedback is greatly appreciated.
Reply
#2
There's a few angles here.   

Practical side - once the battery is in operation and you charge/discharge over a few cycles - it may behave better than you think.  Your BMS should help you with this - e.g. do they easily stay in balance once they've settled in.    I think an 'initial' 10% difference is noteworthy but not alarming.    If it doesn't balance OK after you put it into operation - then it needs attention.  Or even before you put it online - you can just add (or disconnect) a few cells on the extreme packs - perfectly reasonable - to bring them closer.      

*For me, I leave extra slots on my base packs and have added a few cells to bring up the low ones after I see how they perform in about 20% of my packs.   I just put them online and use Batrium 'balance info' to see where I need to tweak things.   Most required a 3% adjustment but 1 of them (out of 80) required a 15% bump - not sure why Smile

On the more technical side I think you'll get better info here than I can provide - but there are some basic causes:
- maybe you have some self-discharging cells in the low packs - e.g. did you let the cells sit for a few weeks and check for this before building the packs?
- cells/packs will test differently on each charge/discharge in the same tester - to be expected.  Is 10% too much to expect - not sure but as I said above, its not out of relm of possibility.    You don't mention your pack test procedure - maybe that introduced some differences as well.
- mixed cells / mixed chemistries / mixed IR can cause differences in charge/discharge performances compared to other packs - even if the original test mah all add up the same.

My electrical engineer friend says he just doesn't like batteries because they are 'messy'..   and yes batteries can be messy, so its good to focus on the goal - e.g. do they stay balanced under charge/discharge operations and deliver the power you designed for Smile
Reply
#3
Repaclr is not magic... It doesnt compensate any errors. It can actually är error instead. Best is to use random based on a grouped value. Like 50mAh ranges. Also note that you egen having assembled packs you need to test with same current.

Lii do 0.5a so you need 25a current now.

Smaller packs är also more prone to this error. Can be seen IF you use a scattered diagram and document it.
The Ultimate DIY Solar and build place
YouTube / Forum system setup / My webpage  Diy Tech & Repairs

Current: 10kW Mpp Hybrid | 4kW PIP4048 | 2x PCM60x | 100kWh LiFePo4 | 20kWh 14s 18650 |  66*260W Poly | ABB S3 and S5 Trip breakers
Upcoming: 14S 18650~30kWh
Reply
#4
Several questions need to be asked.

  1. Did you check the IR of each cell? With a 4 wire kelvin 1kHz mΩ tester.
  2. What is the lowest and highest mAh of cells used. 
  3. What manufacturers are they.
  4. What is the SOH of each battery, i.e. 80% of capacity, lower or higher?
  5. Are the cells used mixed chemistry, i.e. ICR, INR, NCA, etc.
If you want to build a successful pack from used cells the IR needs to be within a window (I personally use 40mΩ to 55mΩ) and cells should be within a 400mAh difference of each other. As in 2250mAh to 2650mAh.

I personally use another repacker that was developed by Nemo and can be found on this board.
The results of following those rules create a very balanced pack.



I'm sure there will be a bunch of people that will pu pu this technique but the results are in the pudding. 
The procedure takes a bit more time but is well worth it.
So far the pack I have built is showing very good balance charged and discharged. The variation at max I have seen is 0.06V and that is mostly with moderate discharge or charging. Batrium very rarely needs to go into bypass at the top of the charge.



So my advice is check the IR of each cell, stick to an IR window and keep all cells within that range.
Stay within a mAh range, use daromers method of random picking, or use either nemos repacker or SLS repacker does not matter.

That should give you a relatively balanced pack.

Wolf
For Info Google Drive

Great ABB shunt breaker seller River City Industrial
Great prices and superb service.
 [Image: em2566%20(1)__19172.1521541365.jpg?c=2]

Not your average Wolf       
            Cool
Reply
#5
Thanks for all the insights. I see that i have a few issues that i did not take too seriously, that could come back to bite me in the A$$.

I did not measure internal resistance.
I have a mix of 18650 cells with a range of 1800 to 3000 mah, but did not take into account the original capacity, and therefore the state of health is not going to be 80%+ overall

All the cells are salvaged from laptop batteries exclusively, and they are all capacity tested and then self discharge tested at around 4 weeks.

Am i in trouble, or will i just have a system that will be higher maintenance ?
Reply
#6
My approach is a bit different.
Capacity testing, sorting, building packs according to Repackr takes way too much time. And after all that there still is some capacity spread, since there are so many variables that can affect the capacity tests.
I build my packs pretty much randomly, then during actual use I see which packs have too little/too much capacity. Then swap low/high capacity cells until the packs are balanced.
Modular PowerShelf using 3D printed packs.  60kWh and growing.
https://secondlifestorage.com/showthread.php?tid=6458
Reply
#7
Wow, so you use all cells untested ?

(06-07-2020, 11:28 AM)ajw22 Wrote: My approach is a bit different.
Capacity testing, sorting, building packs according to Repackr takes way too much time. And after all that there still is some capacity spread, since there are so many variables that can affect the capacity tests.
I build my packs pretty much randomly, then during actual use I see which packs have too little/too much capacity. Then swap low/high capacity cells until the packs are balanced.
Reply
#8
In my opinion you should test and sort properly. Also look 4 weeks after the test what the battery can still do or whether it has discharged itself (I sometimes have batteries here that after 2 months only have 0-3 volts that go into the garbage can)
Reply
#9
(06-07-2020, 07:28 AM)ShortcutShawn Wrote: Thanks for all the insights. I see that i have a few issues that i did not take too seriously, that could come back to bite me in the A$$.

I did not measure internal resistance.
I have a mix of 18650 cells with a range of 1800 to 3000 mah, but did not take into account the original capacity, and therefore the state of health is not going to be 80%+ overall

All the cells are salvaged from laptop batteries exclusively, and they are all capacity tested and then self discharge tested at around 4 weeks.

Am i in trouble, or will i just have a system that will be higher maintenance ?
I wouldn't get discouraged Smile    You have a range of 91 - 102.   You don't give the count - e.g. are most of them the average of 96.5?  or  7 of them 91 and 7 are 102?
5-10% of 50cells is 2.5 - 5 cells.   So worst case...    add 3 cells to the low ones and/or disconnect 3 cells from the high ones.

But really, if it works for your project,  just balance them out (hook in parallel for a week) and then see how it performs.   If your load is small and you don't have a large  DOD - it might just be fine.

>Am i in trouble, or will i just have a system that will be higher maintenance ?
I would guess 'worst case' is higher maintenance
Reply
#10
(06-07-2020, 01:00 PM)ShortcutShawn Wrote: Wow, so you use all cells untested ?

(06-07-2020, 11:28 AM)ajw22 Wrote: Capacity testing, sorting, building packs according to Repackr takes way too much time.

Badly worded.  I still test for capacity, but just to make sure it has at least ~75% of the original capacity.  Then I put them in a dated box and store them for 2 weeks ~ 3 months, depending on my schedule.
I then take cells from the oldest box, check voltages to make sure they've not self discharged, then put them straight into my 14 new packs in a round robin fashion (like a card dealer).  That tends to mostly spread cells from one source over all 14 packs.
With avg 2000mAh, 100 cells/pack, the capacities end up remarkably close at the end.  I don't keep statistics, but I'd say 10 packs don't need any adjusting, the other 4 need maybe +/- 3000mAh adjustments.

Also, the cells degrade at varying rates, unless you have fairly new perfectly matched cells from the same manufacturer.  What's the point of spending so much time on getting it perfect, when it's going to go out of balance after a few months~year of use anyway.
Modular PowerShelf using 3D printed packs.  60kWh and growing.
https://secondlifestorage.com/showthread.php?tid=6458
Reply


Forum Jump:


Users browsing this thread: 1 Guest(s)