Mixed Cells. 2ah + 3Ah


MASSIVE restock of EV Batteries at www.batteryhookup.com 5% coupon "POWERWALLS"

Harry_001

New member
Joined
Jan 15, 2022
Messages
4
First post. Throwing out a puzzle, to me anyway,.

I built a 10S-5P pack out of old Panasonic 22P's. Originally rated 2200 mah,, Six years old. Tested around 2100 ah on a Liitokala 500S @ 500A. IR around .038 ohms on a YR1035. Spot welded. Series nickel was double layer of .15mm and .10mm, ,60 inch wide, more than enough for 5A per cell.

I only had 48 cells. I made 9 rows with the 22P's. The last one was three 22P's and two LG MH1's, also used but tested at 3100 mah and ,032 ohms IR.
While I didn't go into the details of a repackr selection, I arranged the nine rows to be at least 10.4ah each with the last one at 12AH, I expected to get 7-8AH out of the battery on an ebike,

Next, I ran the battery on my ebike. From past experience, I know 20 miles will take about 5Ah at 12 mph. At 19.3 miles, the battery voltage was down to 35.7, about half way down. Half of a nominal 10Ah is 5Ah. Looking OK,. Then the BMS shut down the battery. Uh oh. it's unbalanced.

I went home on my backup battery and then checked the balance voltages. The first nine averaged 3.61 volts, low of 3.59, high of 3,62, but the tenth row was only 3.3V. I took the tenth row cells out, and all of them were 3.31 volts, so no broken spot welds, I tested the three 22P's from row 10. Discharge capacity was 2040-2090 mah, so nothing different. The two MH1's were above 3100 mah.

I surmise that the tenth row supplied 5AH like the other nine before going flat,. It should had at least 3-4 AH left, Where did the power go?

My plan is to make the tenth row all MH1's, as that's all I have left,
 

OffGridInTheCity

Active member
Joined
Dec 15, 2018
Messages
2,073
Cells don't act exactly the same even if they have the same capacity. IR (heat), and load (some cells do better than others), and manufacturer/design specs all contribute to variance in operations.

If the 10th row are all 3.31v and all rows were balanced to start with, then you'll likely get more even results across the board if you distribute the 3.31v evenly thru the other rows (maybe every other row) and then take those 5 cells and make a new 10th row.
 

Korishan

Moderator
Joined
Jan 7, 2017
Messages
6,880
Every parallel group should be as close to the same overall capacity to each other as possible. And with this, the count of parallel cells should be the same for each parallel group. The capacity has to do with charge/discharge balance. The number of cells in each group has to do with the current balance capabilities.

So if you have 9 packs of cells and each cell is 2200mAh and in 5p arrangement, which would make each parallel group 11000mAh, then your 10th pack should be as close to 11000mAh as well.
Likewise, if you have 5p on the other 9 packs, then your 10th pack should also be 5p.

With that said, if the 9 packs have 5 cells with 2200mAh, and your 10th pack has 3 cells at 2200mAh, and 2 cells at 2000mAh, then your 10th pack is not 11000mAh, it is instead 10600mAh. This will mean that the 10th pack will become fully charged before the other 9 packs, and will also become discharged before the other packs. This can cause some serious misbalancing if left unchecked. Not to mention the amount of power that is wasted trying to maintain balance (even with active balancers)
 

Harry_001

New member
Joined
Jan 15, 2022
Messages
4
Thanks for the replies!

Row 10 was three 2050mah cells and two 3100 mah cells, +/- 50mah per cell, so it really was 12ah. I figured my balance BMS could take care of it but the surprise was row 10 being the lowest voltage group after one run, not the highest voltage.

My thoughts about making row 10 five 3100 mah cells was a bad idea. It will stress my BMS. I cannibalized another old battery , got more 22P cells and made a new Row 10.

I'll have to think about why row 10 didn't work out. Mostly riding around in level 1 pedal assist with a 20A oontroller, drawing 100-120W. Mild mannered stuff.
 

italianuser

Member
Joined
Feb 25, 2020
Messages
502
What about putting an MH1 cell in each of the ten series (9x 22P and 1x MH1). That way you're sure series have the same capacity and they should charge/discharge in the same way.
 

Harry_001

New member
Joined
Jan 15, 2022
Messages
4
What about putting an MH1 cell in each of the ten series (9x 22P and 1x MH1). That way you're sure series have the same capacity and they should charge/discharge in the same way.
I don't think the problem was one row being 12AH and the others 10AH. It's all in series. They all pass the same current/

The problem was two 3000mAH cells and three 2000mAH cells in parallel delivering the capacity of a 6000mAH. I know all five were connected. They started at 4.2V and wound up at 3.3V.

Maybe my final B+ tap was poor and the spot welds concentrated the current thru the MH1' cells forcing them to do more of the work. Then the remaining cells did their work equaliizing the voltage when the load got lighter.If I had to rate them, I bet the MH1 cells got cleaner welds.
 

Wolf

Active member
Joined
Sep 25, 2018
Messages
1,715

italianuser

Member
Joined
Feb 25, 2020
Messages
502
I don't think the problem was one row being 12AH and the others 10AH. It's all in series. They all pass the same current/
From the practical experience here in the forum I'd bet it is! Having same capacity in all packs is one of the basic rules.
 
Top