Cells Behavior

[Honuz]

Hi all,
I just builded a pack of 16 cells and graph their voltage near their 100% SOC : (will not say anything on what i did or did not do)



Between 13:00 and 16:00 they are under charge, the interesting part is at 13:45, cells voltage .. diverge, even if under charge, some lose voltage while others gain voltage.. isn't that strange...? During this interval, charge voltage do not change, always 54.2V - 54.3V.
Balancing is 1A.

 

[Nobodybusiness]

Hi all,
I just builded a pack of 16 cells and graph their voltage nearl their 100% SOC : (will not say anything on what i did or did not do)

View attachment 126303

Between 13:00 and 16:00 they are under charge, the interesting part is at 13:45, cells voltage .. diverge, even if under charge, some lose voltage while others gain voltage.. isn't that strange...? During this interval, charge voltage do not change, always 54.2V - 54.3V.
Balancing is 1A.

Take a smarter person than I to explain that..

 

[Honuz]

Take a smarter person than I to explain that..

Common .. do not sell you short like that .. !

 

[Honuz]

I give the 2 days graph, to see that this only happend on top of the charge
The total pack voltage is not on the same scale .. green bolg line is the total voltage (right scale), all 16 cells got their scale on the left.

 

[Nobodybusiness]

Common .. do not sell you short like that .. !

There are a few EEs and Other assorted Engineers on here that will explain it to the molecular level.
I’m just a layman.

 

[brum]

Can you add a chart with the current going in or out of the battery during that time? What is the BMS? Does it have active balancer?

 

[Samsonite801]

Have you tried another test and make a chart with balancing disabled to see how it may differ?

Or maybe the cells are just a bit out of balance and it doesn't manifest until it arrives close to the knee?

Have you tried to set balance on charge and discharge, or is it only set to balance on charge?

Also, did you capacity test each cell independently to see if they are all close?

 

[Honuz]

I added Power and SOC (as stated by JkBMS => Not accurate)
I use a JKBMS 1A balancing active.
Balancing is always active while over 3.3V and not within the 0.010V delta.
I think the same as Samsonit do : Out of balance and they got this "attitude", do not know if we can call that out of balance... cause every pack will deviate at a certain point .. but what's trange here is that some cells see their voltage decrease .... certainly cause other cells see their voltage increase and ... well .. i do not undertand, maybe it have to do with their internal resistance... dunno.



Maybe a reason of this beavior, they have not been top balanced. But still .. at this point i can't see why not top balancing them would generate this behavior ... i can't explain it logically.
Between 13:30 and 15:00, the average power is something like 350W => 525Wh => 9Ah ...?
So i think i will just stop charging this one just before the deviation => 3.39V/cell => 54.24V => 54.20V.

 

[MisterSandals]

Can you post a pic of your battery wiring?

 

[Honuz]

Can you post a pic of your battery wiring?

If you want to see how they are connected .. here it is
If you asked the general drawing of the system, i do not have one. BMS is connected as the documentation say to do it. I Used 2x18650 in serie to start it up ... and work at the first attemps. I do not think there is a wiring problem but i may be wrong..

From the battery negative pole => Shunt => BMS B- => BMS P- => Inverter -
From the battery positice pole => Knife Fuse => Inverter +

 

[MisterSandals]

I do not think there is a wiring problem but i may be wrong..

Looks great.

Is there a pattern to which cells climb and which drop? What cell position is the one that rises the most and what cell is lowest?

Just grasping for clues at this point.

 

[Honuz]

Looks great.

Is there a pattern to which cells climb and which drop? What cell position is the one that rises the most and what cell is lowest?

Just grasping for clues at this point.

At 15h30, starting from the top and going down (higher voltage cell to lowest one) : 15, 4, 3, 2, 10, 16, 9, 6, 1, 8, 13, 5, 14, 11, 12

 

[MisterSandals]

looks pretty random to me.

Did you clean and scour the contact points between bus bars and cell tops?

If you get them apart, it would probably help to do a parallel top balance.

 

[Honuz]

Tomorow, if there is enough sun, i will disable balancing and used my bench charger to make them all 3.5V and see what happend

 

[MisterSandals]

to make them all 3.5V and see what happend

If you can hold them at that voltage for an hour or until charge current drops to under an amp, they will absorb to 100% full. That will be the best chance you give your cells.