diy solar

diy solar

Top Balance - JKBMS - drift voltage between pair and odd cells

Set the actual charging voltage 100 mv lower than the battery voltage noted when the cell went over volt. Place in service. Check back in a week to see if battery voltage can be increased to the expected charging voltage. Repeat weekly until the voltage can be increased to the correct setting.
 
I was just following the "simple" process suggested by the community.
Cell in series, charge with BMS til 3.5V per cell, in my case the charger was set to 8*3.5V.
Switch cells to parallel, charge to 3.65V.

Are you suggesting, with this voltage difference between cells, to put them in parallel ?
Your resting voltage seemed not that bad. By connecting them all in parallel you then take out any anomaly of the BMS trying to balance them out.

Also feel free to then top balance at 3.55v let it bake and note when current falls off.
 
Hello, thanks again for all your reply's.

This morning this was the cell voltages:
1651481224123.png

As soon I can I will connect them in parallel for the final balancing, and then in parallel, charge till 3.6V.
 
Last edited:
3.38 and 3.66 is pretty big delta. It appears balancing current is still high. I’d almost let it sit for the morning and let the BMS balance without the bench supply connected until these values get closer.
 
I just watched Andy from off-grid garage discuss this issue and how some BMS's by default only activate balancing during charging current - which is not helpful for you from the best I can determine.

I'm not sure if posting links to other other youtube channel is kosher here, BUT if you open youtube and search for this
"off grid garage charge balance", the video in question will be the top result.

I think as others have suggested, either ....

pull the pack apart and top-balance in parallel

or

set your BMS to allow balancing when NOT charging.

So charge until your 'runners' hit the high state of charge and then disconnect the power supply and let your bms shuffle the charge between the low SOC cells and the high cells. Once your pack shows a low cell voltage differential, reconnect the power supply. rinse / repeat until you get all the cells to a much closer soc.
 
I just watched Andy from off-grid garage discuss this issue and how some BMS's by default only activate balancing during charging current - which is not helpful for you from the best I can determine.

I'm not sure if posting links to other other youtube channel is kosher here, BUT if you open youtube and search for this
"off grid garage charge balance", the video in question will be the top result.

I think as others have suggested, either ....

pull the pack apart and top-balance in parallel

or

set your BMS to allow balancing when NOT charging.

So charge until your 'runners' hit the high state of charge and then disconnect the power supply and let your bms shuffle the charge between the low SOC cells and the high cells. Once your pack shows a low cell voltage differential, reconnect the power supply. rinse / repeat until you get all the cells to a much closer soc.
Hello @MorganCarey

Thanks for your reply and for pointing me the video from Andy.
At this moment the power supply was disconnected and the BMS is balancing the cells.

What could be more efficient/quick, leave the BMS balancing the cells or connect them in parallel?

The state of the cell now
1651498103642.png
 
Hello @MorganCarey

Thanks for your reply and for pointing me the video from Andy.
At this moment the power supply was disconnected and the BMS is balancing the cells.

What could be more efficient/quick, leave the BMS balancing the cells or connect them in parallel?
Quick and efficient are different but related things.
The BMS will eventually balance your pack.
Since you have 2 low cells and the other 6 are close together you could speed this up by apply charge directly to the low cells.
You don't even have to break the pack to do it.

Just set the psu to 3.65 volts and watch your polarity.
 
Yes and charging to 28.00 volts would be fine to put in service and let the balancer complete the process over time.
 
You don't even have to break the pack to do it.

Just set the psu to 3.65 volts and watch your polarity.

Thanks for stating that Joey

That is something I've always wondered, but I've been too nervous to try.

So you can disconnect the power - supply from the bms negative and the pack positive. Move the power supply to the specific cells that are low and target those cells without isolating them from the rest of the pack..

So in theory this way you can leave the bms fully connected and use it to monitor the cell voltages to a better accuracy compared to some people standard off the shelf multi-meters.
 
Thanks for stating that Joey

That is something I've always wondered, but I've been too nervous to try.

So you can disconnect the power - supply from the bms negative and the pack positive. Move the power supply to the specific cells that are low and target those cells without isolating them from the rest of the pack..

So in theory this way you can leave the bms fully connected and use it to monitor the cell voltages to a better accuracy compared to some people standard off the shelf multi-meters.
I think every diyer should have a dvom that they trust as their reference.
But if the BMS and the dvom agree then its an easy way to monitor the cell voltages.
 
Thanks for stating that Joey

That is something I've always wondered, but I've been too nervous to try.

So you can disconnect the power - supply from the bms negative and the pack positive. Move the power supply to the specific cells that are low and target those cells without isolating them from the rest of the pack..

So in theory this way you can leave the bms fully connected and use it to monitor the cell voltages to a better accuracy compared to some people standard off the shelf multi-meters.

Hello guys,

I followed the @smoothJoey advice to charge the lowest cells individually without disassemble the whole pack, and after 3h I already got one at 3.65V. Works brilliantly! ;) (don't forget to turn off the active balance of the BMS during this procedure)

1651524152863.png

Edit: I'm charging one at a time, cell 2 will be next.
 
Last edited:
Hello guys,

I followed the @smoothJoey advice to charge the lowest cells individually without disassemble the whole pack, and after 3h I already got one at 3.65V. Works brilliantly! ;) (don't forget to turn off the active balance of the BMS during this procedure)

View attachment 93313
What happened to the rest of the cells?
Now you have a low cell and a high cell.
 
What happened to the rest of the cells?
Now you have a low cell and a high cell.
???

I just follow your advice:

Since you have 2 low cells and the other 6 are close together you could speed this up by apply charge directly to the low cells.
You don't even have to break the pack to do it.

Just set the psu to 3.65 volts and watch your polarity.


The 3.65V cell will not settle down to +/- the other cells voltages (3.4...V) ?
 
I'm not sure if posting links to other other youtube channel is kosher here

Perfectly fine to do so.

@Andre Magro Just be patient. Let the balancer do it's thing, charge the pack a bit, disconnect, let it balance. It will get there. Remember that these are large cells and even a 2A balancer will need time if the cells are way off.
 
???

I just follow your advice:

Since you have 2 low cells and the other 6 are close together you could speed this up by apply charge directly to the low cells.
You don't even have to break the pack to do it.

Just set the psu to 3.65 volts and watch your polarity.


The 3.65V cell will not settle down to +/- the other cells voltages (3.4...V) ?
The other cells have dropped in voltage significantly.
Now #2 is now the lowest.
I would have expected those cells that were at 3.5x volts to settle no lower 3.45 volts.
 
The other cells have dropped in voltage significantly.
Now #2 is now the lowest.
I would have expected those cells that were at 3.5x volts to settle no lower 3.45 volts.
#2 was second lowest and settled like the others about 100 mv.

Boost that #2 and put the battery in service. Should take even less time.
 
Hello,

I did not charge cell 2# individually, but let them all during the night in balance mode.
This morning I gave them another charge of 28.8V (3.6V per cell).

The picture on the right was taken 3h after the charge ended.
1651586143181.png

This morning charge graph.
1651586376062.png



Should I let them settle and give another 28.8V charge and balance, or should I switch to parallel for the final 3.65V charge?
Tks.
 
Hello,

I did not charge cell 2# individually, but let them all during the night in balance mode.
This morning I gave them another charge of 28.8V (3.6V per cell).

The picture on the right was taken 3h after the charge ended.
View attachment 93388

This morning charge graph.
View attachment 93389



Should I let them settle and give another 28.8V charge and balance, or should I switch to parallel for the final 3.65V charge?
Tks.
What was your charge voltage?
I would say put the pack into service and see if the BMS can keep the cells top balanced.
 
Cell 2 looks like it is going to have the highest rate of self discharge.
 
Back
Top