Battery charging while remaining capacity doesn't increase

[marcello]

Hi,
Sorry for the confusing title, but in short I have a 200Ah lifepo4 battery, as you can see on the pic it is charging fine via my solar panels and dcdc, but the JBD bms shows the remaining capacity doesn't increase anymore.
It shows 100% but the remaining capacity should increase as the batt is still charging at a consistent 20-25A. Or the bms should cut the charging if the battery is full.
I attach a pic of my bms settings too.
Thanks for your help.

 

[DIYrich]

Nominal capacity 200.81a

 

[GXMnow]

The battery state of charge reported by a BMS or battery monitor is only an estimate. At 199.81 amp hours on a 200 amp hour battery, it is clearly within 1% of full, but the cells may still be accepting some more charge. And it is very possible that amp hour count has drifted off as well. My JK-BMS has drifted the other direction. When my cells are near full, the BMS is only reporting 40% state of charge now. Each BMS or battery monitor has it's own rules for what must happen to trigger a reset to full 100% charged. In most cases, it is a combination of the voltage reaching a full threshold while the charge current also falls below a certain level for a given amount of time. Once it does a reset to 100%, it literally counts the amp hours removed during discharge to estimate how much is left. In a perfect world, if you discharged 40 amp hours out of a 100 amp hour battery, it would be at 60% state of charge. But what if the battery capacity is really just 98 or 102 amps hours? Or, when it the current sensor is a little off calibration? And what about some resistance loss that is losing power while charging so if thinks more power went into the battery but some of it was lost as heat? Each cycles that does not trigger a 100% reset can cause the state of charge to drift away from perfect. That is what happened on my JK BMS. After about 20 cycles, the SoC drifted down about 20%. But I know from the cell voltages that it is hitting 90% each day from sunlight. But for some reason, that is not causing the BMS to reset the SoC counter. After several months, mine got to where it was hitting zero SoC each night, even though the cells are really just dropping to about 50%, again, confirmed by the resting cell voltage and the fact they are hitting the absorb voltage after the correct amount of charging. So now each day the MPPT charger goes into float, the BMS only shows about 40% SoC. That sort of makes sense as I am trying to run the cells from 50% to 90% on daily cycles. That is a 40% change, so when it drifted down and hits zero, the charge pushes it back up to 40%. I have tried a few things to get it to reset back to 100 when full, but still no luck. I trust my voltage readings, and the Schneider charging to absorb limit, and the MPPT charge controller going to float.

 

[marcello]

Nominal capacity 200.81a

Yes I put 200 in the settings, i don't know why it shows 200.81 now. Strange. Anyway it should charge until this value then. From the moment i posted the message, it has now increased to 199.84 so i guess it is still charging but a bit slower than when it is discharged.

The battery state of charge reported by a BMS or battery monitor is only an estimate. At 199.81 amp hours on a 200 amp hour battery, it is clearly within 1% of full, but the cells may still be accepting some more charge. And it is very possible that amp hour count has drifted off as well. My JK-BMS has drifted the other direction. When my cells are near full, the BMS is only reporting 40% state of charge now. Each BMS or battery monitor has it's own rules for what must happen to trigger a reset to full 100% charged. In most cases, it is a combination of the voltage reaching a full threshold while the charge current also falls below a certain level for a given amount of time. Once it does a reset to 100%, it literally counts the amp hours removed during discharge to estimate how much is left. In a perfect world, if you discharged 40 amp hours out of a 100 amp hour battery, it would be at 60% state of charge. But what if the battery capacity is really just 98 or 102 amps hours? Or, when it the current sensor is a little off calibration? And what about some resistance loss that is losing power while charging so if thinks more power went into the battery but some of it was lost as heat? Each cycles that does not trigger a 100% reset can cause the state of charge to drift away from perfect. That is what happened on my JK BMS. After about 20 cycles, the SoC drifted down about 20%. But I know from the cell voltages that it is hitting 90% each day from sunlight. But for some reason, that is not causing the BMS to reset the SoC counter. After several months, mine got to where it was hitting zero SoC each night, even though the cells are really just dropping to about 50%, again, confirmed by the resting cell voltage and the fact they are hitting the absorb voltage after the correct amount of charging. So now each day the MPPT charger goes into float, the BMS only shows about 40% SoC. That sort of makes sense as I am trying to run the cells from 50% to 90% on daily cycles. That is a 40% change, so when it drifted down and hits zero, the charge pushes it back up to 40%. I have tried a few things to get it to reset back to 100 when full, but still no luck. I trust my voltage readings, and the Schneider charging to absorb limit, and the MPPT charge controller going to float.

Thank you much for sharing your experience. I wish i had your knowledge to make that kind of analysis.
Thanks to a member here i did a complete discharge of the battery by increments of 20% to get more precise values for the bms (the ones i entered in the settings) to prevent the bms from displaying wrong infos. But I think it still has to go through several cycles to display correct values.
Maybe my battery has more capacity than 200Ah, probably 210-220, but I don't want to modify the value in the settings because it will reset the bms memory.
I am expecting the charge to go to 14.5v, that's how I setup the bms and the victron mppt. I know it's a bit much and should decrease the numbers of cycles I can get over time but it's not a problem for me to get 2000 cycles instead of 3000, still plenty for my needs. I prefer to have a 100% charged battery in my campervan.
I will see if it reaches 14.5v.
Thanks for your help guys.

 

[Bob B]

Since the pack only shows 1 cycle .... it may not be tracking SOC via amps in vs amps out yet. Once it reaches the cell full voltage setting .... which is set at 3.504 .... it should start coulomb tracking and ignore the % cap settings unless a setting change is made and then the pack will have to go thru another cycle to full before coulomb counting starts again.

P.S. 3.2 is pretty low for starting balancing ... I would have that set closer to 3.4V ... to prevent the risk of unbalancing the pack by trying to balance at to low a voltage.

 

[GXMnow]

Since the pack only shows 1 cycle .... it may not be tracking SOC via amps in vs amps out yet.

Cycle count may only be counting "full cycles". Again, my JK-BMS does this. If I only cycle 35%, a day, it takes 3 days before it adds another cycle count. How far does it run down in a day? Run it down 50% and fully charge it, and it will then take 2 days to add 1 cycle count.

 

[marcello]

Since the pack only shows 1 cycle .... it may not be tracking SOC via amps in vs amps out yet. Once it reaches the cell full voltage setting .... which is set at 3.504 .... it should start coulomb tracking and ignore the % cap settings unless a setting change is made and then the pack will have to go thru another cycle to full before coulomb counting starts again.

P.S. 3.2 is pretty low for starting balancing ... I would have that set closer to 3.4V ... to prevent the risk of unbalancing the pack by trying to balance at to low a voltage.

hi thanks for your reply. The battery is still new I didn't use it a lot I've just finished preparing my van and started playing with it
I will wait for the cells to reach 3.5v to see what happens then.
For the balancing advice thank you. I didnt know there were a risk if we start the balancing too low.
I will change this value for the one you suggest. But I'm afraid the bms resets. I read if you modify some values in the settings, the app will make the bms forget all the history.

Cycle count may only be counting "full cycles". Again, my JK-BMS does this. If I only cycle 35%, a day, it takes 3 days before it adds another cycle count. How far does it run down in a day? Run it down 50% and fully charge it, and it will then take 2 days to a

Thanks for this info