0.08V cell differential at 3.4V?

[kolek]

According to @Steve_S who seems to have a lot of expertise on the subject, "a healthy cell differential is <20mv when static (not charging or discharging)" (see here)
and I'm seeing 0.08V differential at 3.4V with a small 3A load. At lower voltages, the cell differential is usually around 0.003V.

Is this an issue and if so how to correct?

This is a new battery build, 16S, Grade A 280AH cells from Luyuan, JK-BMS, the cells were top balanced to 3.53V (not to 3.65V).

 

[kolek]

Always the same cell is too low, and the JK-BMS has highlighted the cell indicating it's trying to boost the voltage, but the voltage seems stuck on that cell, the BMS doesn't seem able to raise the voltage.

 

[kolek]

selected JK-BMS settings:
start balance voltage: 3.0
max balance current: 2.0
continued charge current: 100A
continued discharge current: 200A

 

[sunshine_eggo]

start balance at 3.0V? HUH? balancing below 3.40V is pointless and potentially counterproductive.

I prefer to charge to 3.45V/cell with balancing starting at 3.40V and 20mV difference.

It can take awhile to balance cells even with active balancing.

 

[kolek]

Measured cell 9, the cell which supposedly is low according to the JK-BMS.
3.344 JK reported voltage
3.358 actual using multimeter
3.358-3.344=0.014

for reference measured cell 8
3.429 JK reported voltage
3.440 actual using multimeter
3.440-3.429=0.011

So I guess that's a normal differential between multimeter measurement and the BMS
presumably JK is measuring with load, multimeter is not

 

[AshleyL]

Measured cell 9, the cell which supposedly is low according to the JK-BMS.
The JK-BMS says the voltage is 3.344. My multimeter says it is 3.358

for reference measured cell 8
3.440 actual using multimeter
3.429 JK reported voltage

So....which one do you trust?

Granted that you can calibrate the JKBMS......but if the multimeter is inaccurate....would you still calibrate jkbms to the multimeter?

 

[kolek]

Cell 9:
Cheapo Chinese multimeter: 3.376V
Klein Tools $100 multimeter: 3.358V
JK-BMS: 3.346V

Bottom line though, cell 9 is too low and seems resistant to correction by the BMS.

 

[kolek]

balancing starting at 3.40V and 20mV difference

Ok changed balancing start voltage from 3V to 3.4V

By 20mV difference, you mean setting Balance trig. voltage = 0.020?
My setting now for that is 0.003

 

[400bird]

I think you are worrying about nothing.
You have 280 amp hour cells and 2 amps max balancing.
That's not going to balance immediately. It will take many cycles. Even then 0.008 volts is tiny.

I would measure all the cell voltages and compare each to BMS readings. In theory, they should all show roughly the same delta (about 0.011 from your posts above.

 

[kolek]

Even then 0.008 volts is tiny.

Thanks for your input. By the way, it's 0.08V, or 80mv, not 0.008.

 

[SeaGal]

Bottom line though, cell 9 is too low and seems resistant to correction by the BMS.

IMHO, if it is just the one cell that is too low, then, rather than wait for the BMS to balance you could disconnect the inverter and just charge up that one cell up using an isolated bench PSU.

I detailed that a bit more in this posting (and also see earlier in that thread re discharging a cell that is too high)

One EVE 280Ah cell tripping BMS high voltage at top end

I have a self-built 280Ah 48v battery pack, that was working great. I initially did a top balance, parallel to 3.65v For a couple weeks now I've been having issues with one cell, #16 which happens to be my main positive for the entire battery. At the top end when the inverter is pushing the...

diysolarforum.com

 

[kolek]

@SeaGal (or anyone else) Thanks for that suggestion!
In your opinion does a 0.08V differential represent a concern that needs to be manually addressed, or should I just wait for the BMS to try and sort out the difference over time? @Steve_S says less than 20mv is healthy, this is 4X that value, but not sure how concerned I should be about it. Again, I'm only seeing the difference at higher voltage, when the SoC drops the differential drops to about 0.003V.

 

[sunshine_eggo]

dV severity depends on cell voltage, and if it interferes with normal operation.

If you're 3.52-3.60 (.08 dV), then it's mostly a non-issue.

If you're 3.32-3.40 (.08dv), then you're not balanced.

My goal is to be < .02V at 3.45V/cell.

 

[SeaGal]

... if it interferes with normal operation

LOL.. it's a DIY Solar Forum... we're all a bit OCD or we wouldn't be on here. So, yes, when you think about 80mV differences at night, when having a shower or sat on the loo, then yes, it does interfere with normal operation

 

[sunshine_eggo]

LOL.. it's a DIY Solar Forum... we're all a bit OCD or we wouldn't be on here. So, yes, when you think about 80mV differences at night, when having a shower or sat on the loo, then yes, it does interfere with normal operation

HAH! "normal operation" = normal cycling of the system without BMS intervention.

You have to keep the inside stuff on the inside...

 

[Lt.Dan]

Yes, 20mV is a very good deviation. This is if ALL cells are over 3.4v though.

Klein Tools $100 multimeter: 3.358v

If this cell is 3.358v, then it is SEVERELY out of balance. Like recommended above, I would charge this single cell, isolated, with a benchtop power supply to get it up over 3.4v.

Me personally, I charge to 56.8v, and if all cells are over 3.45v, and none have hit HVD (3.65v), then thats good enough for me. And that could be a 200mV deviation. The battery is absolutely charged to 100%, and is still functioning within its operating parameters.

 

[Hedges]

I haven't worked with these, but it seems to me one cell low (not high) won't be a runner, so I would charge pack to a higher voltage. The low cell would then stand out and could be charged by bench supply to line up with the others.

If charging low cells was done at low voltage, before it broke away from the pack (dropping behind), it would be too easy to charge it above the others.

While boosting it with a supply, aim low so as to not overshoot. Or not, because it will settle back? That's for the experienced members, not he LiFePO4-naive like me, to say.


What Lt.Dan said.

 

[hwy17]

I'm confused how you're seeing 3.4 under any load, but I have never closely observed cells under light loads right after a full charge.

 

[Ampster]

cell 9 is too low and seems resistant to correction by the BMS.

As mentioned earlier. I can take a long time for even an active balancer like the JK to balance one cell. Have you adjusted the settings on your inverter as suggested and given it a few cycles?

 

[upnorthandpersonal]

First thing to do: calibrate the JK.

 

[kommando]

Started charging my new 48V DIY LFP battery today, got close to 55.2V and the delta was 0.090V, it had been 0.012 when first built and resting but increased as it was charged today. Balancing 2A active on the JK PB kicked in set to 3.45V per cell and 2 hrs later it is 0.049V. When it was at 0.070V delta I checked the cell voltages with a multimeter and the JK BMS reported cell voltages are correct but rechecked and retorqued the lowest cell anyway, voltage did not change. I had recalibrated the JK BMS yesterday and it did not need recalibrating today vs multimeter. I will monitor over next few days but with bad weather forecast the 3.45V balancer threshold may not be breached for 3 or 4 days so doubt there will be much change on where it gets to today. Will just have to be patient until the balancer does its thing, it only transfers between the current low and current high cell so its not in a rush. I have a lab power supply to charge single cells if needed.

 

[kolek]

It's 2AM in Japan. Will respond tomorrow, thanks everyone for the tips!

 

[sunshine_eggo]

It's 2AM in Japan. Will respond tomorrow, thanks everyone for the tips!

Ummm.... 2am is today... are you going to make us wait like... 22 hours until it's tomorrow?

 

[kommando]

Delta is now 0.041V, cell 15 was always the lowest but now cell 6 is the lowest. The balancer works.

 

[Lt.Dan]

The JK 2a BMS actually balances very well. My main BMS's that i use in my house are only 200ma passive balancers. I do have 1x JK BMS for testing and such and MAN does it balance quickly. I usually charge until a HVD, then leave the power supply on overnight and let the balancer do its work. I've never had a pack take longer than 12 hours to get under 20mV at 58v doing this.