JKBMS balance lead voltage fluctuating

[alexcl]

I've noticed that when charging/discharging my eve cells (24 304Ahs in 3p8s), some of my cell blocks have much higher voltage fluctuations according to the bms. It seems to be the furthest blocks from the BMS that have this issue. For example, I see fluctuations of 3.23V to 3.251V in the 8th block while discharging and fluctuations of 3.360V to 3.386V while charging (both over the course of a few seconds). Both of these were at ~50 amps. My multimeter doesn't show these fluctuatoins. Is this normal?

A few things to note:
I did extend some of the balance leads
The 6th cell block is connected to the 7th with 4/0 cable instead of bus bars (as the 7th and 8th block had to be in a different location)
The cells are not yet top balanced

I checked and tightened all my bus bar connections, tried multiple different charging sources (the problem lessens dramatically at lower charging amperage), and tried multiple cables connecting the 1-6s to 7-8s.

Any idea what is causing this? Is it a problem?

 

[sunshine_eggo]

It's my understanding this is somewhat normal behavior until the BMS calculates and iterates on the sense lead resistances to apply a correction factor when displaying voltage.

What does the BMS report for resistances?

 

[alexcl]

It's my understanding this is somewhat normal behavior until the BMS calculates and iterates on the sense lead resistances to apply a correction factor when displaying voltage.

What does the BMS report for resistances?

Interesting - is that iteration something that would take longer than a couple minutes?

The resistances are anywhere between 52 and 62 milliohms:

 

[sunshine_eggo]

Those look pretty typical for good connections. I have seen those values change as users have made follow-on posts after a day or several.

I also had it in my head that this was about active balancing, but it's clearly not as you're not in the balance range (or at least you shouldn't be balancing in this range).

Bad connection of some kind - maybe even at the BMS connector plug OR a bad BMS.

 

[alexcl]

Okay, so just to make sure these per cell block voltage swings of (0.02-0.03V) are not typical?

 

[sunshine_eggo]

Okay, so just to make sure these per cell block voltage swings of (0.02-0.03V) are not typical?

If you're saying that under a constant current charge or discharge, the individual cell voltages fluctuate across a 0.03V range, that feels wrong. If you're talking about the 3rd decimal place, most BMS are pretty spastic there.

 

[alexcl]

Yes, it's fluctuating 0.02-0.03V. I actually bought a backup BMS in case anything goes wrong when I'm on the road (this is being installed in a campervan). I just hooked it up and tested it (different balance leads from BMS, same busbar connections across cells). I'm seeing the same types of fluctuations.

It does look like the input current is not quite constant (but within 5%). I saw this same thing with my benchtop PSU however (to a lesser degree since the problem seems to scale with amperage and the PSU maxes at 10A).

 

[sunshine_eggo]

Yes, it's fluctuating 0.02-0.03V. I actually bought a backup BMS in case anything goes wrong when I'm on the road (this is being installed in a campervan). I just hooked it up and tested it (different balance leads from BMS, same busbar connections across cells). I'm seeing the same types of fluctuations.

It does look like the input current is not quite constant (but within 5%). I saw this same thing with my benchtop PSU however (to a lesser degree since the problem seems to scale with amperage and the PSU maxes at 10A).

If it's the same BMS using the same harness/connector, it only excludes the BMS as the likely fault.

 

[alexcl]

It's the 8 lead version of the same JKBMS using it's owning harness/connector. Interestingly, the problem is almost gone when using the PSU (at 10A) but still exists when charging with the inverter (at 50A). Not sure what else to do/try here.

 

[Quattrohead]

Looks all good to me, any movement less than 10mv is fine, I typically see 3-5 mv jumps.

 

[sunshine_eggo]

Looks all good to me, any movement less than 10mv is fine, I typically see 3-5 mv jumps.

he's saying 20-30mV unless I've been mistaken. My JBD can be spastic in the 3rd decimal place.

 

[alexcl]

Exactly - the jumps are 20-30mV.

Update: I trimmed back some of the rubber on my 7-8 connection and it looks like the problem improved (but not entirely) for the 8 cell block. I'm still getting the problem with the 7 cell block (which has always been the chief offender). As mentioned the connector between the 6 and 7 cell blocks is 4/0 awg cable and I noticed that there is a ~25mV difference between the block 6 end of that cable and the block 7 end of that cable when charging at 50 amps.

Would higher resistance cell connections cause this kind of voltage fluctuation on the leads?

 

[sunshine_eggo]

How long is the 4/0?

 

[alexcl]

How long is the 4/0?

~10ft of 4/0 awg copper cable. It's resistance (as measured by charging at 50A) is ~0.5 milliohm which is consistent with online calculators. I understand why that would cause a voltage drop, but am confused by why it would cause voltage fluctuations.

 

[sunshine_eggo]

You have 10' of 4/0 connecting two adjacent cells? Wow. I guess I just never imagined that.

4/0 @ 10', 50A and 3V has a 0.050 drop.

I would attribute it to the wild disparity in resistance of that "bus bar".

 

[alexcl]

Interesting - so the resistance of a bus bar affects how the voltage reading of the corresponding cells fluctuates? Why don't I see this with a multimeter? Is it something about how the BMS calculates voltage?

It's not actually 3V though, IIUC. It's the voltage between the two ends of the cable (0.025V). Using the multimeter I saw 25mV between the ends of the cable with 50A passing through meaning 0.0005 ohms.

 

[sunshine_eggo]

Interesting - so the resistance of a bus bar affects how the voltage reading of the corresponding cells fluctuates? Why don't I see this with a multimeter? Is it something about how the BMS calculates voltage?

It's not actually 3V though, IIUC. It's the voltage between the two ends of the cable (0.025V). Using the multimeter I saw 25mV between the ends of the cable with 50A passing through meaning 0.0005 ohms.

Been on the Internet before there was a web, and I had to look up IIUC.

I looked again after International Islamic University, Chittagong just didn't see to fit.

The bus baaaaaaaaaaaaaaaaaaaaaaaaaaar (see what I did there) is part of the circuit across the cell. Consider also that large LFP cells are around 0.0002Ω, so your wiring resistance is actually higher than the cell itself.

Unfortunately, I think the extreme nature of the installation is resulting in atypical behavior.

Have you charge to the point that max balancing is attained?

 

[alexcl]

Been on the Internet before there was a web, and I had to look up IIUC.

I looked again after International Islamic University, Chittagong just didn't see to fit.

Hahah, TIL

Unfortunately, I think the extreme nature of the installation is resulting in atypical behavior.

Interesting I'm curious as to why if you have any thoughts. Unfortunately the closest I can get the two boxes is ~2ft. I just observed the same behavior with a 2 ft cable:

https://imgur.com/a/9GXCjEn

Have you charge to the point that max balancing is attained?

What is max balancing? I'm not finding a defintion.

Btw, thanks for all the intel and help debugging/diagnosing.

 

[sunshine_eggo]

Hahah, TIL


Interesting I'm curious as to why if you have any thoughts. Unfortunately the closest I can get the two boxes is ~2ft. I just observed the same behavior with a 2 ft cable:

https://imgur.com/a/9GXCjEn

and that's just sitting there, no charge/discharge...

What is max balancing? I'm not finding a defintion.

JK is an active balancer with a maximum balance current. It should only balance above 3.40V. "max balancing" means where the BMS engages balancing to it max current.

I just checked my JBD sitting there doing nothing, and it wasn't as pronounced as I remember. It's only "twitching" .001V every 4-5 seconds.

Btw, thanks for all the intel and help debugging/diagnosing.

Heh... FWIW...

 

[alexcl]

and that's just sitting there, no charge/discharge...

Yep, though I had been charging it at 50 amps a few minutes beforehand

JK is an active balancer with a maximum balance current. It should only balance above 3.40V. "max balancing" means where the BMS engages balancing to it max current.

Ahh, I turned off balancing so I could isolate a charge source. Why would balancing make a difference in terms of the fluctuations?