JK BMS configuration - charging problem?

[smobley]

So have have built up two of four DIY boxes using 314ah cells with JKBMS V19 (19.09) and am now stepping into the wonderous world of BMS setup.
I have used the app to set the batteries to LiFePo4 and then used Off-Grid Garages JK Inverter BMS (JK-PBxx Series) setup initially for the BMS settings that you can find on his website.

My Settings

I share here the settings of all devices I have configured for LiFePO4 batteries. As a beginner, it is very confusing to understand all these different options solar charge controllers, BMS, shunts…

off-grid-garage.com

I have connected a Chargeverter (120v supply) directly to the isolated battery and started the charge.

I measured the voltage at the pack outputs and set the calibrated voltage to that value.
The Chargeverter shows 10.1amp, my Klein clamp meter shows 12.5amp on the cable.
I set the calibrated current to 12.5.

Everything seems fine as charge starts. Charged overnight, and in morning cell voltages seem fine and climbing as it continued charging.
Checked several more times, and then got busy with other stuff.
Went out and checked cell voltages and they were up in the 3.71v range!? I immediately stopped the charge to figure out what is going on.

So I am confused. Why did it go over the OVP setting of 3.65? Is that not Over Voltage Protection? I assumed that the charge would drop off as the batteries reached that BMS setting. Now I am questioning life, battery life. Newb Help Needed!

 

[Hedges]

I have connected a Chargeverter (120v supply) directly to the isolated battery and started the charge.

Connected directly to the cells? Or through BMS?

BMS should disconnect if any cell exceeds maximum limit it is configured for.
But you should set charger voltage such that no cells are expected to hit that limit.
However, some "runner" cells might do that if not balanced. So could start at lower charge voltage, give time to balance, step up pack voltage gradually.

Everything seems fine as charge starts. Charged overnight, and in morning cell voltages seem fine and climbing as it continued charging.
Checked several more times, and then got busy with other stuff.
Went out and checked cell voltages and they were up in the 3.71v range!? I immediately stopped the charge to figure out what is going on.

All cell voltages? Or just some 3.71V? List all cell voltages.

Andy's settings you linked do show "Cell OVP(V) 3.650V"
Double-check yours.

Compare BMS report of voltages to your DMM measurements.

 

[smobley]

Hey Hedges, Thanks for looking.

First its connected to the battery (should I have said pack?), not a cell (would seem odd connecting a chargeverter to a cell no?) However, I am sure it wouldn't be the first time something like that was done.

All cells in the pack display page showed at the time within .002 of the 3.71.
Already double checked the OVP parameter and its showing the entered value of 3.65

I haven't done individual cell volts comparison meter to display, so I'll go do that now and report back.
Okay, cell voltage display vs meter differences is only about .002v.
Pack voltage showing the same as well.

 

[kommando]

First its connected to the battery (should I have said pack?), not a cell (would seem odd connecting a chargeverter to a cell no?)

You could connect directly to the pack between the BMS and the cells, leaving the BMS not in control and unable to disconnect the pack from the charger..

 

[smobley]

Ah, I see, if I had a custom battery build that could very well happen. I am using the Yiziang DIY box and using the big red and black terminals on the front. I think next time I might just include a picture as well as it would clarify a lot of possible hardware configurations.

 

[Hedges]

First its connected to the battery (should I have said pack?), not a cell (would seem odd connecting a chargeverter to a cell no?) However, I am sure it wouldn't be the first time something like that was done.

Connected directly to the cells? Or through BMS?

"Cells", as in 16s string of cells for nominal 48V.

My question is whether ChargeVerter output goes through BMS (so BMS could disconnect), or ChargeVerter output bypasses BMS.
Because I figured BMS should disconnect, and if it couldn't, maybe you wired it wrong.

I haven't charged mine up to the point of disconnect, so I can't confirm JK does what it is told.

Next theory, you were reading 3.71V with DMM but BMS read less than 3.65V so didn't disconnect.

All cells in the pack display page showed at the time within .002 of the 3.71.

This addresses that question, 3.71V was reported by BMS, so we do expect it to disconnect.

Good news is well balanced charge on cells.

Is this 48V, 16s battery? 3.71V x 16 = 59.36V
Your charger voltage setting is too high.

I haven't done individual cell volts comparison meter to display, so I'll go do that now and report back.

That should show if any error in BMS measurement, but won't explain failure to disconnect.

I'm left with two theories:

1) BMS fails to disconnect when max cell voltage exceeded
a) Functional bugs
b) Damaged transistors

2) Operator error
a) Wired wrong, bypassing BMS (as kommando also proposes is possible reason)

Maybe there is something else we haven't thought of.

 

[Hedges]

Photo is good, although sometimes not traceable.
Schematic also good, it is our way of communicating intent.

 

[smobley]

On the charger, it was all default whatever that setting was.
Here is a shot of the power connections and the Chargeverter settings

 

[Hedges]

"Voltage 54.0V"
Is that the present voltage? Or the limit that is set?

"Battery Stop 54.0V"
Oh, that is the limit. And it is NOT 59.36V

Confirm it is 16 cells? (I can't count in picture)

54.0V / 16 cells = 3.375V/cell

I could have imagined one cell to be a runner and hit 3.71V, but not all 16 cells.
Double-check 54.0V with DMM

Outside battery wiring does not look wrong.
Now need to see what's inside. How the terminals to to BMS and cells.

I figure 3.71V still charging for a short time won't cause damage. Sitting there longer would bring to high state of charge, closer to damage. Good you found it early and stopped.

 

[smobley]

So I can now see how 3.71v came about, but shouldn't the BMS limit the input voltage?

I guess I might be thinking to much about how an alternator puts out 14ish volts for charging a car battery that is 12v.
Is the not the case for charging LiFePo4 batteries?

Should the max charge voltage be whatever you want the max battery voltage to be?
In this case I need to change the Chargeverter to output 3.65v max?

 

[Hedges]

BMS doesn't limit voltage, only disconnects for over-voltage or other issue.
(Some closed-loop with inverter let BMS request voltage from inverter.)

Yes, max charge voltage the target for battery. Less than the over-voltage disconnect for battery.
You might want 3.55V per cell, so 56.8V on chargeVerter? Or less if only to avoid draining all the way, if you have inverter with PV as well.

I don't see how 3.71V happened. What was the explanation?

 

[smobley]

I kept looking at what you posted for voltages and what the Chargeverter (CV) settings were and where the problem could be, because how can you get 3.71 out of 54.0v with a 16s...the answer is you can't.

With multimeter and calculator in hand I set off in my expedition, and I discovered a measurement anomaly between my left ear and my right ear. Apparently there is an open circuit as I read no potential.

Sooo in a nutshell, there was a 371, but it was 3.371. A Chargeverter that isn't broken has a max output voltage setting is only 57.0v so it could never overvolt these battery packs. (OVP set at 3.65v per cell * 16s = 58.4v, 57.0v / 16s = 3.5625v per cell).
I'll go back to my corner now and write 100 times "I will check my decimal points before posting."...in cursive!

Now, I changed settings on the CV to 57.0v and stop at 57.0v, Almost all batteries charged to around 3.548 (13 of 16), but one charged to 3.65 and the BMS cut the charge current and set the "battery charged" message on phone app so that is working, however two only charged to 3.480. It is also showing the error light, but I can't see anyway in the phone app to see what the error is. I am assuming its the cell delta, but what do I know as I can't read decimals.
So now I have to figure how to see the error, and get those cells balanced up to the others.

 

[Hedges]

OK, no big problem!
Being within couple mV at 3.371 is nothing to brag about.
Now you have 170 mV imbalance.

By setting chargeverter to something which doesn't quite hit 3.650V for the runner, it should sit there while BMS balances.
Wash, rinse, repeat.

 

[smobley]

Thanks, I'll try that.

 

[smobley]

So sneaking up on 57.0v on the CV worked. The runner hardly moved.
Pack went from a diff of around .063, down to under .010.