Mysterious battery voltage spikes

[DrBourne]

Hey Guys,
I was hoping that someone could give me an idea as to these battery voltage spikes. No inverter in play at the time. see image below.
The 'Charging Status' reading "Not Charging" was there before I disconnected the PV. But, I could be wrong. Should have left it on before taking the screenshot.
When I use the inverter(24V 3000W) it would beep and shut off momentarily. Not under a heavy load, under 500W. This makes me afraid to use it to power my computers, which is my primary reason for setting it up.

gear;
eight 100W panels wired 4s2p
Epever Tracer 4215BN controller
Wzrleb 24v 3000W inverter
two 12.8V 200Ah in series

settings;
Maximum battery voltage today: 36.45V
Minimum battery voltage today: 26.7V
Battery type: USER_DEFINED
Battery capacity: 200AH
Over voltage disconnect voltage: 29.4V
Charging limit voltage: 28.8V
Over voltage reconnect voltage: 28.4V
Equalize charging voltage: 28.6V
Boost charging voltage: 28.6V
Float charging voltage: 27.4V
Boost reconnect charging voltage: 27V
Low voltage reconnect voltage: 24.8V
Under voltage recover voltage: 24V
Under voltage warning voltage: 23.2V
Low voltage disconnect voltage: 21.6V
Discharging limit voltage: 21.6V
Battery rated voltage: 24V
Default load on/off in manual mode: OFF
Equalize duration: 0 min
Boost duration: 60 min

Thanks,
Dale

 

[sunshine_eggo]

Did you charge each of the 12V to 14.4V for 2 hours and then again in parallel before putting them in series? If you didn't, you should now.

Extremely common for new/new-ish batteries with imbalance at the top. BMS triggers charge protection, and the MPPT can't clamp down in time to prevent a voltage spike.

Do the 12V individual/parallel charges.
Reduce boost to 27.6V, 120 minute boost time.
Reduce float to 27.0V

The above should reduce/eliminate the BMS protection events and allow the BMS time to balance the cells.

 

[timselectric]

And repeat, every 6 to 12 months.
Or less, if the issue comes sooner.

 

[Shimmy]

Looks to me like the battery is just responding to the no-load solar transients. The energy needs to go somewhere until the MPPT catches up with the PV to throttle things back.

 

[sunshine_eggo]

Looks to me like the battery is just responding to the no-load solar transients. The energy needs to go somewhere until the MPPT catches up with the PV to throttle things back.

I'm not following you. It's possible you're making perfect sense, but it's a bit late, and I'm getting older. "energy needs to go somewhere" makes no sense whatsoever. Please help me with this. Per the settings, should charge to boost voltage, hold for 60 seconds and then drop to float voltage. I'm not seeing that at all.

Here's what I see:



All lines are voltage. Red is lower because current is being drawn to charge. Suddenly, a battery voltage spike to > 35V and the PV voltage peaks via a vertical line. Then no current whatsoever is being drawn - BMS charge protection. Stays there for awhile at Voc. Then it starts to draw current again, PV voltage lowering, then BAM! Voltage spike and PV voltage goes to max. Repeat two more times.

The best explanation for the above is BMS charge protection.

 

[Shimmy]

Suddenly, a battery voltage spike to > 35V and the PV voltage peaks via a vertical line.

I guess my question is if the PV Voltage is cause or effect of the Battery voltage change; I can see no reason why it would be an effect.

The brevity of the event does look like BMS prevents charging, but that also looks like a secondary effect.

I guess my question is what is causing the PV Voltage to behave like that; I am not surprised that the behavior would cause instability on the battery rails.

 

[McKravitts]

The charge cut off voltage on an EPever tracer is slightly highr than the shut off voltage on many inverters.
I had to set mine to the user settings and set the max charge voltage lower to match the inverter chargers cut off voltage.

 

[DrBourne]

Did you charge each of the 12V to 14.4V for 2 hours and then again in parallel before putting them in series? If you didn't, you should now.

Extremely common for new/new-ish batteries with imbalance at the top. BMS triggers charge protection, and the MPPT can't clamp down in time to prevent a voltage spike.

Do the 12V individual/parallel charges.
Reduce boost to 27.6V, 120 minute boost time.
Reduce float to 27.0V

The above should reduce/eliminate the BMS protection events and allow the BMS time to balance the cells.

Yes, I got the settings for the battery from Redodo, who also advised me to charge each one separately prior to paralleling them.
Made suggested setting change, will keep an eye on it.
Thanks for the assist.

 

[DrBourne]

Looks to me like the battery is just responding to the no-load solar transients. The energy needs to go somewhere until the MPPT catches up with the PV to throttle things back.

The above occurred with no-load.
Any suggestions as to remedy? When in use(seldom), the inverter momentarily shut down even under light load, 20in. box fan, or slightly higher desktop pc and monitor.

 

[DrBourne]

I'm not following you. It's possible you're making perfect sense, but it's a bit late, and I'm getting older. "energy needs to go somewhere" makes no sense whatsoever. Please help me with this. Per the settings, should charge to boost voltage, hold for 60 seconds and then drop to float voltage. I'm not seeing that at all.

Here's what I see:

View attachment 169988

All lines are voltage. Red is lower because current is being drawn to charge. Suddenly, a battery voltage spike to > 35V and the PV voltage peaks via a vertical line. Then no current whatsoever is being drawn - BMS charge protection. Stays there for awhile at Voc. Then it starts to draw current again, PV voltage lowering, then BAM! Voltage spike and PV voltage goes to max. Repeat two more times.

The best explanation for the above is BMS charge protection.

This all makes sense(sort of). The BMS are internal on the batteries. Any idea as to how to figure out which is which, cause an/or effect?
Thanks for the reply. All you guys are really ready to help. And I greatly appreciate the efforts.

 

[DrBourne]

The charge cut off voltage on an EPever tracer is slightly highr than the shut off voltage on many inverters.
I had to set mine to the user settings and set the max charge voltage lower to match the inverter chargers cut off voltage.

I did similar, but .5V lower than inverter cutoff.

 

[DrBourne]

I really don't care for the epever 'Solar Station MonitorV1.95" software. While it provides 'real-time' stats, I want more.
Am working on something to provide longer term tracking. Hopefully, if/when it's working, I can get more details.
I am thoroughly impressed how you guys jump in to help. MUCH better than most of the Linux forums I've utilized!

 

[Shimmy]

Did you charge each of the 12V to 14.4V for 2 hours and then again in parallel before putting them in series? If you didn't, you should now.

Extremely common for new/new-ish batteries with imbalance at the top. BMS triggers charge protection, and the MPPT can't clamp down in time to prevent a voltage spike.

@sunshine_eggo got it right at the start and I missed it.

Do the 12V individual/parallel charges.
Reduce boost to 27.6V, 120 minute boost time.
Reduce float to 27.0V

The correct action.

Another approach would be to get a second charge controller and dedicate a 4s string of PV to each charger and a charger to each battery. Balancing them should be enough, but that is a backup plan.

...it's a bit late

That was part of my problem as well... I saw what looks like cloud edge on the PV voltage and thought that was the source of the problem on the battery bus. While that is still possible I think battery balance needs to be addressed first.

 

[DrBourne]

Did you charge each of the 12V to 14.4V for 2 hours and then again in parallel before putting them in series? If you didn't, you should now.

Extremely common for new/new-ish batteries with imbalance at the top. BMS triggers charge protection, and the MPPT can't clamp down in time to prevent a voltage spike.

Do the 12V individual/parallel charges.
Reduce boost to 27.6V, 120 minute boost time.
Reduce float to 27.0V

The above should reduce/eliminate the BMS protection events and allow the BMS time to balance the cells.

I offer my heartfelt gratitude. Your suggestion to alter the boost and float voltages worked like a charm so far. Been using under light inverter load with NONE of the spikes. Soon to put under significant load, wish me luck!
Thanks again.

 

[sunshine_eggo]

I offer my heartfelt gratitude. Your suggestion to alter the boost and float voltages worked like a charm so far. Been using under light inverter load with NONE of the spikes. Soon to put under significant load, wish me luck!
Thanks again.

Glad to hear it improved. In a week or so, try running to a higher voltage and check to see if the BMS protects again. Or, if you're happy with the performance at 3.45V/cell, just leave it be. You should be able to get to 98%+ SoC @ 3.45V/cell with a 2 hour absorption.