How to get Pace BMS to reliably reset to 100% SOC

[CourtJesterau]

I have a deye inverter and 4 x 280 AH batteries in parallel each with a 200A pace BMS

Over the last couple of weeks I have been playing around with the settings on the Pace BMS trying to get the batteries to switch over to a 100% SOC

The Deye inverter uses the pack OVP alarm voltage set in the pace BMS to set its charging voltage

I have this set to 57.4v and OVP protection set to 58.4v and release at 57v
cell OVP is 3.65 and release is at 3.55v

the "pack full charge" voltage is set to 57v and 2000mah set for the full charge current

either battery 1 or 2 generally does not reset to 100% SOC even though they are at 57.4v and are not drawing any more current i.e. well under 2500mah

screen shot of bms below The only reliably way I can get it to reset to 100% soc is to force the pack into OV protect by settings the OV protect to 57.3v and then it resets the SOC to 100% but surely this is not how it should be done my understanding the settings I ahve now should be reliably setting it to 100% soc all ready and ideas as to what might be going on?

 

[MisterSandals]

either battery 1 or 2 generally does not reset to 100% SOC even though they are

SoC “calculations” from a BMS are almost always crap. Don’t waste your time (our time?) trying to make sense of it.

 

[marionw]

Need a screenshot of the "Realtime Monitoring" tab

 

[CourtJesterau]

Need a screenshot of the "Realtime Monitoring" tab

when I get home I will take some actual screenshots and post them up but I do know it wont be marked as fully charged in the monitoring tab as I have looked at this everyday for a week now trying to get it to reset.

 

[CourtJesterau]

SoC “calculations” from a BMS are almost always crap. Don’t waste your time (our time?) trying to make sense of it.

the problem is it creeps down and fluctuates and I have automation etc based on SOC so I do want this as accurate as possible or at least consistently inaccurate which needs the starting point to be reset to 100 to be consistent

 

[CourtJesterau]

I have 2 x different hardware version of the BMS pack 1 and 2 are the same version as easht other but are different to back 3 and 4



here are the screen shots of pack 1 and pack 4 -- they are different versions of the BMS but pack 4 as you can see has the "fully"lit up and the pack 1 does not have it lit.

the parameters are the same on all 4 packs


Pack 1 details






Pack 4 details

 

[robby]

SoC “calculations” from a BMS are almost always crap. Don’t waste your time (our time?) trying to make sense of it.

It worked fine for me for three years and matches the shunt numbers I am seeing.
Your statement is based on which BMS or do you really think all of them are inaccurate?

 

[MisterSandals]

Your statement is based on which BMS or do you really think all of them are inaccurate?

Most of my BMSs are JBD and in my experience the SoC calculations are inconsistent between units as well as compared to my Victron shunts (which are very accurate and consistent).

 

[CourtJesterau]

It worked fine for me for three years and matches the shunt numbers I am seeing.
Your statement is based on which BMS or do you really think all of them are inaccurate?

its not the soc calculations its the fact it does not reset to 100% when it should be that is the issue where 2 x of the packs do and 2 x dont with the exact same settings I dont care so much at the calculations but the fact it does not reset back to 100% It can never be accurate if the starting point is always changing

 

[marionw]

1) I would lower the "Pack FullCharge Voltage(V)" to 56 (16 cells x 3.50vdc per cell).
2) Depending on where your computer is connect to the batteries you may or may not be reading the configuration parameters from the battery you want. If for example you are connected to the RS232 jack on the "Master" you may be reading the "Master" configuration parameters rather that those of Pack 4. If connected to the Master then select 4 for the "Pack" before opening the serial port. To be sure you can connect your computer directly to the Pack 4 RS232 jack.

If you connect to the "Master" RS232 jack and select FF for the pack number before opening the serial port you will be able to view each battery by clicking the button near the top of the screen for the battery you want to view. In addition when you change a parameter is is applied to all batteries.

Cycle the batteries from a full change down to 20%-30% SOC then re-charge. I have found that at times a full charge and discharge is required.
In addition I generally try to keep the charge current to around 10 amps which helps to prevent a larger "Max Cell DIff Voltage"

 

[hybe022]

The Deye inverter uses the pack OVP alarm voltage set in the pace BMS to set its charging voltage

I wonder if this is the same for other brands like luxpower / eg4
I will try changing it to test

1) I would lower the "Pack FullCharge Voltage(V)" to 56 (16 cells x 3.50vdc per cell).

This is only for SOC estimation right?
I don't really care much about soc.
I charge at 55.2V so i think i'll set this at 55.2?

 

[CourtJesterau]

I wonder if this is the same for other brands like luxpower / eg4
I will try changing it to test


This is only for SOC estimation right?
I don't really care much about soc.
I charge at 55.2V so i think i'll set this at 55.2?

I have automations set at based on SOC so I want it to be as accurate as I can get it

and yes in the BMS info on the Deye inverter is changes the charge voltage according to what the pack OVP alarm is set at in the pace software

I fully discharged and re charged the batteries over the weekend and pack 1 and 2 are still not resetting to 100% ( though they are @ 99.4%) but wont fully reset correctly so I dont know

packs 3 and 4 are fine there must be some difference in hte firmware for these BMS and these 2 x packs just wont reset unless it hits the OVP or cell OVP protect status.

 

[CourtJesterau]

Over the last week the 4 x batteries in the so are creeping apart even though both are fully charged Bay 3 and 4 reset to 100% and bat 1 and 2 don't and stop charging now when sic isn97% fully charged and balanced but does not reset the so this is an issue for me with the automations etc any one have any other ideas as to what I can't try?

 

[LydMekk]

12V * 4 will always be creeping apart. Bad solution.
If you need 48V batteries buy a 48V battery directly.
Sell the ones you have and buy a proper 48V battery bank which is big enough for your normal loads during dark hours.

 

[CourtJesterau]

12V * 4 will always be creeping apart. Bad solution.
If you need 48V batteries buy a 48V battery directly.
Sell the ones you have and buy a proper 48V battery bank which is big enough for your normal loads during dark hours.

they are 48v batteries there is 4 x 280Ah packs all in parallel with their own bms i mean just clicking on one of the images posted above would have shown you that ........

2 x of the BMS in the batteries refuse to reset to 100% even though they are fully charged

 

[RCinFLA]

Two most common ways for resetting BMS Columb counter for SoC calculations are:
- Tripping a cell overvoltage limit.
- Reaching total battery stack absorb level voltage .AND. absorb voltage held for the set absorb cycle timer.

Three most common reasons for SoC error.
- Lack of achieving 100% full battery Columb counter reset over a long time period causing large error accumulation.
- Current measurement accuracy, particularly at low currents less than a few amps.
- DC offset in current shunt sense voltage operational amplifier.

 

[CourtJesterau]

Two most common ways for resetting BMS Columb counter for SoC calculations are:
- Tripping a cell overvoltage limit.
- Reaching total battery stack absorb level voltage .AND. absorb voltage held for the set absorb cycle timer.

Three most common reasons for SoC error.
- Lack of achieving 100% full battery Columb counter reset over a long time period causing large error accumulation.
- Current measurement accuracy, particularly at low currents less than a few amps.
- DC offset in current shunt sense voltage operational amplifier.

they are pace BMS I ahve the same settings on all 4 batteries full voltage is set to 57v with a current of 2000mah 2 x of the packs reset to 100% under these conditions 2 x dont

I charge at 57.1 v all the packs get to this voltage and the current drops well below the 2A to trigger the reset but they dont ( see BMS settings above in screen shot) I have changed these settings I can reliably have ti reset if I hit over voltage protection on the pack or cell but dont want to constantly be bouncing off that to get the packs to reset and should not need to as per my understanding of the Pace BMS

 

[RCinFLA]

Many BMS's only reset Columb counter based on cell overvoltage trip.

I also think it is stupid way to reset Columb counter.

One way folks live with this is to temporarily lower the overvoltage trip setting after a normal full charge and cell balancing.