[EG4 + Schneider Conext] Help troubleshooting Alarm & Warnings Events

[Old_Skewler]

This is for an off-grid 15 KWh 48V battery bank (3x EG4s) connected to a Schneider 4048 Inverter in split-phase. The system has been running well for over a year. I would like to get feedback on how you'd deal with these warnings if you were running this system. I have a total of (3) alarm and warnings:




Here is the battery BMS information as read by the inverter. The most recent event happened when battery around 50% and 48V. Very small load on system, less than 200W.



And here is the battery setting screen with a 47V LBCO:




1. DC Under Voltage: i suspect the LBCO setting is too high and needs to be lowered? I don't remember where I got this 47V setting from (@SignatureSolarTech or Schneider).

2. AC Output Under Voltage: is there a way in InsightHome to access historical data and verify the AC load at the moment this event happened? Anything else I should be looking into regarding this event?

3. DC Over Voltage: this must've been the Inverter when in charging mode. I currently have the following charger settings.

Bulk/Boost Voltage Set Point: 57.6V
Bulk Termination Voltage: 56.8V
Absorption Voltage Set Point: 56.8V

I am not sure whether i should be looking into updating the charging numbers above OR increasing the HBCO. Any thoughts?

Thanks in advance for any feedback!

 

[hwy17]

Is 21-33v in the error description just an SW bug cause it was written for the 24v? You have a 48v SW right?

For DC undervoltage I would inspect your cables and connections. A loose connection could cause voltage drop.

 

[BentleyJ]

1. DC Under Voltage: i suspect the LBCO setting is too high and needs to be lowered? I don't remember where I got this 47V setting from (@SignatureSolarTech or Schneider).

NO, your batteries need to be fully charged. At 48.8V the EG4's are under 10% SoC. The inverter is doing its job by stopping discharging the battery and displaying the warning code

2. AC Output Under Voltage: is there a way in InsightHome to access historical data and verify the AC load at the moment this event happened? Anything else I should be looking into regarding this event?

There should be a graph of Watts or Amps output going back a least a few days??

3. DC Over Voltage: this must've been the Inverter when in charging mode. I currently have the following charger settings.

Bulk/Boost Voltage Set Point: 57.6V
Bulk Termination Voltage: 56.8V
Absorption Voltage Set Point: 56.8V

I am not sure whether i should be looking into updating the charging numbers above OR increasing the HBCO. Any thoughts?

Thanks in advance for any feedback!

If the battery cells are out of balance and the BMS cuts out on HV, there is no were for the current to go and the inverter bus voltage will spike then shut down and display the OV alarm.

EDIT: The BMS data, last item on bottom right, is showing a Warning for Cell Voltage difference Too High. The cells need to be fully charged AND balanced.

Recommend reducing maximum charge current to 75A. That would correspond to 0.25C charge rate for the 3 batteries.

 

[Old_Skewler]

NO, your batteries need to be fully charged. At 48.8V the EG4's are under 10% SoC. The inverter is doing its job by stopping discharging the battery and displaying the warning code

How could the battery be at 10% when BMS SOC tells me it is about 50%? Only thing is, 15 mins after I post this thread and the system goes down. I connect to the EG4s and they are now reading 0% SOC.


How is this possible? I've been relying on the BMS SOC info and that showed 48% a bit over 30 mins ago. Now all three packs show 0%.

If the battery cells are out of balance and the BMS cuts out on HV, there is no were for the current to go and the inverter bus voltage will spike then shut down and display the OV alarm.

All three packs were balanced a the beginning of the season, back in May/June.

EDIT: The BMS data, last item on bottom right, is showing a Warning for Cell Voltage difference Too High. The cells need to be fully charged AND balanced.

Okay, understood. How often should this happen though? They are less than 15 cycles and were topped individually a few months ago.

Recommend reducing maximum charge current to 75A. That would correspond to 0.25C charge rate for the 3 batteries.

The actual Max Charge Current the inverter can do is 45A, so about 15A a battery. The charger let me set 100% of charge and the inverter tops of at 45A.

 

[Old_Skewler]

Is 21-33v in the error description just an SW bug cause it was written for the 24v? You have a 48v SW right?

You are correct, it is 48V SW. It must be a typo in the description.

For DC undervoltage I would inspect your cables and connections. A loose connection could cause voltage drop.

I checked all the DC connections, no issues.

But I am puzzled that the system turned off on low voltage and the EG4s were reading out 47% SOC and then instantly went to 0%.

 

[JBertok]

Do you have a PC set up with a serial port connection to the BMS in the EG4 batteries? A battery BMS that is faulting into disconnect can cause both the DC undervolt and overvolt errors you're getting. As was mentioned, you might have a few cells out of spec that the BMS isn't able to compensate for with balance. One cell too high or too low will shut down the whole BMS.
There could also be an imbalance between the batteries themselves, and that could result in one unit seeing a greater duty cycle than the others, and that could lead to BMS disconnection if, for example, one unit is discharging more deeply or overcharging.

The events in the logs are just effects but I'd be looking at the battery health as a cause.

What's the DC cabling to the batteries look like? Particularly interested if you're using a cascade or a busbar strategy paralleling the batteries, the lengths and gauges of cable, and how/where the main DC feeds connect to the batteries.

EDIT: I see you posted BMS data as I was typing this. Looking at it now....

 

[JBertok]

Each battery pack has a low cell. That's likely what's shutting down the BMS in each of them.
Pack 1 - Cell 8
Pack 2 - Cell 9
Pack 3 - Cell 6

 

[Old_Skewler]

Each battery pack has a low cell. That's likely what's shutting down the BMS in each of them.
Pack 1 - Cell 8
Pack 2 - Cell 9
Pack 3 - Cell 6

Max V diff is about 0.628V. I don't know if this is too high or normal.

But how does this explain the SOC going from 48% to 0%? The Battery BMS is currently reading 0% SOC.

Do you have a PC set up with a serial port connection to the BMS in the EG4 batteries?

EDIT: I see you posted BMS data as I was typing this. Looking at it now....

Yes, the last screenshot is straight from the EG4 BMS.

 

[BentleyJ]

Max V diff is about 0.628V. I don't know if this is too high or normal.

628mV is a VERY BIG cell imbalance. However, this is happening at the Bottom so either the cells are highly unbalanced OR some of the cells have lost more capacity than the the others OR both. This probably deserves more explanation. Each cell is listed as 100Ah and we just have to trust they are all indeed about that capacity within a couple of Ah of each other. Since the batteries are relatively new and the State of Health is 99%, one would not immediately assume a bad cell but rather a poor state of balance. Given the large cell Voltage delta, its going to take a long time for the BMS to correct. Long time as in maybe a few days with the batteries float charging at about 55V. Check the cell voltages on the BMS after the first day and verify the cells are getting balanced. Ultimately they should be within 10 to 15mV.

If the SoC value from the BMS is not accurate you may need to get something like a Victron Smart Shunt and monitor it so the batteries never go below 20% SoC.

But how does this explain the SOC going from 48% to 0%? The Battery BMS is currently reading 0% SOC.

The explanation is the BMS malfunctioned and indicated an incorrect SoC. Based on all the other cell voltage info from the BMS as well as the warnings from the inverter the batteries are completely discharged.

 

[Old_Skewler]

628mV is a VERY BIG cell imbalance. However, this is happening at the Bottom so either the cells are highly unbalanced OR some of the cells have lost more capacity than the the others OR both.

Thanks for the clarification. And good to know acceptable values are 10 to 15 mV.

This probably deserves more explanation.

I already reached out to the vendor, (SS) and I will follow up if the issue persists.

Given the large cell Voltage delta, its going to take a long time for the BMS to correct. Long time as in maybe a few days with the batteries float charging at about 55V. Check the cell voltages on the BMS after the first day and verify the cells are getting balanced. Ultimately they should be within 10 to 15mV.

I have a 48V 8A charger that runs at 58.4V constant charging voltage. I do not have a 55V charger. Can I still make this balancing with it? And how would I know when they are topped off? From going over this balancing early in the season I realized it is a very very fine line between topping off the batteries and triggering them on over voltage. Any thoughts on this?

 

[BentleyJ]

From going over this balancing early in the season I realized it is a very very fine line between topping off the batteries and triggering them on over voltage. Any thoughts on this?

58.4V on a 16 cell battery is an average of 3.65V per cell. If the cells are already balanced this voltage will work fine since all the cells will reach full charge at the same time and the current will rapidly decrease indicating the full charge has been achieved.

In this case, your cells are most definitely not balanced so the charging voltage has to be lower otherwise the first cell to reach full charge will rapidly run up to the HVCO limit on the BMS causing the BMS to disconnect from the battery. Charging and balancing will stop.

Two Solutions:
1) An adjustable Constant Voltage/Amperage bench top power supply can be used to charge the battery and the voltage can be ramped up slowly over a few days to prevent the BMS from triggering on HVCO. This is a slow process because it depends on the balancing capacity of the BMS built into the battery.
2) A stand alone active cell balancer. JK BMS sells a 5A and a 10A Bluetooth balancer with connectors, wires and alligator clips. I have one of each size. They are a bit expensive depending on your budget and would require taking the covers off the batteries to gain access to the cell terminals. On the upside, this would be a much faster way to balance the cells.

Based on the number of Forum members that eventually have this cell imbalance problem, maybe I should start business renting out my 10A balancer!!

 

[Old_Skewler]

58.4V on a 16 cell battery is an average of 3.65V per cell. If the cells are already balanced this voltage will work fine since all the cells will reach full charge at the same time and the current will rapidly decrease indicating the full charge has been achieved.

In this case, your cells are most definitely not balanced so the charging voltage has to be lower otherwise the first cell to reach full charge will rapidly run up to the HVCO limit on the BMS causing the BMS to disconnect from the battery. Charging and balancing will stop.

What a conundrum! I have found a 48V charger with 54.6V @ 8A output. Could this work and bring the cells to a balance then I'd top it off with my 58.4V charger?

It only costs about $70.

 

[BentleyJ]

Yes, that voltage should work well. 54.6V is 3.41V per cell.

 

[JBertok]

Effectively, a single cell at low-voltage cutoff is considered criteria for the BMS to be the discharge floor for the entire pack. An entire series string of batteries will be limited by the capacity of the lowest cell in the series. It's especially precise with any lithium chemistries since only a very slight voltage deviation represents a huge change in state of charge.

Are EG4's firmware upgradable? Maybe newer firmware would address the balancing scheme. Allowing the batteries to spend more time at full float voltage and an apparent 100% SoC should help too, over time. The BMS will report 0% SoC if just one cell reaches low-voltage cutoff, and 100% SoC if just one cell reaches high-voltage cutoff.

 

[JoeHam]

Thanks for the clarification. And good to know acceptable values are 10 to 15 mV.



I have a 48V 8A charger that runs at 58.4V constant charging voltage. Any thoughts on this?

One comment and two thoughts :

I believe your battery bank is 300 Ah not 100 Ah as you have set in your configuration. Please confirm.

First, you are charging at 150 Amps or 0.5 C and that is way too fast IMHO and may aggravate your runner cells.

Second, using your 8 Amp charger would be charging at a rate just under 0.03 C and may get the cells closer to balance without the runner taking off at such a slow charge rate.

I would give your charger a try and watch the individual cell voltages to see when the runner takes off.

Since you already have it I would give it a try.

 

[Old_Skewler]

I believe your battery bank is 300 Ah not 100 Ah as you have set in your configuration. Please confirm.

This setting won't stick and reverts back to 100Ah. Not sure why.

First, you are charging at 150 Amps or 0.5 C and that is way too fast IMHO and may aggravate your runner cells.

The battery BMS charging rate comes from the battery itself and it is not a variable setting. I believe it is only informational. Regardless, my charger is limited to 45A which divided for 3 batteries is equal to 15A a battery.

Second, using your 8 Amp charger would be charging at a rate just under 0.03 C and may get the cells closer to balance without the runner taking off at such a slow charge rate.

I would give your charger a try and watch the individual cell voltages to see when the runner takes off.

Since you already have it I would give it a try.

Yes, I will try it as soon as I am able to haul all packs to the urban environment.

 

[Old_Skewler]

Effectively, a single cell at low-voltage cutoff is considered criteria for the BMS to be the discharge floor for the entire pack. An entire series string of batteries will be limited by the capacity of the lowest cell in the series. It's especially precise with any lithium chemistries since only a very slight voltage deviation represents a huge change in state of charge.

Thanks for the info!

Are EG4's firmware upgradable? Maybe newer firmware would address the balancing scheme. Allowing the batteries to spend more time at full float voltage and an apparent 100% SoC should help too, over time. The BMS will report 0% SoC if just one cell reaches low-voltage cutoff, and 100% SoC if just one cell reaches high-voltage cutoff.

They are, believe my packs have the latest firmware, as of now.

 

[Sanwizard]

How could the battery be at 10% when BMS SOC tells me it is about 50%? Only thing is, 15 mins after I post this thread and the system goes down. I connect to the EG4s and they are now reading 0% SOC.
View attachment 166265

How is this possible? I've been relying on the BMS SOC info and that showed 48% a bit over 30 mins ago. Now all three packs show 0%.



All three packs were balanced a the beginning of the season, back in May/June.


Okay, understood. How often should this happen though? They are less than 15 cycles and were topped individually a few months ago.


The actual Max Charge Current the inverter can do is 45A, so about 15A a battery. The charger let me set 100% of charge and the inverter tops of at 45A.

Your cells are very unbalanced. Cell 8 is just about at cutoff.

 

[Old_Skewler]

Your cells are very unbalanced. Cell 8 is just about at cutoff.

I believe you are correct. I was relying on battery BMS SOC and suddenly got the carpet pull from underneath my feet. Working on slowly charging and balancing the batteries.

 

[n2aws]

I'm not sure if this was mentioned, I skimmed the thread quickly. I'm not sure if you are using Lifepower4's or LL's, but there are updated firmwares for both.

I know the latest "schnieder firmware" for my EG4 LifePower4 batteries eliminated some cell imbalance issues for me, but I had to keep the batteries charged up long enough to allow them to top balance. Your cell imbalances are massive, so I'd suspect you need these batteries to be able to top balance for a significant period of time.

At a guess, I'd imagine your batteries don't see 100% SOC all that often? You likely want to let them hit 100% occasionally, so the BMS can do the top balancing on a somewhat regular basis. Maybe monthly? Otherwise, when your "lowest" cell it's the undervoltage cutout, the whole battery shuts down, and whatever energy is in the remaining cells is essentially unusable (ie, you're likely seeing MUCH lower than 100ah from each one)

I'd get the firmwares updated, and get those batteries to 100% for a while until the cell voltages are really close to each other.