Runaway cells, sudden overvoltage

[2potatoes]

Thank you for the abundance of knowledge you share on this forum. I'll be reading for months. I only wish I'd found you before I got in so far over my head. I thought I knew what I was doing. I don't.

I have 4 brand new 100ah 12v weize lifepo4 batteries connected in series to a 48v renogy 3500 watt inverter/charger. The charger happily charges the batteries all day from solar panels. In the evening, a battery BMS beeps for overvoltage and since the inverter is 48v only, the system shuts down. The battery has over charged to an astonishing 36 volts. As per the manufacturer's instructions, I disconnected the battery and the BMS sheds voltage back down to 13.6. i wire the batteries in parallel to rebalance. Then the cycle repeats the next day with a different battery. If I follow the manufacturer's advice and reconnect immediately after the BMS resets, the battery over charges within minutes of powering up the inverter/charger.

After reading awhile here i have learned that I know nothing about batteries and it will be sooner time before I know enough to fix this problem. Maybe the cells in the batteries are out of balance?

I would be very grateful for any advice on how to proceed. We are entirely off grid and have been without power for a few weeks now. I'd love to be able to use at least some of the power the panels are putting out, even if I can't get the battery bank problem fully resolved.

2potatoes ( well 2 varieties anyway)

 

[MrThisIsME]

Few questions
1. How is 36v over charging a 48v battery?
2. Assuming the Weize batteries are self contained, where is this BMS you mention?

No current BMS (worth its weight) will allow a battery to charge to 36v.
Are you sure you have connected correctly and understand 4 12v batteries in series?
What voltage is the Renogy telling you?

 

[2potatoes]

Thank you for your consideration! The batteries are each 12 volts nominal; the BMS is built into each of the 4 batteries. It does seem awfully strange that the BMS would let the voltage get so high. I am forced to consider that the BMS is perhaps not worth it's weight. I am pretty sure that they are wired correctly in series. The cables are all the same length except those running from the batteries to the charger/inverter. I have no battery bus bar. The renogy reads around 51 to 52 until the battery cuts off, then it just gives an incorrect voltage error.

 

[SpongeboB Sinewave]

Are these batteries made to be able to put 4 in series to make 48V nominal ?

The BMS's have a maximum voltage they can say, turn off at with other batteries in series.

36V + 12V = 48V so maybe there is something wrong with only one of those BMS/batteries ?

boB

 

[2potatoes]

They do claim to be able to be stacked in series up to 48 volts max. It is possible that one is bad. I had the overvolted battery replaced by the manufacturer, but the problem recurred right away with the new battery, so either the overvolted battery wasn't the problem battery, or else the battery was fine all along?

 

[Supervstech]

You have learned why it is not recommended to put 12V lithium batteries in series.
It can be done with a 12V battery balancer, but it is tricky.
#1, the batteries all need to be balanced together. Or individually charged up to 14.4V
Most lithium batts won’t handle over 13.8V without a cell going into BMS disconnect… if yours are like that, you need to set your charger settings so it never outputs more than 4 times that amount. 55.2V max will fully charge 4 lithium batteries in time, but it takes a while.
A higher setting will drop out a weak BMS in one of the series and your inverter will slam the rest with more voltage than they can handle before the controller stops charging.

 

[2potatoes]

In retrospect that very much makes sense. I think that is most likely what is happening.

I assume a battery balancer is a separate piece of equipment?

The manufacturer says the batteries can handle up to 14.6, which would be like 58.4, but I've never had them that high. I can't get them there individually without grid power and a 12 volt charger. When I try with the renogy I just get the high voltage issue.

Perhaps I should just cut my losses and either switch to a 12v system with these batteries or else go back to lead acid? I definitely value reliability and can't afford to have things be finicky.

 

[GXMnow]

I agree this all looks like a balance issue. The BMS is doing it's job to protect the cells by going open circuit when a cell goes over voltage. That will then allow the external terminals of the battery to go very high, but the cells inside are each at or below 3.65 volts.

You have 2 separate but similar problems all at once. The 4 cells inside each 12 volt battery are likely now out of balance. AND the the 4 12 volt batteries are out of balance from each other. Do these batteries have BlueTooth monitoring? Can you see each cell voltage?

I think your best bet is going to be placing the 4 12 volt batteries in parallel, and then very slowly charging them all up to 13.8 volts, and then holding them there for a full day or more. 13.8 is 3.45 per cell. It is into the upper knee, so it should be triggering the internal balancers to get all the cells matched. But the internal balancers could be a very small current, maybe just 30 to 100 milliamps. But the cells are 100 amp hours. So it can take a very long time for them to pull those cells together. If you can monitor the cell voltages, you would be able to see when they achieve a decent balance. It is quite possible that even going up to 13.8 volts that a cell could run and trigger a cell over voltage disconnect. That should be at 3.65 volts per cell. If one cell is out of balance and is hitting that, the other 3 in balance would be at 3.383 volts. Holding the battery terminals at 13.8 will then keep trying to balance the cells. When one battery goes into a cell over voltage, it's balancer should be applying a load on the high cell until it drops enough for the BMS to turn back on. It's not ideal, but if you can't connect to each cell's terminals, it is the best you can do to get it to balance.

Once all the cells are in a fairly close "Top Balance", then you can wire them back in series for your 48 volt system. A 4 x "12 volt" balancer can help keep them together, but with LFP cells, they should hold a decent top balance for a long time. You don't want a balancer to pull on the voltages when the cells are not at the top. That will actually hurt the top balance. The cells are very likely not going to be perfectly matched in capacity. As you discharge the cells, the weakest cell is going to drop voltage the fastest. When that cell hits it's low cel voltage limit, that is your usable capacity for the entire string. At that point, they cells will look way out of balance, but don't worry about that. Since they are all in series, you took the same amp hours out of every cell. As they charge back up, they all get the same amp hours, and should all hit full charge at the same time again. They will come back into balance near full charge again. That is the best you can do.

 

[2potatoes]

I agree this all looks like a balance issue. The BMS is doing it's job to protect the cells by going open circuit when a cell goes over voltage. That will then allow the external terminals of the battery to go very high, but the cells inside are each at or below 3.65 volts.

You have 2 separate but similar problems all at once. The 4 cells inside each 12 volt battery are likely now out of balance. AND the the 4 12 volt batteries are out of balance from each other. Do these batteries have BlueTooth monitoring? Can you see each cell voltage?

I think your best bet is going to be placing the 4 12 volt batteries in parallel, and then very slowly charging them all up to 13.8 volts, and then holding them there for a full day or more. 13.8 is 3.45 per cell. It is into the upper knee, so it should be triggering the internal balancers to get all the cells matched. But the internal balancers could be a very small current, maybe just 30 to 100 milliamps. But the cells are 100 amp hours. So it can take a very long time for them to pull those cells together. If you can monitor the cell voltages, you would be able to see when they achieve a decent balance. It is quite possible that even going up to 13.8 volts that a cell could run and trigger a cell over voltage disconnect. That should be at 3.65 volts per cell. If one cell is out of balance and is hitting that, the other 3 in balance would be at 3.383 volts. Holding the battery terminals at 13.8 will then keep trying to balance the cells. When one battery goes into a cell over voltage, it's balancer should be applying a load on the high cell until it drops enough for the BMS to turn back on. It's not ideal, but if you can't connect to each cell's terminals, it is the best you can do to get it to balance.

Once all the cells are in a fairly close "Top Balance", then you can wire them back in series for your 48 volt system. A 4 x "12 volt" balancer can help keep them together, but with LFP cells, they should hold a decent top balance for a long time. You don't want a balancer to pull on the voltages when the cells are not at the top. That will actually hurt the top balance. The cells are very likely not going to be perfectly matched in capacity. As you discharge the cells, the weakest cell is going to drop voltage the fastest. When that cell hits it's low cel voltage limit, that is your usable capacity for the entire string. At that point, they cells will look way out of balance, but don't worry about that. Since they are all in series, you took the same amp hours out of every cell. As they charge back up, they all get the same amp hours, and should all hit full charge at the same time again. They will come back into balance near full charge again. That is the best you can do.

That's a lot to consider; thank you very much for your help. It seems like the more I learn here the more questions I have.

I have no way to check individual cell voltage short of opening the case. I haven't tried that yet out of concerns for the warranty. If things get any more dire I'll do it anyway.

Would it be possible to balance as per your instructions straight from the solar panels if I were to buy a small pv to 12vcd charger controller, with no load on the batteries? That would be preferable to getting something like the Renogy 20A AC-to-DC Charger, loading up the battery bank, and hauling it off on search of grid power. Especially if I would have to do it regularly.

Assuming I can get each cell balanced and the batteries balanced together, about how long do you think I could reasonably expect to go before needing to balance from the grid again? Weeks? Months? Years?

Do I understand correctly that I could have avoided this entire problem by buying a single 48v rack mounted battery instead? Or does that have its own pitfalls?

Finally, you mentioned that the voltage at the terminals could be higher than the total cell voltage? I really don't understand that at all. Is there any reading material you could refer me to that might help me understand how that works?

Thank you again, everyone, for your help.

 

[GXMnow]

Once they get into a decent top balance, they should hold for a very long time. Th internal BMS units will keep the 4 cells in each battery balanced, but if the batteries start drifting apart, then maybe a 4 x 12 volt balancer may be needed.

 

[2potatoes]

Hello folks. I bought a 12v 20a ac charger, hauled the batteries to town, charged them, equalized for 24 hours, then brought them back. Although I can't check the individual cells, the batteries as a whole were in balance when I connected them back up.

3 hours in use, I have 13.89, 13.82, 13.68, and 13.62.

As far as I can tell, I can either buy a $50 balancer, sell the batteries and get an eg4 48v rack mounted battery, or abandon the 48v idea and buy 12v inverter, charger, and wire the batteries in parallel.

Is there anything I'm missing? Thank you again for your help

 

[100 Proof]

Are you taking your main positive and main negative from your battery bank to the inverter from the same battery? That will definitely cause battery imbalance.

They should be wired per the attached drawing.

 

[2potatoes]

They are wired exactly as in the diagram. I'm now past max charge for the day and into evening with about .8v difference between the highest and lowest batteries. I think I'll try a balancer before making any more drastic decisions.

 

[TacomaJoe]

I think what's happening is that one of the batteries hits a high voltage (cell or battery) and opens the charging MOSFETs. The inverter will complete a series circuit. Batt1, Batt2, Batt3, Batt4, Inverter and back to Batt1. The 36V reading is actually the sum of the 3 batteries that didn't cut out. The BMS is doing its job. I would lower the charging voltage to reduce the chance of a cell/battery OV condition.

 

[Berseker]

Thank you for the abundance of knowledge you share on this forum. I'll be reading for months. I only wish I'd found you before I got in so far over my head. I thought I knew what I was doing. I don't.

I have 4 brand new 100ah 12v weize lifepo4 batteries connected in series to a 48v renogy 3500 watt inverter/charger. The charger happily charges the batteries all day from solar panels. In the evening, a battery BMS beeps for overvoltage and since the inverter is 48v only, the system shuts down. The battery has over charged to an astonishing 36 volts. As per the manufacturer's instructions, I disconnected the battery and the BMS sheds voltage back down to 13.6. i wire the batteries in parallel to rebalance. Then the cycle repeats the next day with a different battery. If I follow the manufacturer's advice and reconnect immediately after the BMS resets, the battery over charges within minutes of powering up the inverter/charger.

After reading awhile here i have learned that I know nothing about batteries and it will be sooner time before I know enough to fix this problem. Maybe the cells in the batteries are out of balance?

I would be very grateful for any advice on how to proceed. We are entirely off grid and have been without power for a few weeks now. I'd love to be able to use at least some of the power the panels are putting out, even if I can't get the battery bank problem fully resolved.

2potatoes ( well 2 varieties anyway)

AM assuming the battery manufacturer told you, you can wire the batteries in series??. what bms type is inside?. the avg lithium battery doesn't like being wired in series. until recently that jbd claimed to release a bms that has that capability, though they are yet to release the science behind this claim.

your best and easiest option based on my perceived evaluation of your knowledge level is to get a 12v inverter and put them all in parallel.

getting a 48v battery bank is my recommended solution, but would be more expensivbe