Title summarizes what's going on, my 2x parallel battery banks are discharging unevenly. Here's my setup:
Sol-Ark 15k Inverter
2x16s1 LifePO4 battery banks (EVE LF304), Overkill Solar 48v 100A BMS
Since there's no way to get the BMS to communicate with the inverter - or each other - this is an open loop system.
A little more context: I assembled these batteries at the beginning of this year when I brought my system online for the first time. I bought all the cells from a reputable supplier and they were all new. I was a bit lazy and decided to let the BMS do the balancing for me instead of doing a top balance myself - and to be fair, until recently it seems everything was working fine, and based on the data I've gathered from many charge/discharge cycles the capacity seems to be just fine. So this doesn't seem to be some issue with the way things were set up initially as it was all pretty smooth at the start. The Sol-Ark was remarkably accurate at "guessing" the SOC and was usually just about spot on with what the BMSs were reading. The only odd thing I did notice is that one battery always seems to discharge/charge at a higher current than the other. Although, given that this didn't cause any problems, I didn't pay too much attention to it. I have the inverter set to draw a maximum of 185A from the banks, which if roughly divided in half, should not trip the overcurrent protection.
Recently however I've been running into some issues that I wasn't seeing before. I initially noticed that the SOC indicated by my inverter would suddenly drop from whatever it was reading (somewhere above 50%, no consistent number) to 0% (or close to it).
First I did some digging and (thanks to some help from some folks here in another thread I posted), I identified a loose connection on one of the voltage leads that was causing one of the BMSs to constantly flip between over/undervoltage shutoff due to the faulty readings. I fixed this with a few turns of the wrench, thinking that I identified the problem.
However, this problem is still happening on a pretty frequent basis - and it's pretty annoying because once it happens, the inverter will not 'correct' the SOC until the next day - meaning if we have power bumps at night (which has been happening a lot), suddenly I don't have a battery backup because the inverter thinks the battery is drained.
tl;dr here is basically what I think is happening:
1. A large load starts drawing a lot of power in the evening or when there isn't a lot of solar power
2. One of the battery banks discharges at a higher enough rate than the other that it trips the overcurrent protection and cuts off that battery
3. as the inverter is now trying to power the same load from only one bank, it also overcurrents the second battery - on the inverter side this registers as a very sharp voltage drop
4. The inverter sees this and stops trying to use the batteries, and then proceed to set the SOC to 0%, thinking (understandably) that the batteries are depleted.
5. This SOC mismatch is not corrected until the next day, and I'm not exactly sure what causes the inverter to 're check' the SOC, but at some point it figures it out and the SOC will jump back up.
And now I'm seeing the benefit of a closed loop system...
Right now I see my options as:
1. Throw more battery banks at the problem so that it's just less likely to happen (as I was planning to expand anyway).
2. Replace my BMSs with ones that can handle 200A discharge (don't really want to do this as overall I like these ones and they come recommended by Will).
3. Take the batteries apart and top balance them, check for bad cells and re-shuffle them into the banks (would this even help though?).
4. Give up and buy prebuilt batteries that can operate in a closed loop with my inverter.
Can anyone recommend some troubleshooting steps to figure this out?
Sol-Ark 15k Inverter
2x16s1 LifePO4 battery banks (EVE LF304), Overkill Solar 48v 100A BMS
Since there's no way to get the BMS to communicate with the inverter - or each other - this is an open loop system.
A little more context: I assembled these batteries at the beginning of this year when I brought my system online for the first time. I bought all the cells from a reputable supplier and they were all new. I was a bit lazy and decided to let the BMS do the balancing for me instead of doing a top balance myself - and to be fair, until recently it seems everything was working fine, and based on the data I've gathered from many charge/discharge cycles the capacity seems to be just fine. So this doesn't seem to be some issue with the way things were set up initially as it was all pretty smooth at the start. The Sol-Ark was remarkably accurate at "guessing" the SOC and was usually just about spot on with what the BMSs were reading. The only odd thing I did notice is that one battery always seems to discharge/charge at a higher current than the other. Although, given that this didn't cause any problems, I didn't pay too much attention to it. I have the inverter set to draw a maximum of 185A from the banks, which if roughly divided in half, should not trip the overcurrent protection.
Recently however I've been running into some issues that I wasn't seeing before. I initially noticed that the SOC indicated by my inverter would suddenly drop from whatever it was reading (somewhere above 50%, no consistent number) to 0% (or close to it).
First I did some digging and (thanks to some help from some folks here in another thread I posted), I identified a loose connection on one of the voltage leads that was causing one of the BMSs to constantly flip between over/undervoltage shutoff due to the faulty readings. I fixed this with a few turns of the wrench, thinking that I identified the problem.
However, this problem is still happening on a pretty frequent basis - and it's pretty annoying because once it happens, the inverter will not 'correct' the SOC until the next day - meaning if we have power bumps at night (which has been happening a lot), suddenly I don't have a battery backup because the inverter thinks the battery is drained.
tl;dr here is basically what I think is happening:
1. A large load starts drawing a lot of power in the evening or when there isn't a lot of solar power
2. One of the battery banks discharges at a higher enough rate than the other that it trips the overcurrent protection and cuts off that battery
3. as the inverter is now trying to power the same load from only one bank, it also overcurrents the second battery - on the inverter side this registers as a very sharp voltage drop
4. The inverter sees this and stops trying to use the batteries, and then proceed to set the SOC to 0%, thinking (understandably) that the batteries are depleted.
5. This SOC mismatch is not corrected until the next day, and I'm not exactly sure what causes the inverter to 're check' the SOC, but at some point it figures it out and the SOC will jump back up.
And now I'm seeing the benefit of a closed loop system...
Right now I see my options as:
1. Throw more battery banks at the problem so that it's just less likely to happen (as I was planning to expand anyway).
2. Replace my BMSs with ones that can handle 200A discharge (don't really want to do this as overall I like these ones and they come recommended by Will).
3. Take the batteries apart and top balance them, check for bad cells and re-shuffle them into the banks (would this even help though?).
4. Give up and buy prebuilt batteries that can operate in a closed loop with my inverter.
Can anyone recommend some troubleshooting steps to figure this out?
