Saw on of my batteries get down to 2.6v recently

[jedics]

My 24v 2p 8s system got down to 20% soc recently and saw one of my battery pairs get down to 2.6v when the rest were at 3.2v which was probably caused by my system getting run to zero 3 times before I got things under control after I first installed it. Its been running fine for 6 years of daily use and because I have almost never run the batteries that low with my usage I never saw this until now. Obviously I am loosing some capacity which I dont care about that much but I would like to be able to get my moneys worth for another 4 years at least before upgrading. 

What would you guys do, take out the battery pair and find if one or both batteries is damaged and replace one or both, new or second hand?

Or just carry on as it is seeing I havent had a problem with the damaged battery in 6 years? (not ideal in winter to never be able to drain them below about 30% safely)

[Dacian Todea (electrodacus)]

I will like to see a photo of the SBMS screen. 3.2V can be just around 10 to 15% so the other cells where likely also almost discharged.

The issue is that your one cell that got to 2.6V should never have been there if the SBMS has control ON/OFF over all Loads and settings are are default with switching OFF all loads when any cell gets to 2.8V. Since it gets to 2.8V under load usually the voltage will recover to around 3V and will stay there with just the SBMS as load until the next day when sun should recharge the battery.

Likely the battery is fine and can last a few more years. You just need to setup the system so that no cell can get below 2.8V  (2.6V is not an issue but below 2.5V is a problem).

[jedics]

When I built the system the Giandel inverter I bought had the on/off switch that you must push and hold so the sbms couldnt control it which is why I had the draining to zero because the inverter stupidly has a 21v cut off that you cant change. Everything is charged up now so a pic probably wont help you now but here you go:

[Dave McCampbell]

We also have Giandel inverters (3 ea) and a large (900ahr) LFP bank.  But we only turn one at a time on-off while monitoring our display for short periods of time for cooking, so there is little danger of significantly depleting our bank.  Couldn't you control yours using a big relay on your load buss (as we do) or in the DC cable from there to your inverter?  Either set to execute a default LVD at 2.8vpc.   Your cell delta at 130 mv is large compared to ours at around 7mv at charge termination.  What is the cell delta at near LVD?  If you wanted to reduce that, doing a cell capacity test would help determine if you have low capacity cells and which ones they are.  Another easier thing you could do is use a large active cell balancer to see if you can reduce the cell delta to a more reasonable figure.  Also, have you checked your cable and internal cell resistances lately?

[Dave McCampbell]

Dacian just reminded me that cell voltages, while being balanced, are not accurate.  Therefore the 130mv delta I mentioned above should be disregarded.  The real difference is probably much less and within acceptable limits.  So you might not want to do anything more with balancing.  Instead concentrate your efforts on getting your SBMS0 On-Off control over all loads and charging devices so there is no possibility of a cell getting to a dangerous voltage.