Will not hold a charge

[Bandannaman76]

My battery bank has 16 150ah 12v AGM. I am running them in series for 4/48v sets of 4 batteries each and then parallel to make a 48v bank of all 16 batteries. What happens is the entire bank will charge up to capacity just fine, but will drop down rather quickly with little load on system, so charge/discharge cycles are frequent. I checked each battery and there were 5 that were at about 10.5v, but after a maintenance cycle and charge they recovered and all hold 12.5~13.5v each after I let them sit for about a week disconnected. I did a pile load test on each and all passed. I reconnected the entire bank and problem still exists.

Should I reconfigure to have 4/12v sets in parallel and then connect in series to make a 48v bank with all 16 batteries?

[chrisski]

My battery bank has 16 150ah 12v AGM. I am running them in series for 4/48v sets of 4 batteries each and then parallel to make a 48v bank of all 16 batteries. What happens is the entire bank will charge up to capacity just fine, but will drop down rather quickly with little load on system, so charge/discharge cycles are frequent. I checked each battery and there were 5 that were at about 10.5v, but after a maintenance cycle and charge they recovered and all hold 12.5~13.5v each after I let them sit for about a week disconnected. I did a pile load test on each and all passed. I reconnected the entire bank and problem still exists.

Should I reconfigure to have 4/12v sets in parallel and then connect in series to make a 48v bank with all 16 batteries?

Really sounds as if your batteries are toast. Won't matter how you configure them, they will still sag ;(

It would be odd for a battery to hold 13.5 volts given a time off a charger.
Screenshot 2022-12-14 163017.jpg
That is a Trojan AGM Battery which says should not exceed 12.84 resting volts, after a while off a charger. It will read high, perhaps 13.5, that for a bit immediately after being removed from. Not a week though. Chart also says a battery at 11.64 volts is at 0% SOC / 100% DOD, and yours was lower, also pointing to a dead battery, or at least 5.

There's nothing I know of to revitalize an AGM battery like an equalization phase is for batteries.

Depending on what type of batteries they are, you may be able to have one or two tested at an auto parts store. That is not a standard first step, but there's enough pointing to saying the batteries are at the end of their life cycle that a load test may show this to be true.

Best you can hope for is to individually test the batteries to see the best ones and maybe there are 4, 8 or 12 that can be put in parallel for some usage, but I don't see any way to tie all 16 in. Perhaps two sets paralleled will work better than including the five dead batteries.

[Bandannaman76]

Thank you for the reply! Looking for suggestions. I have 16 250w panels ran in series/parallel to make 4 1000w banks for the arrays. I have a 12kw inverter with a epever 5000w controller. If I need to get new batteries, would it be best to get 48v units and gang them together, or go like I did initially with 12v? Any suggestions for a good source to purchase from? I've attached some pics of my system.

[azdave]

I did a pile load test on each and all passed.

That makes no sense. If they all passed individual load tests then there should be no issues when they are combined. There must be some bad batteries in the bunch and your mentioning five of them that needed maintenance cycled and charged to bring them up to voltage is your first hint that you've probably found them already.