Growattt SPF 12000T DVM charging for a very long time after an outage

[jfunk]

I have a Growatt DVM12000T with 8 Discover EVL16A-A 6V batteries in a 48V configuration. There is no solar input, as I am only using it for backup.

Over the weekend I experienced 3 outages. The first 2 were only relatively short but the 3rd was a bit longer, and I plugged in a generator when the batteries got low. My Victon smart shunt said it was 20 or 30% at the time, I think. When I have my generator on, I set the Growatt max charge current to 30A since any more causes the generator to stall when my well pump kicks in. I manually set it back to 80A once grid power is restore.

Normally after an outage, the Growatt charges the batteries for a few hours and goes back to float charging, but this time it appears to be doing something else. It looked like it went to 100% in a few hours, according to the Growatt and the smart shunt, but kept charging. The batteries got quite warm. It did start to taper off for a while, even apparently stopping for a minute, but then went back up again.

It's been going for nearly a day and a half which is way longer than I've ever seen it and I am getting worried about it. I have attached the charging power and voltage charts.

Does anybody else recognize this behaviour?

 

[jfunk]

It took a few days, but it did slowly go back to normal.

 

[Vigo]

You say you have a shunt on this battery pack. I would think just looking at the ah/wh in and out info would tell you whether the pack SHOULD be being charged or not?

Overall it does look pretty normal given what's on those charts. It charged at ~6kw for a brief time then started tapering down to 2000w, and the next day down to 1000w then back to normal 'maintenance' charging, i guess i would say. I think this is probably normal for your pack and you just didn't know because you perhaps have never drawn it down to 45v before.

 

[jfunk]

Yeah, I did a bunch of searching and couldn't figure out whether it was normal or not. Both the shunt and the inverter reported the battery at 100% within a couple of hours even when it was pulling 2000W for charging. The room with the setup got incredibly warm and the batteries bulged a bit. I had a strut keeping them against the wall via threaded rod as a safety measure, and it pulled out of the wall. It all made me rather worried, but I guess it's normal.

I think I really need to get the shunt reporting to Home Assistant. I can only see it on my phone via bluetooth when I am next to it right now.

Thanks.

 

[Vigo]

Well, considering that info i would say it actually IS a problem! The batteries should not bulge under normal charging conditions.

The most likely scenario for why the batteries would get hot and bulge when the overall charging voltage is normal, is that one or more of the batteries or cells of batteries are severely out of balance, in this case lower voltage than the rest. Each cell in this string is ~2.1v nominally, so if one cell were to completely crap out, the pack goes from being ~48 to ~46v (more or less) and whatever charge settings you had before are now 2 volts too high the rest of the cells. You could also have several cells being low by a little bit, and the cumulative effect is the same.

The pics i googled of that battery made it look like you can't measure each cells voltage individually, but you can still compare each 6v battery in the string and make sure they are all very close to each other.

Unfortunately there is some chance that the batteries which did get overcharged, could have overpressurized and vented some gas out of a pressure limiting relief valve somewhere. With a sealed battery there is no way to put anything that comes out, back in. So there is a chance some permanent damage was done. But the starting point would just be compare all the battery voltages and see if one is lower than the others.

 

[jfunk]

According to the manufacturer, some amount of bulging is normal on larger batteries. https://discoverbattery.com/compone...gm-or-gel-batteries-bulge-or-appear-sucked-in

The shunt is is reporting a mid-point deviation of 0.10% which is unchanged from before the last event and when I measure the batteries directly the highest deviation is 0.02V.

I suspect these are probably OK.

I have also just ordered a Daygreen 12v-56v DC-DC charger to I can hook up my Bolt EV and keep the bank from getting so low in future outages.

 

[Vigo]

Sounds good. How much bulging is normal is probably highly contextual. I do think if all the heat in the room came from the batteries it might be a bad indicator of some kind, but if you also had inverter/charger in the room, i wouldn't be surprised at the heat. If you figure something like charging at 2kw and the charger is 90% efficient and the batteries themselves are 90% charge efficiency you've got almost 400w of heat going into the room for hours and hours. But if that heat was ONLY coming from the batteries i'd be paranoid about it. Hard to quantify 'walking into a hot room'.