low battery discharge issue

I've been having some battery issues and have been through a couple of calibrations in the last few weeks, latest last night, and now I'm getting an odd issue where the battery only seems to want to discharge at a very low rate.

Battery is reporting as 83% full (of 9.5kWh), but is only supplying 110W to the house, the rest is drawing from the grid, and I'm unable to get it to go any higher for more than a few seconds.

The only switch that seems to make any difference is the "Battery Reserve % Limit", if I set it to a high value then change it back to 4% the battery picks up the full house load for a short time.

Battery discharge is set to 3600W and I've tried restarting & resetting everything I can get my hands on, with no effect. Any ideas?

set it to eco?

Yep, toggling eco off/on was the first thing I tried but it does not appear to get me above 110W when it comes back on.

chriswright What sort of system (hybrid, ..?) and what firmware is it running?

It's unlikely, but .... a few weeks back, my inverter rcd tripped, which the inverter took to be a power cut. So it was using the battery only to power itself, and the rest of the house was running from grid. Flicking through the portal data, I can't immediately spot when that might have been, but I think the symptom was similar - reporting about 100W discharge. But I can't think why that would then behave in the way you describe when adjusting the reserve.

if you force it to charge from the grid, does anything happen ?

Its a GIV-HY3.6 and a gen2 battery, inverter firmware is D0.309-A0.309 and battery is on 3017.

I just forced the battery reserve to 95% and turned off eco, that resulted in a slow charge (_290w) for a few mins then I dropped the reserve back to 4% and turned eco back on. And now everything appears to be working?!

What the hell??!

Still not perfect, if I boil a kettle I don't appear to be able to draw more than ₁.8kw from the battery, with the grid making up the rest.

I'll have to try some forced charge/discharge cycles tomorrow and see what behaviour I get.

Still ongoing problems unfortunately, I did another calibration and again all was well for a short time.

I suspect what is happening is that a problem is being detected and a discharge limit applied, which can be reset/cleared by:

• calibration, or simply charging for a short time
• turning the whole battery unit off and on again

It will then meet home demand for a time before starting to limit output and quick dropping to around 108 watts output, which it will happily sit at all day.

Pulling stats from GivTCP Cell 5 appears to be tracking at a lower voltage than the other 15 cells, 3.25v vs 3.33v, and occasionally sags below 3v (it briefly dropped to 2.57v yesterday). Not sure it this is a large enough variance to indicate a hardware fault in that cell?

Unfortunately I've heard nothing from support as yet, so I'm feeling kind of on my own trying to coax some life out of this unit.

chriswright What is the variance between cells when you're discharging/charging at full rate?

A variance of 100mV (0.1V) between cells is usually going to indicate a damaged cell and would explain your problems.

Post a picture of the GivTCP/HA graph of battery cell voltages over the last 24-48 hours.

the yellow trace is Cell 5,

The left side of that will be the last calibration, which appeared to get Cell 5 somewhat "in sync" for 7 or 8 hours and gave me a period of false hope ...

The 00:00 - 06:00 period in the middle was an overnight charging session, which again looks like it "synced" the voltage, but when the load picked up at 07:45 it drifts again.

So it looks like it tracks with the other cells, mostly, during charge but goes properly awry under discharge load, then you get that flat line where the discharge rate is limited.

(that chart is huge, open it in a new tab to read the keys)

chriswright this a 9.5 battery by any chance? Makes a change from it being cell 13 for once…

chriswright The busbar is probably loose but that in turn will probably have knackered the cell. Engineer visit probably required is the short story.

Try running it through a few cycles of charge/discharge at a limit of 800-1000W and see what happens. By that I mean AC charge the battery to 100% then force discharge at 800-1000W until it cuts out at 4%. Give it a couple of hours to idle at that level (it'll rebalance cells) then rinse/repeat a couple of times.

That's basically what GE are going to do (slow calibration) when they eventually respond to you.

FWIW I don't think it'll make any difference but you may as well try.

hoggy Makes a change from it being cell 13 for once…

Cells 3, 5, 8 and 13 appear to be the ones most likely to be "stressed", probably because they've got the busbar/cabling connected. All of those on my 9.5 took a hammering with v3013/3015 on a G1 (400 series f/w) inverter.

Vestas I've discovered recently (getting rusty) there's atleast 2 versions of the G2 9.5.

One has solid busbars everywhere all the way back to the BMS (I think this is the newer type)

Other has wired links between modules (that I've pictured before)

Not sure which one is more flakey!

hoggy It'll be the solid busbars and lack of the correct loctite-type product* applied to the thread which will be the problem. Vibration in transit followed by heating/cooling as its no longer secured properly and there we go. Poor mechanical design but that's not uncommon.

Its not rocket science - there were products we used 40 years ago to lock threads which coped with temps from -55C to +70C with 15G vibration. They weren't expensive then and they aren't now.

*you'll remember the method they use to secure the USB stick - doesn't really indicate much thought.

This is a Gen2 9.5 installed early this year, so probably the newer type.

I've got it current limited to 855 watts now (both ways) and charging, since I've not been able to discharge properly I was already at 63% so I thought I'd go to the top and probably leave it there for a few hours before I do a discharge.

Wondering how well the discharge is going to work, I have a horrible feeling that its going to be a 108 watt discharge.

Hi Chris, I had this issue with a battery that was just installed, replacing a faulty one. it's a firmware issue that needs updating. after that it has worked fine

the firmware is already on 3017.

As of Wednesday the questionable battery is now the secondary in a two battery pair (and as such has been through yet another calibration, both batteries are Gen 2 9.5) and isn't discharging to keep pace with the primary. Currently seeing the primary at 72% and the secondary at 90%.

Hopefully that is obvious enough an issue to get GE to act on it, if only I could get them to answer my emails!