Unexpected battery calibration

Is it normal behaviour that batteries run a calibration cycle every 3 months or so?

I've got a 9.5kWh battery on v3012 which is currently calibrating. It also ran a calibration cycle about 3 months ago. There are no calibration notifications showing in the portal apart from the initial one 7 months ago when the portal was comissioned.

If this is expected behaviour is it possible to improve the scheduling of it as I'd been charging overnight when the discharge stage kicked in. If it's not expected does anyone have any ideas on what may be causing it?

No this is not normal behaviour. The battery should only calibrate when its initially commissioned or when you do a firmware update.

Might be worth contacting Givenergy support to ask why its happened. They can kick them off remotely but of course shouldnt do so.

It has been known for the register that triggers them to get set randomly, presumably though a misconfigured/garbled request and/or network error.
GivTCP does seem to be a common denominator but equally I've seen a few either here or on FB that it's just done it from the portal for some reason.

Support can stop it if you contact them.

Thanks, I'm not using GivTCP, just the Octopus beta R&D scheduling site.

The calibration has finished now but I'll contact them if this happens again. I might alos update the firmware and see if that fixes the issue.

Shini I believe latest inverter firmware now includes a crc check, which was intended to reduces risk of garbled messages. Unfortunately, the newer battery firmware may have other issues - I've been holding off upgrading for the moment.

I've had this happen when I've sent several configuration instructions (changing charge time, start or stop export, etc) in succession. I've learnt to wait for the requested action to take effect, which may take a minute or more in some instances, before thinking it has failed, getting impatient and sending again.

SimonP Is that when using portal, or API, or app, or local control over modbus ?
(With my API scripts, I tend to be changing several parameters together.)

DD That was using the portal.

I have posted this elsewhere but seems to be relevant here also:

Today (Thurs 14th March) was a day of Calibration woes.
I have two GE 8.2 kWh batteries daisy-chained to a AC 3.0 coupled inverter. Running on 3015.
At 5:45 pm, I checked everything in advance of the saving sessions. I found that my batteries were charging at full pelt, whereas the Predbat Plan had them discharging.
I checked Predbat to find my current status shown as 'Calibration'.
I checked the GE portal and spotted an inverter warning that I had experienced a BMS low voltage at 11:25 am. On checking, both batteries had recorded 1 kWh. And the batteries had been charging ever since on a calibration.
The calibration finished at 6:35 pm just after the saving session and having charged during 2.5 hours of the peak tariff.
Checking back, the same behaviour happened on 12th Feb: Low BMS warning following by an automatic calibration.
For me, I guess it was just one of those things. At least it kicked in when the batteries were empty. Pity it didn't start 2.5 hours earlier, before the peak tariff time and the saving session.
The GE portal does not record either calibration from 12th Feb and 14th March.
The screenshot shows the drop in voltage at 11:25 am (and also a rise when calibration was nearly done).

☹️

Rob

I had a new battery installed a couple of weeks ago, and had to do a full calibration again.

Good that Predbat spotted it was calibrated and ‘butted out’

What I found I could do was to turn predbat to read only and manually drop the charge rate to zero to stop charging in the peak period, then turn it back up when the peak had passed.