Excess PV being exported to the grid instead of charging the battery

Hi. Long time lurker, first time poster etc.

I have solar with the GivEnergy inverter and 9.5kW battery, along with an iBoost (currently powered off though, to eliminate it from the suspect list).

All has been working as expected, only today it's been acting a little weird. I've been WFH and have been idly monitoring the dashboard when I noticed that the Grid Export figure is waaaay high (a kW or two) and the battery isn't charging. Looking at the power graph, from 11:15ish today I see this:

Highlighted just for PV, export, load shows this:

Basically, the differential between generation and load is being dumped out to the grid.

I've raised with my installer, who cannot see anything wrong with my config (Eco mode, nothing else), and have raised it with GivEnergy, who've told me:

_"I have just sat and watched your system for around 2 minutes and this export is showing because every now and again your load spikes above the solar generation which causes your battery to go through a change over process from charging into discharging, shortly after (around 5 seconds) your load drops down again and your inverter then has to reverse this change over process back to charging during these change over processes the excess solar is being exported to the grid.

It is being displayed more drastically on your end as you see 5 minute data points so all of these change over periods are not being shown/displayed as they are happening numerous times through out that 5 minutes."_

Thing is, I've never seen it do this - and I love watching the dashboard. Additionally, I've checked previous days going back a few weeks, and these spikes aren't there, which would be the case if the explanation given by GE was right.

Being a bit of a data nob, I exported the last few weeks of data to Excel, and ran a query to show me any time the PV was higher than the load, but the PBat was 0, meaning the battery wasn't charging. I'd average 1 a day, usually a delta of a dozen watts or so, and usually before the PV kicked in.

Today I've had 46 instances and counting. I've restarted the inverter through the portal and it continues after. I've powered off the iBoost, but it continues after.

The system was only installed in November, and since then I've exported a total of 15kWH, an average of about 120W a day. Today it's exported just shy of a kW, and I know, I know, it's only a kilowatt, but to me it's an indication of something awry, but I seem to be getting the brushoff..

Can anyone shed any light on what I should be looking at? It seems a bit daft that it's just elected to start dumping power out to the grid for no reason I can see..

Help?

And apologies for the lengthy post...

The battery seems to take some time to adjust to spikes in usage or generation. Oftentimes that ends up with deployment to grid if it can’t go anywhere else. But this shouldn’t last long.

It did seem particularly erratic today due to the sun constantly peaking in and out behind clouds.

That’s all I’ve got sorry. I’m learning still too…

pacemaker I’d go along with this, only today’s sunshine has been no more or less erratic than any other day in the last… 8? And the patterns of export have been consistent as anything since we’ve had it. It’s only today it’s started to play silly buggers, and like I say, we’ve exported nine times as much as any other day since we’ve had it.

I could pick any day since we’ve had it and the graphs look similar, it’s only today it’s shifted to this pattern of exporting instead of charging.

I think I’ve come to learn the patterns of what the inverter is doing for the most part, and this is.. off.

Has your system shown the same pattern today of exporting rather than charging?

Atkinstagram Have a look at My Inverter > GRID DATA . Are you seeing regular 250+ volt spikes that match the exports? If so suggest you report it to the DNO

Atkinstagram

Mine was doing both yesterday but wasn’t exporting for long. But it did seem to happen often as where I am the sun kept peaking as clouds moved fast from wind.

SilverArt

Hiya. Checked grid data, and it's fairly stable at around 237, 238V, with a max of 240V and a low of 233V.

Today is very similar weather-wise, with spikes of PV generation, but it's acting as expected, charging the battery and not exporting a drop (or so little as to be inconsequential).

The only difference is today I'm not at home, so the load is very low, in the order of 80-120w. I'd hope that the fact I was in the house and using power wouldn't be sufficient to act so out of character. I get the point about the load changing and the battery reacting to it, but.. I'm home 50% of the work week and all weekend, and it's not usually acting weird like it was yesterday.

Maybe it's something I need to put down to being a(nother) quirk of GE..

I have seen exactly the same, clear time of day, no clouds, no change in house load, battery half charged and sudden spike of power being sent to the grid.
Never seen it before despite much looking frequently at what’s happening.
In the figures table from inverter data just one reading went to grid.
WHY? Not high mains spike as suggested above, regular at 248ish