Dual Inverter Voltage Control

TheDragon (GivEnergy) Some good news, we have had tested the Grid over voltage issues. Where the inverter over 253 goes into alarm.

You've got quite a few issues on your hands at the moment but want to chuck in my "two pen'orth". We have dual hybrids, not on beta and I can see that between them, they jack up the local grid voltage way in excess of 253V causing the Zappi to trip on overvoltage.

On Tuesday this week, we generated over 64kWh from solar. It would have been more but we'd consumed everything we could (EV charge, washing, dishwasher, oven etc) and clipping started in the afternoon. Once self consumption reduced, the inverters were pushing each other to export, the voltage started to go up. I set the "Inverter Max Output Active Power Percent" to zero on one inverter and the other inverter reduced voltage nearer to the grid voltage but took over the export. After ten minutes, I reset the "Inverter Max Output Active Power Percent" back to 100% and reduced it on the other inverter to zero. The inverters swapped over house load and export; the voltage remained reasonable and export was limited to 3.68kW. After another ten minutes, setting "Inverter Max Output Active Power Percent" back to 100% on the second inverter again saw the voltage start to rise as they fought each other to export.

I can see why this happens and it's only going to affect a few installations. My work around will be to do some automation in HA (if I can get "Inverter Max Output Active Power Percent" exposed as an entity). However, for me, it would be useful if, when the inverter voltage exceeded 251V, export reduces inversely proportional to voltage so that by the time it hits 253V, export is zero.

I know this would disadvantage Givenergy inverters compared to other inverters who could continue to export at rates in excess of 253V and maybe a change in the G100 regulations is needed to see some throttling as the 230+10% limit approaches. Whatever the decision within Givenergy, it's good to see that there will be formal logs of high voltage.

I like this idea. The possibility of high grid voltages will only increase, as more people invest in home solar systems (plus local solar farms) - which in turn will result in an increasing number of grid-powered devices squealing or tripping or generating error messages.

Presumably this is more of an issue for those those consumers who have higher impedance connections to the grid (e.g. due to under-rated grid wiring, or particularly long runs from the local transformer). Perhaps we will all need to invest in servo-driven variacs between our meters and consumer units 🙂