Micro Inverter Battery Backup

Good Day everybody

I posted this here as im not sure it fits in any specific category.

I would like to pick the brains of some of the people that are more knowledgeable than I would be.

This is a topic ive thought about after a recent discussion with a friend.

Would you use mico inverters combined with a battery backup type inverter.
Specifically an axpert type inverter, thus having batteries charged by AC that is supplemented by the micro inverters.

To me what seems to be a downfall of microinverters would be that they shut down when there is no incoming AC. Thus the opportunity could arise where you would be losing out on some production as a result of loadshedding.

Now what seems to be puzzling my mind, and yes I know there would be some cost involved and possibly no point to doing this, but would it be possibly to power the micro inverters from the axpert during loadshedding, using the right contactors to switch between the two AC sources under certain conditions.

I would like to hear your thoughts on both


I may be mistaken, but I’m pretty sure you can’t put a grid-tied inverter on the output of an Axpert. You will possibly blow up the Axpert.

I think the Victron inverters can handle micro inverters if they support the sunspec protocol (I think Hoymiles do a little)

The problem is not if you can to this but if it makes sense. If you have micro inverters then it all is in AC, but you have many smaller inverters and they all sit outside in the weather. And the nasty is if you need backup power they do not work. Now if you think of adding another backup power inverter you now how another inverter meaning you have all the extra cost. I had the same thoughts.
In then end it is way cheaper just to buy the Victron that has power backup.

If you put an AC source on the output of such an inverter, what I expect will happen is the voltage on the HV-bus will rise. Best possible scenario, there are plenty worse ones.

These inverters are HF designs. They have a boost stage that turns the 48VDC into 330VDC (the HV-bus), and then they have a final stage that chops that into 230VAC. Why 330VDC? Cause 230VAC is the RMS value, it is like an average. The peak is actually at 330V, give or take.

But these inverters also raise an alarm and switch off if that HV bus goes out of bounds. And since it is not expecting energy to arrive by those means, this is the best case scenario I would expect: The inverter would just switch off.

Furthermore, it is going to depend on the semiconductors used in that final stage. I’m not an electronic engineer and I don’t know for sure, but if they are using IGBTs, those are directional.

Finally, for the whole thing to work the inverter needs to raise the frequency proportional to the battery voltage, so that the grid-coupled inverters throttle back. Again, the Axpert cannot do that either.

So this is going to fall anywhere on the spectrum from does-not-work to I-blew-up-an-axpert.

Don’t need to support sunspec. You just need to support GFPR (gradual frequency-dependent power reduction). That’s simply a method whereby the island inverter gradually ramps up the frequency between 50hz and 53hz to control PV-inverters on the network. At 52.7Hz the PV-inverter is throttled to zero, and at 53Hz they switch off.

Yup. I think Hoymiles is the only one that supports GFPR. At least the only one I know of. It’s not a very common feature in micro-inverters.

1 Like