So I’ve been playing around a bit more and I’ve discovered a few things. Apparently anyone can sign up for an installer account at solaredge.com. I used a different email than the one I use for monitoring, but I used my same name and address. It also optionally asked for an inverter serial number during the sign up and I just skipped that part. But with the installer account you can sign in to Set App and not need to do the work around of connecting through the browser.
I spent a bit of time trying to get the inverters connected through WiFi, but it doesn’t look like it’s possible. I spent a lot of time reading the documentation and as near as I can tell the inverters either need a separate WiFi module (which mine don’t have) or a SolarEdge WiFi gateway box that I also don’t have. It doesn’t look like a WiFi connection will be possible without getting some sort of additional hardware.
However, I did discover that you can connect multiple inverters with RS485 and then only run an ethernet cable to one of them. That way if you have a number of inverters you only need one ethernet connection. I’m debating whether I want to go back and reconfigure my two inverters this way or not. I’m a bit tight on ethernet ports, so getting one port back wouldn’t really be a bad thing, but I’ve already run the two wires anyway, so maybe I should just leave well enough alone.
Finally I also was able to find the P(f) settings in the inverters and I configured that. This should allow it to work with the powerwall to scale back solar production as the powerwall gets full rather than just cutting off solar production entirely. As as been previously discussed, when you are off grid, as the powerwall starts getting full it will slowly start raising the frequency, In my case as the powerwall gets up to 98% full I saw the frequency rise to 60.1 and 60.2, then as it continued to charge the frequency would keep rising until it got to the point where the inverters would shut off.
However the solaredge P(f) setting allows the inverter to scale back as it sees the frequency rising. You set a low point and a high point and it will linearly scale back between those settings. For example you could set 60.2hz to 100% and 62.2hz to 0% and as long as the frequency is 60.2hz or less the inverter will supply 100% of the available power. But if the frequency rises to 61.2hz (which is halfway between 60.2 and 62.2) then it will only supply 50% of the available power. If the frequency gets all the way to 62.2hz then the inverter will shut down entirely.
Normally when you are off grid the system will charge the powerwall to 100%, then turn off the inverters to let the house run on the powerwall until it drops down to about 97%, then the inverters will come on again until it gets charged back to 100% and keep cycling as long as you are off grid and there is more solar power than your house can use.
Instead, using P(f), as the powerwall charge nears 100% it will slowly increase the frequency which will cause the inverter to slowly scale back the amount of power it’s feeding to the system. In theory it should reach an equilibrium where the power being fed in from the inverter is about equal to the amount of power actually being used by your home. So the powerwall will stop charging and the home will just operate on solar power. Obviously there may be a little fluctuation as your home’s energy use rises and falls, but this should still allow the system to increase and decrease the solar output as needed without going through that cycle of turning the inverters off, then turning them back on over and over.
I was hoping to test this today, but as luck would have it today has been very overcast and rainy and there’s no way my powerwalls will get up to 100%. Perhaps I can test tomorrow if it’s sunnier.
Also, note that this only comes into play when you are off grid and have nowhere to send any excess solar production. If you are on grid and the powerwalls get to 100% then things will continue to operate normally as the excess power will just be fed into the grid.