SolarEdge HD Wave inverters arc faulting on Smart Meter comms

[NukeEngineer]

I have a self-installed 13KW DC, 12KW AC roof-mounted system using two 6KW HD Wave inverters, that has been online for almost 2 weeks now. Starting from day 2, and every sunny day since, they have both reported arc faults several times, beginning after noon and stopping around 3-4pm.

I was standing in front of them when they did this once, and I heard the sound of sizzle coming from the meter, then both arc faulted. I recognized this sound in my ceiling fan motor and LED lights, as it is the sound of the smart meter communication.

I had the meter guy from my POCO come out and manually initiate the meter's comms, and sure enough, they both arc faulted! The meter tech tried another meter, but it did the same thing. He is now contacting the meter supplier to see if there's something wrong.

I have also contacted SolarEdge, and am waiting for their Tier 2's to look at this. Certainly, I can't be the only one experiencing this problem with the HD waves, right? Anyone else seen this?

[max2k]

... I recognized this sound in my ceiling fan motor and LED lights...

I would blame the source instead of inverter since you can hear it elsewhere in the house.

Did you hear this sound before you installed the system? Did you try to turn inverters off when tech was there and initiate the comm? If your fan would still produce the same sound it would eliminate inverters entirely, they would be just another victim of interference form the meter.

[NukeEngineer]

Yes, its not really audible in the LED lights, but there's a clear flicker. In a ceiling fan, it sounds like a bad bearing, for 2 seconds. It definitely roughs up the 60hz wave.

Yes, I have heard this sound before the solar went up, as well as hearing it in my previous home, without solar.

[Ben25]

Are you sure everything's wired and torqued correctly? I've had led lights mess with my 10kW solaredge (DC injection on AC line) but not the other way around.

[NukeEngineer]

Are you sure everything's wired and torqued correctly? I've had led lights mess with my 10kW solaredge (DC injection on AC line) but not the other way around.

Everything is wired and torqued correctly. The one connection that I was skeptical about was the Ilsco IPC, which taps my 4/0 aluminum with a 1/0 copper. However, I took these off when this issue first arose to inspect. They made nice clean square bites into both cables, with no signs of arcing.

Anyway, I've also had my POCO meter guy come out to initiate a manual meter read, and these also set off the arc fault detection. Even using a different meter, did the same thing.

I've ordered some ferrite chokes to put on the AC lines, hopefully will help.

[NukeEngineer]

Heard from meter guy at POCO. He talked with meter manufacturer. They have heard of this issue before, involving Enphase micros. He said Enphase released a firmware update to address this specific issue, but of course SolarEdge can't see what their solution was. He will be sending a signal to the meter to reconfigure it's firing angle to be lower, closer to the zero crossing. It will probably be tomorrow before I get a result.

[emartin00]

Does it matter if the inverters are producing power or not? Maybe they could program the meter to communicate at night instead?

[NukeEngineer]

Does it matter if the inverters are producing power or not? Maybe they could program the meter to communicate at night instead?

Yes, they need to be pushing at least about 4KW. Also, it's not that you need at least 8KW going through the meter, you just need at least one producing 4KW.

What I mean by that is this... I had one inverter off while I was doing something one day, but the other was on pushing about 4KW. I heard the meter's communication sizzle sound, and that one inverter that was running arc faulted.

I would like to suggest to them to read my meter at night, if their system will support that, if their recent change in it's communication method doesn't work.

[NukeEngineer]

Still having this issue, and I think I've gotten all the help I'm gonna get from SolarEdge, my POCO, and Aclara (meter manufacturer). I would have thought, for sure, SolarEdge would have been interested in resolving this, but they didn't seem too concerned about it.

So, I'm stuck with my inverters tripping off at least twice a day, at 12:26pm and 1:10pm +/- a minute or two. They actually report "HW Error" instead of arc fault now, but that's because SolarEdge disabled the AC side arc fault detection (or effectively anyway). If they didn't do this, someone would have to turn the switch off and back on twice a day, as arc faults shouldn't be cleared automatically.

I'm less concerned about the 60 or so seconds, twice a day, of missed production as I am about having the AC side arc fault protection bypassed. My state uses NEC 2014, and I couldn't find if there is a requirement to have AC arc fault protection from a grid interactive inverter. Does anyone know if this is required by code? If it is, I may have more leverage with my POCO to get them to do something about their meter.