That is exactly the error my unit displayed but it didn't matter if it was charging or not. This turned out to be some relays which had welded their contacts closed. How they did that remains a mystery.
Despite the error code saying "open relay", I suspect you actually have a stuck relay like I did, one that should be open but is closed. The system detects this and flags the error. The error code should be more accurately stated as "relay error".
If you take the top cover off the unit, you will see this large relay board (this one removed from my unit, yours will have wires attached to it all over):
View attachment 278013
The power relays are the large black boxes, 11 of them in this picture. I bet one of them is stuck closed.
To test this, remove ALL power from the inverter, PV, battery, and grid. Let it sit for a minute or two for caps to discharge.
Then take a multimeter in resistance or beeper mode and measure from the sea of little silver dots on each side of the relays. Those dots are PCB vias (holes through the PCB that conduct) in large wide traces carrying the power. Using a long straight dentist pick or other long sharp metal probe will help you get in between the relays to touch the trace dots. With power off, there should not be conduction across any relay.
If you feel brave, you can remove the board and measure them on the backside. The wiring and PCB contacts are actually quite nicely labeled (at least my unit was), but be VERY sure to get all the wires back to the right connections. Take pics! Also, handle the board with care, it is heavy and there are small parts on it that will break off easily if the weight of the board runs through them.
The top row of relays connects the inverter L2, L1, N to the LOAD1 L2, L1, N.
The next two rows are doubled up relays that connect the grid L2, L1, N to the LOAD1 L2, L1, N. The implication here is that the grid to load connection has two relays to provide redundancy in case one relay fails.
The two relays on the right connect the generator L2, L1 to LOAD1 L2, L1.
There are two more relays on the communication board in the lower section. They connect LOAD2 L1, L2 to LOAD1 L1, L2.
Your likely culprits are the 9 in the center, the grid and inverter relays. I doubt the generator or the LOAD2 relays matter for your issue.
You may be able to free the stuck relay by tapping on them. That is usually only a temporary fix as whatever caused them to stick in the first place will likely recur. Try to find the guilty relay first, though, you want to know which one it is.
For me, I had 4 busted relays, 3 of which had welded. This is what a good set of relay contacts look like inside the cover in the open position:
View attachment 278014
And this is what a set of welded contacts looked like:
View attachment 278015
This happens on an overload of sufficient excess to heat up the contacts that they melt and stick. You can see arc flash discoloration, too. There was no apparent damage to the rest of the inverter, nothing else seems fried, though I can't really be sure of that.
I am still puzzled by this happening, but it clearly did somehow and I know of at least one other case like this. You might be a third case.
You can buy new relays here:
HF167F 12-HF
All the big black relays are the same part number.
The issue is replacing them requires some skills and tools for heavy trace PCB work. A typical soldering iron won't cut it, the big traces take all the heat away and you can't melt the solder, you really need a fairly large heat gun and the right technique to soften all the terminals at once and pull the old relay out, then solder wick to clean up the holes, and then a big soldering iron to solder in the new part. Not something your average solar DIY type is gong to be able to do effectively.
I have found that Watts247 is well meaning and supportive, but not equipped with engineering information that would be helpful, and the Amensolar factory support is meager at best.
These relays seem to be a weak spot in the inverter design, or there is some sort of significant overload fault that the inverter can create somehow on its own that ruins them. It is weird.
Also, each relay uses about 2 watts of power. In "normal" use, 11 relays are on (3 to the inverter, 6 to the grid, and 2 to LOAD2) which totals 22 watts. The unit idle power is about 70 watts (measured on mine), so 22 watts of relays is a significant portion of that idle power, nearly a third. Just an interesting fact.
Mike C.
Any ideas on how to fix this? Ian at Watts247 forwarded my problem to the factory, but figured I would check here as well.
