Error messages not matching the error (Goodwe)

Inverters Goodwe-AlphaESS
1

Earlier this week I’d had an electrician out to do a fresh COC for the PV system. He was unhappy and would not sign a COC. Because

  1. Inverter frame not grounded
  2. Panels not grounded
  3. Cables in the ceiling from the panels not in conduit.

All of this was rectified on Friday.

There was some confusion when the system wouldn’t restart. It would go into self-check mode (as it always does), then the red error LED came on. Then it went back into self-check mode, then the error LED came on. Wash. Rinse. Repeat.

One of the errors was “Isolation failure”. I read the user manual. It suggests high humidity, a broken DC string as possible causes. Gives some tests that you need to be an electrician to carry out and know what you’re doing.

The other was “Ground failure”. To quote from the manual: “Ground failure
could be caused by multiple reasons like that the neutral cable on the AC side is not connected well or the surrounding humidity is comparatively heavy, etc.”

Again some tests are recommended. The electrician tells me he’s checked all that already.

So I say well, let’s shut it down, then just turn the AC supply on. That will start the inverter and we’ll see if there’s an error or not.

No error.

So we reconnect the battery.

No error.

Go and have coffee. Come back. Still no error, but this isn’t a lot of help as there’s no PV. But also it suggests that despite all the stuff about the neutral cable, the problem actually is one of the PV strings. So we agree on that.

Go to turn on string 1 and immediately there’s an error. OK… turn that off. Let the inverter settled down. Go to turn on string 2. Same thing: Immediate error.

Then I notice that there’s a red flag showing on the SPD for string 2.

Now the electrician is inspired. He opens up the combiner box, disconnects both of the SPDs. We turn both strings on. Peace and harmony reigns.

So a new SPD is fitted. Everything OK again. The SPD that sacrificed itself was causing both error messages.

But all the time we could see the voltage from both strings. The electrician double checked with his own meter. So there was no break in the DC strings - you don’t get a voltage over an open circuit. Nothing about an SPD that had shuffled off this mortal coil matched the explanations in the manual for the two error messages.

Not helpful.

2

A bit more scary, the unearthed case of the inverter likely caused the error not to be raised (can’t leak if there is nowhere to leak to), which masked a fault condition. Makes you think doesn’t it :slight_smile:

1 Like
3

Thanks for that insight.

I can’t say for certain when the SPD popped. I had thought that it must have occurred whilst the work was being done, because I had never seen those errors before. But as you say, the lack of earthing may have masked it.

İ suppose I should get into the habit of checking SPDs every now & then. I have 5 in various DBs plus at least 3 on plug points.