Error 8: Ground relay test failed on Victron Multiplus II GX

I have nothing useful to add… but it does remind me of this

Friendly_Switch

1 Like

I can actually tell you what is going on here! Thanks to the wonders of YouTube.

In the US, it is common to have double outlets, where one outlet is permanently on, but the other is switched by a light switch (they call it half-hot). You would then plug in a floor lamp into the correct outlet, and you can then turn on your floor lamps with a switch by the door. Pretty useful in a culture where bringing your own lamps to a rental place is common :slight_smile:

Of course you shouldn’t be switching lights in a completely different room, but who hasn’t done that. In the house I grew up in, the light switch for the kitchen back porch light was in the garage, which was adjacent to the kitchen, but you still had to go into another room to actually use it.

My father wired it. I think it was a simple matter of forgetting to install a wall box for the switch, and that’s just how it was done temporarily, and then it became permanent.

I have a similar non-functioning switch in my own home. After some time, I discovered it feeds a Euro-style outlet on the back patio. The previous owner had a tree decorated with Christmas lights plugged into it. So that light switch… was for the Christmas lights outside. Of course they did it the cheap way by burrying 1.5mm unarmoured twinflex into the garden… so once it started tripping the RCD I simply ripped the whole thing out.

So now I have an outside socket, switched by a light switch in the house :slight_smile:

2 Likes

Because this story just won’t die.
Loadshedding ends last week, inverter trip, error 8, you know the deal.

Back to my debugging protocol, pull out neutrals until it stops tripping.
Found the culprit, left it unattached. Except, everything still works.

Until lunch time, when kitchen lights stopped working.
Until dinner time, when kitchen lights started working
Until morning, when kitchen lights and lounge lights stopped working.
Fun.

Electrician came out this morning, found 200V between the disconnected neutral and earth (which I’m lead to believe is not ideal).

Told him to pull all cloth-clad wires out the roof.

He found this (inside a metal pipe of course):
Nicely frayed neutral

And a live connector with a rogue inner piece jutting out to touch whatever it likes.

Fingers (wires?) crossed that’s the last of it. Old houses…

Edit: New circuit in and… it still trips :see_no_evil:
Remove the new circuit neutral, happy days.
Electrician coming back tomorrow to debug further

2 Likes

To wrap up (all toes and fingers crossed).

Wednesday loadshedding ends, inverter trips, remove the neutral off the new circuit, inverter is happy.
Except, kitchen lights (on the new circuit) now flash like a disco. Concerning…
Electrician arrives, lights are no longer flashing, system won’t reproduce the fault.

Thursday, neutral still removed, kitchen lights now fully on.
Electrician arrives, goes into the roof, tells me to connect the neutral. Immediately he shouts “I see it! It’s sparking!”

Turns out there was still one remaining cloth neutral in a metal pipe between 2 of the kitchen lights.
So that was feeding to ground, and the lights “worked”.

Replaced that neutral, all happy so far.
I believe I’ve now removed all paper/cloth wires from my roof. Fingers crossed that’s the case…

4 Likes

In TTT’s world, the moment he thinks “There we go, Bob’s your uncle!” … he is reminded that Uncle Bob is Murphy’s brother! :rofl:

My point … you HAVE found the cause. THAT is the important thing. If there are more issues, you know where to look for them.

And THAT is Priceless!

Yeah, I’m all too familiar.
Thing is, I’ve had, over the course of the last 10 months, numerous “bouts” of neutral-ground faults.
Every single one of them came back to a degrading light circuit in a metal pipe.

It’s just taken a long time to actually rip all the old wires out (and it’s really had to know that you haven’t missed one somewhere, case in point this week).

So… :crossed_fingers: :crossed_fingers: :crossed_fingers:

So, question. Wouldn’t a good old insulation test have found these very quickly? I imagine it would.

If the faults were consistent, maybe.
But they weren’t. Sometimes fine, sometimes not, presumably as pipes expand and contract, or as condensation builds up in the roof.

On Wednesday, after the disco show stopped, we moved the live and neutral from the troublesome circuit onto the earth leakage, no issues. And the inverter was happy, until that evening when it tripped after load shedding again…

The point of the insulation test is that it uses a higher voltage, that is much more likely to break through weak insulation. A cloth-insulated wire that is in close proximity to a pipe may not arc over at 230V, but it will at 1000V.

I can imagine that some faults would still be missed though, and the result would have been the same anyway: replace all the bad wiring.

Here is an interesting video (skip to 5:30 for the beef) by Mr. Lorton showing why.

2 Likes