GFI on inverter: ground and neutral bonding issue?

[greg_mustang]

Hi,
This is my first post thanks in advance for comments.

I have a small PV setup (400W solar), controller, inverter (Renogy2000W) and 12V storage of about 200A.
The intent is to supply the electric loads in my garage. My garage panel is fully isolated from the house, it is a small panel with three 15A circuits. Within the panel, the neutrals and ground are bonded. The garage panel ground is connected to the same ground copper feed of the house panel. Nothing else is connected between the two panels of course. My PV power system (output of 120V inverter) is wired to feed the garage's electric panel input (hot, neutral, and ground).
The question is likely about grounding / neutral bonding... because at the moment I flip on any breaker on the garage panel, the inverter goes into ground fault.

Is the inverter likely internally ground-neutral bonded? Should it be?
Should the inverter chassis ground by grounded to my panel ground? I never connected it.
Or should the inverter chassis ground be connected to the ground at the 120V outlet?
Or should I disable the GFI detection internally on the Renogy inverter?

Let me know, I would think this is a common question. I watched a bunch of youtube videos but never found a straight answer.
Hope you can help.

Thanks!
Greg

 

[greg_mustang]

I forgot to mention that the inverter (12VDC to 120VAC) works perfectly and does not trigger its GFI when not connected to the garage panel.

 

[TacomaJoe]

I know GFCI doesn't like a neutral/ground connection on the load side. In testing my trailer GFCI, I can take a 3-wire pig tail, plug it in and as soon as I touch the neutral and ground together, the GFCI trips. Mind you the hot is not connected to anything, just a neutral/ground connection.

In your case I don't understand why the garage panel neutral/ground bond doesn't trip the GFCI on the inverter until you turn on the breaker. What sort of loads are on your circuits?

Test yours by removing the panel bond. If it works you're in business. Here is my thinking: The neutral/ground bond in the panel is to provide a path for fault current to return to the panel and trip the breaker, cutting power and removing the hazard of an energized circuit. The inverter GFCI does the same thing, any current imbalance between the hot and neutral will trip the GFCI opening the circuit and removing the hazard.

 

[hwy17]

Within the panel, the neutrals and ground are bonded.

Does the garage panel have a 3 wire feed (hot/hot/neutral) or 4 wire feed (hot/hot/neutral/ground) from the main panel?

If it's a 4 wire feed and bonded in the panel I think that's wrong by any code, current or grandfathered. Bonded with a 3 wire feed could be grandfathered.

In any case, I don't think a bonded inverter is suitable for backfeeding a subpanel whether it's bonded in the subpanel or only in the main panel, either way you have two bonding points connected.

 

[greg_mustang]

I know GFCI doesn't like a neutral/ground connection on the load side. In testing my trailer GFCI, I can take a 3-wire pig tail, plug it in and as soon as I touch the neutral and ground together, the GFCI trips. Mind you the hot is not connected to anything, just a neutral/ground connection.

In your case I don't understand why the garage panel neutral/ground bond doesn't trip the GFCI on the inverter until you turn on the breaker. What sort of loads are on your circuits?

Test yours by removing the panel bond. If it works you're in business. Here is my thinking: The neutral/ground bond in the panel is to provide a path for fault current to return to the panel and trip the breaker, cutting power and removing the hazard of an energized circuit. The inverter GFCI does the same thing, any current imbalance between the hot and neutral will trip the GFCI opening the circuit and removing the hazard.

So with the inverter connected to the garage panel (and ON of course) the panel busbar is energized. The moment I flip on any of the three circuits, the inverter trips its GFI. Just basic lights and outlets, and a tiny cooler. I cannot for the life of me think that there is anything wrong with the circuits. The garage panel was previously powered by a generator with no problem, nothing changed.
Some more info: the panel is in my garage, but the solar power system is at the shed. Between the garage and shed, I have L15-30 receptacles and a 30 foot cable. The receptacles/connectors and cable only extend the three connections HOT/neutral/ground to provide the feed from the shed (solar system) to the garage panel.
Would the cabling distance be causing the imbalance and issue?

Greg

 

[greg_mustang]

Does the garage panel have a 3 wire feed (hot/hot/neutral) or 4 wire feed (hot/hot/neutral/ground) from the main panel?

If it's a 4 wire feed and bonded in the panel I think that's wrong by any code, current or grandfathered. Bonded with a 3 wire feed could be grandfathered.

In any case, I don't think a bonded inverter is suitable for backfeeding a subpanel whether it's bonded in the subpanel or only in the main panel, either way you have two bonding points connected.

Thanks.
The garage panel is not a subpanel. Its totally isolated (only shared its ground to the house).
Would there be an issue if I just remove the GFI protection at the inverter? In case of short to ground, won't it just use the bonding at the panel to ground the current and trip the breaker?

Greg

 

[hwy17]

So this is like a critical loads panel just for solar? If the inverter has a ground neutral bond I would think you could separate grounds and neutrals in the panel an unbond them, leaving only the inverter bond in the system. But this is outside my experience I have only worked with my unbonded inverter.

You could bypass the GFCI and probably it would ok, but I think you would be ignoring the fact that you have neutral current on the ground which isn't great.

 

[Sharing Society]

Did this ever get resolved?

I've seen this issue twice myself with the RENOGY; I think it is because they are designed for MOBILE applications.

Now I found these video modification solutions:

DISCONNECTING Ground Fault system on RENOGY 12V Inverters

ALSO; removing ground neutral bond:

 

[TacomaJoe]

I have a new theory for OPs issue. Feeding his bonded panel was OK until a circuit breaker is turned on to power any load which causes the fault trip. Here's what I think. Renogy didn't implement a grounded neutral transformer and won't fault with the panel neutral-ground bond. When a circuit breaker is turned on, that load current flows through the hot and neutral like it should. What happens is that the panel bond splits the current between the neutral wire and the ground wire (they are in parallel because of the panel bond and the Renogy internal bond). If you have 2A on the hot and it splits, there will be 1A on the neutral and 1A on the ground wire. The 2A on the hot and 1A on the neutral causes the fault.

 

[Sharing Society]

I have a new theory for OPs issue. Feeding his bonded panel was OK until a circuit breaker is turned on to power any load which causes the fault trip. Here's what I think. Renogy didn't implement a grounded neutral transformer and won't fault with the panel neutral-ground bond. When a circuit breaker is turned on, that load current flows through the hot and neutral like it should. What happens is that the panel bond splits the current between the neutral wire and the ground wire (they are in parallel because of the panel bond and the Renogy internal bond). If you have 2A on the hot and it splits, there will be 1A on the neutral and 1A on the ground wire. The 2A on the hot and 1A on the neutral causes the fault.

Yes. Any difference in the HOT/Neutral trips it.
Shown in the 2nd video I posted

 

[jdnttek]

Yes. Any difference in the HOT/Neutral trips it.
Shown in the 2nd video I posted

I also had same issue with GFI tripped on my renogy 3000w inverter. I watched the you tubes, replies. It seems all issues was coming from the ground wire. I've installed a separate load center from house ac panel. The intent of installing this off-grid system was to move a few house circuits to solar as a backup supply. I ended up disconnecting all ground wires from the circuit and the ground wire from the inverter. Everything on the house circuit is now working. Jeff