Growatt SPF5000ES doesn't trip RCD in batterymode

[drbytes]

I'm finalizing my installation and putting in place all the required components to get it pass inspection, so everything has to be to code. My installation now has all the required doodads to pass inspection. So before I invite the pencil pushing civil servant I decided to run trough an inspection as if it were the real thing.
I failed my own inspection when I tested the RCD (ground fault protection?
- When the inverters are not connected to grid and I induce a ground fault none of the RCD's downstream (or upstream) of the Growatts trip.
- When I am connected to the grid, so with the Growatt in bypass, and I induce a ground fault, it immediately trips all the rcd's.

Curiously when I push the test button on the rcd it does trip no matter if the inverters or on or offgrid..

The only variable when inducing the ground fault is the inverters power source, grid or battery.

I have a 300mA rcd between grid and growatt, and I have a 30mA and 300mA after the growatts.


What gives ?

 

[timselectric]

Are you using an auto transformer?

 

[Bud Martin]

Are you in the US or some where else?
Are we dealing with GroWatt SPF5000ES sold by Signature Solar as SPF5000US for US market with extra purchase of the Auto-transformer?
May be you should show us your wiring diagram also.

 

[timselectric]

It would have to be the modified for US model. Otherwise, this wouldn't be their question.
Don't ask me how I know. lol

 

[timselectric]

The solution is to pull a neutral wire along with the two hots. Between the main panel (source feeding the inverter) and the sub panel (that the inverter feeds).

 

[timselectric]

Also, you should have a relay that disconnects the auto transformer neutral, in inverter bypass mode.
But, the inspector probably won't know to look for that.

 

[drbytes]

Thx for all the replies, much appreciated.

I'm using the ES version and I am located in Europe . I have a delta network supplying grid power, so one side of the delta is laid to earth and a neutral is not being distributed by the grid operator. My 230V is obtained between two phases. ( in reality 3 phases are fed into the house, L1,2 and 3 -- HOT1,2 and 3 so to speak) But no neutral. Ground faults are detected because of the one phase to earth in the neighborhood transformer.
I measure from any phase ~130V to ground and 230V between phases.

I have the ES with the screw grounding the circuit board to the housing. Removing it breaks continuity from that solder pad surrounding the screw to the earth/housing.

I had another look at the inverter, in island mode it switches the neutral to the earth by opening a relay. That would mean that the neutral would need to be 0 potential to earth, instead I measure 120v to earth at that moment.

 

[Bud Martin]

It would have to be the modified for US model. Otherwise, this wouldn't be their question.
Don't ask me how I know. lol

Well, how did you know?

 

[drbytes]

Also, you should have a relay that disconnects the auto transformer neutral, in inverter bypass mode.
But, the inspector probably won't know to look for that.

Hold on.. how did you tap into that event? Via the relay contacts on the bottom of the inverter?

I've replaced all the earthing screws. But there are two screws with the earth/PE icon not 3 inches apart .. Do you guys remove both of these ?

 

[timselectric]

Well, how did you know?

I mistakenly bought the non modified version.
It cost me 2 TV's and a video recorder.

 

[timselectric]

Thx for all the replies, much appreciated.

I'm using the ES version and I am located in Europe . I have a delta network supplying grid power, so one side of the delta is laid to earth and a neutral is not being distributed by the grid operator. My 230V is obtained between two phases. ( in reality 3 phases are fed into the house, L1,2 and 3 -- HOT1,2 and 3 so to speak) But no neutral. Ground faults are detected because of the one phase to earth in the neighborhood transformer.
I measure from any phase ~130V to ground and 230V between phases.

I have the ES with the screw grounding the circuit board to the housing. Removing it breaks continuity from that solder pad surrounding the screw to the earth/housing.

I had another look at the inverter, in island mode it switches the neutral to the earth by opening a relay. That would mean that the neutral would need to be 0 potential to earth, instead I measure 120v to earth at that moment.

I believe that in your situation you need to replace the ground screw.

 

[timselectric]

Hold on.. how did you tap into that event? Via the relay contacts on the bottom of the inverter?

The one bonding screw (shown in the video by watts247) is connected to the relay internally. From that spot on the board to "L" will give you 230v, when the inverter is in battery mode.

 

[timselectric]

The solution is to pull a neutral wire along with the two hots. Between the main panel (source feeding the inverter) and the sub panel (that the inverter feeds).

Also, you should have a relay that disconnects the auto transformer neutral, in inverter bypass mode.
But, the inspector probably won't know to look for that.

For everyone else reading this. These solutions only apply to using this inverter in the US. (Split phase systems)

 

[drbytes]

I don't have a neutral that I can jump over the growatt into the main distribution panel.

And while it seems to work and power all the loads, .. when I fault any of the outputs to earth when in offgrid it happily keeps running.
Keeping that setup is beyond stupid, if I can not resolve that it will need to go, no way I'm risking that kind of danger. Suppose the outside lanternpole has a loose wire, anybody that would touch it when I'm running offgrid would get fried until the breaker pops at which point the person holding the pole would be very crispy.

 

[timselectric]

No, not in your situation. But, you should have an earthing wire connected to the inverter case. With the grounding screw in place, it should provide a fault path.

I don't have a neutral that I can jump over the growatt into the main distribution panel.

And while it seems to work and power all the loads, .. when I fault any of the outputs to earth when in offgrid it happily keeps running.
Keeping that setup is beyond stupid, if I can not resolve that it will need to go, no way I'm risking that kind of danger. Suppose the outside lanternpole has a loose wire, anybody that would touch it when I'm running offgrid would get fried until the breaker pops at which point the person holding the pole would be very crispy.

You might have to swap the two hots feeding from your main panel. To make sure that everything is in phase.

 

[timselectric]

Even though you don't have a neutral coming from the utility. Everything after the inverter should be wired as if "N" is a neutral.

 

[drbytes]

True, I haven't swapped the phases yet.

I do use the L en N color scheme.. right from the main panel through the inverters and even in the distribution panel.
"Everything after the inverter should be wired as if N is a neutral" --> I need to tie the neutral from the AC OUT on the inverter straight to ground?

 

[drbytes]

Won't all my grounding pins in my outlets be live that way?

 

[timselectric]

Hmmm, good point. They would be while in bypass mode.
Going to have to think about this.

 

[timselectric]

Don't bond your neutral to ground.
Let the inverter do that with the relay. That should allow the system to work correctly in both modes.
But, definitely test this. To confirm that I am thinking correctly. I'm not fluent with European power distribution.