diy solar

diy solar

No grid grounding/bonding

Moreover if you put the N/G bond downstream of the GFCI isn't the GFCI going to always see that as a ground fault and trip. I was actually thinking ironically he wouldn't be creating an unsafe condition because the GFCI would be tripping all the time :)
Actually, it would see a ground fault as nothing abnormal. Because it would rejoin the neutral before going through the GFCI.
 
Actually, it would see a ground fault as nothing abnormal. Because it would rejoin the neutral before going through the GFCI.
Okay... I agree it is generally a bad idea and shouldn't be done but I just had to test this. I just connected neutral to ground downstream of a GFCI in my house and it tripped. I actually thought it would take some load to make it trip - it didn't or at least I didn't have to cause new load. If GFCI works by tripping when there is unmatched current between hot and neutral wouldn't adding ground as a path to take some of the current off of the neutral cause it to trip?

I'll admit the first time I tried it didn't trip and then I put a GFCI tester on that particular circuit and it appears I have a bad GFI, the tester won't trip it either - so I tested two others and they both tripped in the N/G connect scenario.
 
Okay... I agree it is generally a bad idea and shouldn't be done but I just had to test this. I just connected neutral to ground downstream of a GFCI in my house and it tripped. I actually thought it would take some load to make it trip - it didn't or at least I didn't have to cause new load. If GFCI works by tripping when there is unmatched current between hot and neutral wouldn't adding ground as a path to take some of the current off of the neutral cause it to trip?

I'll admit the first time I tried it didn't trip and then I put a GFCI tester on that particular circuit and it appears I have a bad GFI, the tester won't trip it either - so I tested two others and they both tripped in the N/G connect scenario.
The test was flawed. Because you have two N/G bonds.
One upstream and one downstream.
 
And then Article 680 comes along and builds a massive ufer appendage on the system.
It's still one point of connection. (The main ground bar)
All earth connections (rod/plate, ufer, water, gas) should come together at one connection with the grounding system.
 
The test was flawed. Because you have two N/G bonds.
One upstream and one downstream.
Fair, I presume you are right but that leaves something I don't understand then about why the upstream one matters at all... darn it... more mystery :( oh well.
 
Last edited:
Fair, I presume you are right but that leaves something I don't understand then about why the upstream one matters at all... darn it... more mystery :( oh well.
The double bond you created. Gave a parallel path for neutral current around the GFCI. This is why it trips emidiatly.
With no bond upstream and a bond downstream. All neutral and fault current will travel through the GFCI neutral. And appear as normal (balanced) to the GFCI.
 
If you want to test it correctly. Remove upstream grounding from the GFCI. And bond downstream.
 
If you want to test it correctly. Remove upstream grounding from the GFCI. And bond downstream.
I think I have my head around what you are saying. I was testing at my house though and probably don't want to remove the panel bond for fun. But I could give it a go on my offgrid solar system next time I'm there :)
 
I think I have my head around what you are saying. I was testing at my house though and probably don't want to remove the panel bond for fun. But I could give it a go on my offgrid solar system next time I'm there :)
No need to remove the panel bond.
Just remove the upstream ground from the GFCI location. And test it downstream.
 
Poor Nan, she asked for one thing and this pummeling over ground Fault & other stuff MURDERED THE THREAD....

If you guys want to argue about GFI, CGFI, GCFI and the variants & related applications START A THREAD ON IT !
 
Back
Top