LVX-6048 Ground Fault

[mrzed001]

Can this not be tested by simply disconnecting AC, and operating off grid? Once the unit has no incoming connection to neutral or to ground or to both then any continuity N - G must occur in the unit. Correct?

Only AC L + N needs to be removed then, and G remains. Without G there is no 0V reference.

 

[eabyrd]

Only AC L + N needs to be removed then, and G remains. Without G there is no 0V reference.

I can be a bit slow, so let me repeat what I am understanding the test to be
1. disconnect AC-input L1, L2, N. Leave AC-input G in place.
2. Disconnect AC-output
3. Attempt to operate in Battery mode.
4. check continuity N - G At AC-output

is that it?

 

[Sanwizard]

It looks like his problem was no neutral-ground bond. Mine has a neutral-ground bond at the main panel (connected to the AC input of the inverter).

Check out Dan's video on grounding.

 

[mrzed001]

I can be a bit slow, so let me repeat what I am understanding the test to be
1. disconnect AC-input L1, L2, N. Leave AC-input G in place.
2. Disconnect AC-output
3. Attempt to operate in Battery mode.
4. check continuity N - G At AC-output

is that it?

Yep. If there is OL then inner relay not working (fixed in that position = never inner bond).
To test if it is fixed in other position (always inner bond) ... this is the remove house bond test. Or completely remove all AC (L -) G - N connections (in and out). Sorry I do not know when the relay is operated (without power it is on or off) so I am not sure if the remove AC cables will give relevant info.

 

[eabyrd]

I saw no disc

Yep. If there is OL then inner relay not working (fixed in that position = never inner bond).
To test if it is fixed in other position (always inner bond) ... this is the remove house bond test. Or completely remove all AC (L -) G - N connections (in and out). Sorry I do not know when the relay is operated (without power it is on or off) so I am not sure if the remove AC cables will give relevant info.

you mentioned adding a resistor. For the second test could I simulate the removal of the bond by placing a resistor on the N or the G leg into the inverter?

 

[eabyrd]

Check out Dan's video on grounding.

watched his video, different inverter, and different behavior in that his will run but possibly a similar cause. If the inverter is going to create its own bonded neutral connection then it needs to switch neutral as well as the hot legs in its transfer switch. If it doesn’t it creates a ground loop the fact his current goes away in Inverter mode suggests it does switch neutral between states, but keeps the bond in place even in utility mode creating a ground loop in that mode and generating the current he detected

 

[eabyrd]

Yep. If there is OL then inner relay not working (fixed in that position = never inner bond).
To test if it is fixed in other position (always inner bond) ... this is the remove house bond test. Or completely remove all AC (L -) G - N connections (in and out). Sorry I do not know when the relay is operated (without power it is on or off) so I am not sure if the remove AC cables will give relevant info.

Well then, Houston we may have found the problem. See pics, but with only G as the AC input, there is no continuity between output N and G

 

[mrzed001]

Well then, Houston we may have found the problem. See pics, but with only G as the AC input, there is no continuity between output N and G

Gooood So the relay does not bond N-G in battery mode.
You know this should not cause any problem in line mode. Only when inverter goes to grid separated battery mode.
Maybe a check here between incoming G - outgoing G and incoming N - outgoing N ?

I was starting to think through and make a model in my mind about the resistor solution you proposed ... but it was to much for my processor .

Here is where I was while stearing in a childish schematic paint pic I throw up fast :
Hmmm good question.
G is needed for reference and many reasons so there is not a good idea.
For N ... mayyyybe? If you put inverter into off-grid mode with grid support (not grid-tie) then relay should be in not bonding position (and not producing power out to grid, only load). But then can you measure the resistance between the outgoing G and incoming N ? Not a good idea?
Maybe better with resistor in incoming G and check with outgoing N ?

 

[eabyrd]

Gooood So the relay does not bond N-G in battery mode.
You know this should not cause any problem in line mode. Only when inverter goes to grid separated battery mode.
Maybe a check here between incoming G - outgoing G and incoming N - outgoing N ?

I was starting to think through and make a model in my mind about the resistor solution you proposed ... but it was to much for my processor .

Here is where I was while stearing in a childish schematic paint pic I throw up fast :
Hmmm good question.
G is needed for reference and many reasons so there is not a good idea.
For N ... mayyyybe? If you put inverter into off-grid mode with grid support (not grid-tie) then relay should be in not bonding position (and not producing power out to grid, only load). But then can you measure the resistance between the outgoing G and incoming N ? Not a good idea?
Maybe better with resistor in incoming G and check with outgoing N ?

Gooood So the relay does not bond N-G in battery mode.
You know this should not cause any problem in line mode. Only when inverter goes to grid separated battery mode.
Maybe a check here between incoming G - outgoing G and incoming N - outgoing N ?

well Houston I may have screwed up my prior test, because this time I did get continuity between output N -There is a direct connection between Input G, output G, and Chassis. So continuity between all and output N as well. There is no continuity between Input N and Output N in inverter mode with no AC input L1 or L2 or N

this time I took my measurements twice cycling the power between the exercises

 

[eabyrd]

Gooood So the relay does not bond N-G in battery mode.
You know this should not cause any problem in line mode. Only when inverter goes to grid separated battery mode.
Maybe a check here between incoming G - outgoing G and incoming N - outgoing N ?

well Houston I may have screwed up my prior test, because this time I did get continuity between output N - G. There is a direct connection between Input G, output G, and Chassis. So continuity between all and output N as well. There is no continuity between Input N and Output N in inverter mode with no AC input L1 or L2 or N

this time I took my measurements twice cycling the power between the exercises

 

[mrzed001]

Gooood So the relay does not bond N-G in battery mode.
You know this should not cause any problem in line mode. Only when inverter goes to grid separated battery mode.
Maybe a check here between incoming G - outgoing G and incoming N - outgoing N ?

well Houston I may have screwed up my prior test, because this time I did get continuity between output N -There is a direct connection between Input G, output G, and Chassis. So continuity between all and output N as well. There is no continuity between Input N and Output N in inverter mode with no AC input L1 or L2 or N

this time I took my measurements twice cycling the power between the exercises

Possible that the inverter was not started ?

OK so that is sure in battery mode the relay connects the N-G.
The other part is the hard one ... in line mode the relay disconnects the N-G ?

 

[eabyrd]

More likely a misplaced probe. Anyway moving on. Why can‘t I just exclude neutral or ground (either should work since they’re bonded) from AC input, and disconnect the battery. If I have continuity across output N - G then I know the inverter’s bond remains in place

 

[mrzed001]

More likely a misplaced probe. Anyway moving on. Why can‘t I just exclude neutral or ground (either should work since they’re bonded) from AC input, and disconnect the battery. If I have continuity across output N - G then I know the inverter’s bond remains in place

Neutral in you need, that is sure.
So if you do not connect the incoming ground ... nothing after that would be grounded so caution ...
If inverter starts then out N-G should not be bonded. And I am not sure inverter will start ... not sensing N-G bond on the AC in side. Not sure but possible.

 

[eabyrd]

Neutral in you need, that is sure.
So if you do not connect the incoming ground ... nothing after that would be grounded so caution ...
If inverter starts then out N-G should not be bonded. And I am not sure inverter will start ... not sensing N-G bond on the AC in side. Not sure but possible.

There is nothing after it right now ?. It does start without a ground connected and delivers power as a line connection, or in battery mode. I will reset the wiring and check the output N - G connectivity in Line mode (off), in utility first mode (UPS / bypass) both with and without utility power, and in battery first (inverter mode). But now I gotta do some work

stay tuned

 

[mrzed001]

There is nothing after it right now ?. It does start without a ground connected and delivers power as a line connection, or in battery mode. I will reset the wiring and check the output N - G connectivity in Line mode (off), in utility first mode (UPS / bypass) both with and without utility power, and in battery first (inverter mode). But now I gotta do some work

stay tuned

Good testing

 

[eabyrd]

Good testing

I appreciate your insights and ideas on tests to perform. I will share my results

 

[eabyrd]

@mrzed001 here are my test results. Looks to me like the thing is broken. The continuity between AC Out Neutral and Ground was inconsistent, sometimes reading 0.00 or some modest value like 19, sometimes being OL in the same mode. More worrying was the varying voltages that appeared between N and G on both sides of the AC Block. Maybe those are normal, but if so that's news to me.

Ed

 

[mrzed001]

@mrzed001 here are my test results. Looks to me like the thing is broken. The continuity between AC Out Neutral and Ground was inconsistent, sometimes reading 0.00 or some modest value like 19, sometimes being OL in the same mode. More worrying was the varying voltages that appeared between N and G on both sides of the AC Block. Maybe those are normal, but if so that's news to me.

Ed

Nice test
This resistance cycling thing I do not understand. It goes from 110 to 0 and stays there; or back to 110; or any random number between them ?
A relay can only switch on/off, so it should stay OL or a low value.

I am lazy to read back, what is the error the unit gives when you try to start it AC in connected ?

 

[eabyrd]

There are no error codes, it just cycles on and off, on and off. The fault light comes on, but no codes

 

[mrzed001]

There are no error codes, it just cycles on and off, on and off. The fault light comes on, but no codes

That is strange
Without AC in connected it starts.
What did MPP say about it ?

Let me guess (again lazy to read back ) ... asked for your wiring diagram ?
I remember there was an issue with the common neutral VS separated neutral. Where I said separated neutral is the good solution.
But in current case there is nothing connected on the AC out when the inverter do not start.
So there can not be a neutral backflow problem.
So ... inverter is faulty.
The protected load DC/AC inverter part works.
But the grid-tie and PV/AC part does not (do not know if AC charger is OK too).