Problems with Eg4 6500ex buyer beware.

[AntronX]

You need N-G bond, the LV6548 switches neutral.
The 6500EX does not switch neutral with the firmware update SS pushes for correcting their problem.

Ok, so LV6548 auto bonds NG internally when supplying critical loads panel and 6500EX does not ? Chances are if you force unbond LV6548 it will show same symptoms as 6500EX.

 

[timselectric]

You're really grasping Tim. It's not the wiring around the light, it's not the light, it's not the house wiring.

The " glowing led light with the switch off" occurred when I removed the N-G bond I had added in the loads panel as a test to see if indeed it is a bonding circuit problem.

Watch the videos again in this post. https://diysolarforum.com/threads/problems-with-eg4-6500ex-buyer-beware.54597/post-793269


Been thru all of it since March. Very methodical, it's not the wiring. It is the 6500EX inverters. Installing the LV6548's eliminated the problems.

Signature Solar might have a pair of heavily discounted inverters you can purchase. They can simply put your name and address on the RMA call tag when these are picked up. It will be an educational experience for you.

If it's in the inverter (which is still possible). The inverter is creating a N/G bond, that it shouldn't be.
Did they send you back the same units? Or replacements?
I still have to lean towards the house wiring. Because with the switch off. They should not have power. No matter what the power source is.
Maybe the reason it's so hard to find the issue. Is because you have two issues.

 

[Zwy]

Ok, so LV6548 auto bonds NG internally when supplying critical loads panel and 6500EX does not ? Chances are if you force unbond LV6548 it will show same symptoms as 6500EX.

Yes, if I remove the N-G bonding screw in the LV6548, I would get the glowing light problem with the switch turned off. I'm quite certain of that.

 

[AntronX]

What is the voltage drop when voltage is 7V on 180 feet of #4 wire?

It's precisely 7V. System voltage does not matter. When looking at only this part of the circuit only ohms law R=V/A applies. If you are seeing 7.5Vrms across 180' of #4 wire then by ohms law, regardless of total system voltage, it will require 168A to induce that voltage across it assuming no additional resistance in the wire run. This points to possible loose connection somewhere in that circuit or measurement error.

 

[Zwy]

If it's in the inverter (which is still possible). The inverter is creating a N/G bond, that it shouldn't be.

Inverter is not bonding, otherwise I would not see the glowing light with the switch off. The N-G bonding screw in the loads panel shows there wasn't a bond because once it was removed, the light started glowing while switched off.

Did they send you back the same units? Or replacements?

Replacement manufactured 10/22 which is the same date the member who lost his TV after replacements.

I still have to lean towards the house wiring. Because with the switch off. They should not have power. No matter what the power source is.
Maybe the reason it's so hard to find the issue. Is because you have two issues.

Ever see lightning directly hit a house on the neutral? I did, I was in it at the time. The scorch marks around the receptacles that had nothing plugged in were certainly interesting. Insurance company inspected it, said it was on the neutral. Anything plugged in was fried even though the switches were off.

I never would have thought it could backfeed on the neutral due to the ground. This isn't the house by the way.

 

[AntronX]

Are you certain that you actually have #4 ground wire going between house panel and NG bond at the pole? Could it be 2 unbonded ground rods at each end instead? That would explain this high resistance you are measuring with that 7.5Vrms reading.

 

[Zwy]

It's precisely 7V. System voltage does not matter. When looking at only this part of the circuit only ohms law R=V/A applies. If you are seeing 7.5Vrms across 180' of #4 wire then by ohms law, regardless of total system voltage, it will require 168A to induce that voltage across it assuming no additional resistance in the wire run. This points to possible loose connection somewhere in that circuit or measurement error.

I'm measuring potential between N and G. Not voltage drop. G should be at zero reference and N has potential of 7V. https://www.tutorialspoint.com/difference-between-voltage-drop-and-potential-difference#:~:text=Voltage drop occurs across a,points in an electric circuit.&text=The main cause of voltage,is EMF (Electromotive Force).

One more thing, "Potential difference can exist even without the flow of current in the circuit"

.Are you certain that you actually have #4 ground wire going between house panel and NG bond at the pole? Could it be 2 unbonded ground rods at each end instead? That would explain this high resistance you are measuring with that 7.5Vrms reading.

It's not bonded on the house side but is bonded at the pole.

 

[AntronX]

It's not bonded on the house side but is bonded at the pole.

So you don't have physical #4 wire running from house panel ground to NG bond point at the pole then? If so, we found your problem.

 

[Zwy]

So you don't have physical #4 wire running from house panel ground to NG bond point at the pole then? If so, we found your problem.

I sure do, it is green in color and runs with a red one, a black one and a white one. Take a look at the diagram again. https://diysolarforum.com/threads/problems-with-eg4-6500ex-buyer-beware.54597/post-793791

Any more guesses? It's fun but I'll be back later, have work to do.

 

[AntronX]

I sure do, it is green in color and runs with a red one, a black one and a white one.

In that case verify that each connection on that run is tight, including NG bond at the meter. You should not see 7V N-G potential in your house panel unless your neutral is carrying 167A which is unlikely. Could there be wire damage?

 

[AntronX]

...One more thing, "Potential difference can exist even without the flow of current in the circuit"

It cannot, by ohms law, if that circuit has the resistance of 180' of #4 copper wire. It's on your own circuit diagram you posted.

 

[timselectric]

Ever see lightning directly hit a house on the neutral?

Yes, that's usually how it gets in the house.

I never would have thought it could backfeed on the neutral due to the ground. This isn't the house by the way.

The ground is how it got to the neutral.
Especially if you have an overhead service feeder.

 

[BuffaloTrace]

Zwy said at some point he grounded to the homes water pipes. Can these grounds somehow intersect the neutral such as inside an appliance.

I know my homes water pipes were originally inside the foundation and grounded at the far end last outlet near the electrical main disconnect. Eventually pipes were changed and re routed through the attic.... 1st owners changed the whole grounding system without a thought or care.

 

[AntronX]

Try recreating the conditions for that 7V to appear across NG point and this time use AC clamp meter to measure current on neutral and ground wires going into your panel from the pole.

 

[Samsonite801]

Next, attempted the mobile firmware with one bonding screw in the inverters, flicker was about the same.

Did you happen to try a physical N-G bond jumper right at the inverter terminals (for your testing), since the new firmware holding the relay open would disable their bonding screw?

 

[Samsonite801]

Try recreating the conditions for that 7V to appear across NG point and this time use AC clamp meter to measure current on neutral and ground wires going into your panel from the pole.

Also may not hurt to sanity check your voltmeter and check with another voltmeter to confirm the 7v is for real..

 

[DIYPV]

Inverter is not bonding, otherwise I would not see the glowing light with the switch off. The N-G bonding screw in the loads panel shows there wasn't a bond because once it was removed, the light started glowing while switched off.

I don’t know if I understand the firmwares correctly, but i thought that the mobile firmware passes the neutral through when in hybrid mode, and then creates a neutral-ground bond when in solar/battery mode. Wouldn’t that fix your N-G issue if you installed the mobile firmware on one EG4 inverter?

 

[Zwy]

Try recreating the conditions for that 7V to appear across NG point and this time use AC clamp meter to measure current on neutral and ground wires going into your panel from the pole.

Kinda hard to do, the 6500EX's are no longer installed. I installed the LV6548's last night in a few hours. Boxed up the 6500EX's for their next journey to Tim's address.

But I will give you this information. The 7V potential would only appear when the lights in the house would flicker. If load was increased, the 7V potential would disappear. This video shows the 7V potential, then when the lights are turned on in basement, the 7V potential is gone and it is just noise on the scope. You have to watch the whole video to see it.

 

[Zwy]

I don’t know if I understand the firmwares correctly, but i thought that the mobile firmware passes the neutral through when in hybrid mode, and then creates a neutral-ground bond when in solar/battery mode. Wouldn’t that fix your N-G issue if you installed the mobile firmware on one EG4 inverter?

I did try the mobile firmware and you can see all the steps in this post. https://diysolarforum.com/threads/problems-with-eg4-6500ex-buyer-beware.54597/post-793853

It was the second step and I wish I had not installed any firmware. The LV6548's I bought and received in December 2021. I installed them a month ago after taking out the original EG4 6500EX's and never updated any firmware, just used the units right out of the box except I removed one bonding screw.

 

[Zwy]

Also may not hurt to sanity check your voltmeter and check with another voltmeter to confirm the 7v is for real..

Scope agreed with the multimeter by about 0.4V.