diy solar

diy solar

Power Outage Causes EG4 Operation Faults

If the pedestal has a breaker then you would disconnect the wires feeding into the main panel and then reconnect those wires directly to the AC Input of the Inverter. In this configuration the main panel will get power from the inverter and batteries when the grid is down BUT you will not be back feeding the grid since the Inverter has a disconnect on the AC Input terminals.
This is probably the easiest solution for now. I don't have much that needs power.

Thank you everyone for your clarification.

Curious if you'd recommend a switch to bypass the Inverter in case of some electronic failure. If so, what would you use? I'm concerned about my house being without power because of that.
 
in our place we installed a Reliance manual transfer switch for our critical loads. We then wired the circuits we wanted to run on it and the GEN side of the transfer switch was set to the output of the Inverter. We just keep those loads always set to GEN ... power goes out, we don't know about it till we use non-critical loads. In an event the inverter needs work, batteries need work, or stuff just doesn't work, then we can switch the transfer switch to LINE which is hooked up to the GRID. I have done this many times as I am working out some kinks in our system.

If GRID power goes out we have two choices, run on battery backup alone, or go out and start the generator which is hooked up to a mechanical lockout breaker along with the GRID to feed the input side of the inverter; we don't have solar at this location .... new place will. First image here has layout https://diysolarforum.com/threads/i...or-battery-bank-and-sma-si-load-center.52698/

Since both GRID / GEN go to inverter input we have to set our BMS to a lower charging amperage when we do use the GEN to keep it under the rated loads of the generator as the GRID doesn't have those issues. But that is a couple clicks and your done type setting so no worries for us.
 
This is probably the easiest solution for now. I don't have much that needs power.

Thank you everyone for your clarification.

Curious if you'd recommend a switch to bypass the Inverter in case of some electronic failure. If so, what would you use? I'm concerned about my house being without power because of that.
A transfer switch allows complete bypass of the inverter and the circuits that feed it. It will require a new main panel with 2 breakers that is N-G bonded, one breaker to the inverter, the other breaker to the transfer switch. The other input of the transfer switch is fed by the inverter. The output from the transfer switch would go to your current main panel that will have N-G bond removed.

One thing that needs to be determined is if the inverter has internal dynamic bonding. Original units in the EG4 inverters had dynamic bonding under inverter power. If you have a dynamic bonding inverter, then the transfer switch required would be a 3 pole double throw that allows neutral to be switched. If the inverter doesn't have dynamic bonding, then a simple 2 pole transfer switch can be used that only switches hots. The way to determine whether the unit has dynamic bonding is to check continuity on the inverter output N to G when it is off with no wires connected. If continuity exists, the unit has dynamic bonding. If there isn't continuity, then you may have to install a common neutral (which may not be supported but this may change shortly).

I highly recommend waiting until Signature Solar releases a new video about how to approach bonding with these units. It will save a headache or two, possibly some money and be a system that is safe. My understanding is your model does not have a ground screw in the inverter for dynamic bonding. This creates a dilemma on how to properly bond these units if AC input is used with N-G bond in a main panel ahead of the inverter and a subpanel after the inverter.
 
By system design, the AC feed from the main panel goes to the Inverter and relies upon the Inverter's AC Transfer Relay to detect whether it's receiving AC from the Main Panel / Grid. It if doesn't, it switches over to the Battery for power and sends that back to the Main Panel to power house items during emergency, etc.

How would this be unsafe? An explanation would really help to understand why the system failed as it did. Again, this is new territory for the electrician and if I could provide data, it would help to propose a strategy similar to what you're recommending which is to put the Inverter between the pedestal and the Main Panel. We're not creating a new Sub Panel to feed the house.

Thank you!

If you and your "electrician" don't understand why the above is wrong, no one here can help you. If linemen were working outside, you are lucky you didn't kill someone.

Immediately unwire what you have done here and hire someone else to do a backup install for you.

This situation is exactly why when I work on industrial equipment I don't trust anyone and check everything myself. JFC.
 
By system design, the AC feed from the main panel goes to the Inverter and relies upon the Inverter's AC Transfer Relay to detect whether it's receiving AC from the Main Panel / Grid. It if doesn't, it switches over to the Battery for power and sends that back to the Main Panel.

Please see the original post for the fault codes and description.
Sir, you are very lucky that you did not electrocute somebody wiring it that way you never send it back to the main panel. You should always send it to a subpanel....
 
Is the neutral/ground bond at the pedestal?
Is your home hardwired to the pedestal?
Yes, my home is hardwired to the pedestal and the NG Bond is at the pedestal.
As to your other question, I called him an expert because compared to me he is. He's just never done solar before. He's a retired electrician that ran his own company that's done some pretty major jobs in our county. He's also my neighbor. I've told him about the possibility of back feeding into the system if we feed the main panel from the inverter but he seems to be resistant to believing that could happen based on something esoteric about electricity that is beyond my knowledge level.
 
Last edited:
A transfer switch allows complete bypass of the inverter and the circuits that feed it. It will require a new main panel with 2 breakers that is N-G bonded, one breaker to the inverter, the other breaker to the transfer switch. The other input of the transfer switch is fed by the inverter. The output from the transfer switch would go to your current main panel that will have N-G bond removed.

One thing that needs to be determined is if the inverter has internal dynamic bonding.
Yes, it's internally bonded.

What about putting an ATS after the pedestal and Inverter AC Out, before the Main Panel? The ATS should prevent back feeding and the Inverter will only output current from the Battery when it's internal switch detects that the Main Panel is no longer feeding AC In.
 
Yes, it's internally bonded.

What about putting an ATS after the pedestal and Inverter AC Out, before the Main Panel? The ATS should prevent back feeding and the Inverter will only output current from the Battery when it's internal switch detects that the Main Panel is no longer feeding AC In.
There are two ways shown in this block diagram below.
Whole home panel.png
 
Yes, it's internally bonded.

How do you know? Did you measure continuity between N and G inverter output, inverter off and nothing connected to AC output. and input?

What about putting an ATS after the pedestal and Inverter AC Out, before the Main Panel? The ATS should prevent back feeding and the Inverter will only output current from the Battery when it's internal switch detects that the Main Panel is no longer feeding AC In.
My system is wired like this. https://diysolarforum.com/threads/mpp-lv6548-ground-neutral-safety.36241/post-462580

There is also a guide in the Resources section, hit the Download button at the top right after reading the page. https://diysolarforum.com/resources/grounding-and-bonding-for-mpp-lv6548-and-eg4-6500ex-48.275/
 
I've told him about the possibility of back feeding into the system if we feed the main panel from the inverter but he seems to be resistant to believing that could happen based on something esoteric about electricity that is beyond my knowledge level.
I guess its esoteric enough that I don't get it either.
aio = all in one
poco = power company
Can anyone here share the occult knowledge that explains how aio.ac.in, aio.ac.out and the poco can all be connected to the same distribution panel without any external means of mutual exclusivity?

If we simplify the environment by getting rid of the poco I would still expect a power loop between aio.ac.in and aio.ac.out.
I would not be surprised if the bypass relay were to cycle on and off rapidly.
 
Last edited:
Can anyone here share the occult knowledge that explains how aio.ac.in, aio.ac.out and the poco can all be connected to the same distribution panel without any external means of mutual exclusivity?
Could use a split bus panel ?
Top of the panel is feed by mains and has mains only loads. There would be a breaker here which goes into AIO AC IN.
Then AIO AC Out goes to the bottom part of the panel where any loads which should be on the AIO are found.

Otherwise ... you need the POCO to go to an external, likely expensive ATS, prior to suppling the main lugs on the main panel. But again, as pointed out by several people so far. Most people will put the AIO inline with the main service panel to not have this issue POCO > AIO > MAIN PANEL as Mattb4 has drafted above.
 
Could use a split bus panel ?
Top of the panel is feed by mains and has mains only loads. There would be a breaker here which goes into AIO AC IN.
Then AIO AC Out goes to the bottom part of the panel where any loads which should be on the AIO are found.

Otherwise ... you need the POCO to go to an external, likely expensive ATS, prior to suppling the main lugs on the main panel. But again, as pointed out by several people so far. Most people will put the AIO inline with the main service panel to not have this issue POCO > AIO > MAIN PANEL as Mattb4 has drafted above.
I would call that an external means of mutual exclusion.
Serves me right for being rhetorical.
 
I would call that an external means of mutual exclusion.
Serves me right for being rhetorical.
Solved it with MSPaint. Really big unicorn diodes and the switch of the ATS on the POCO side of the equation ... POCO / AC IN / AC OUT ... one panel ;)
1676992545801.png
 
Last edited:
Again it's an external means.
What if you replace your "main" panel with a 200amp rated ATS Load center panel with 24 slots? ;) Technically it is now all in one panel acting as the "main" ... mind you they are not cheap >__< Generac makes some nice ones that fall into this category as they do whole house systems.
 
Um, I think just wiring it with all the desired loads in the sub panel would make much more sense than trying to get a bunch of extra transfer switches and whatnots. 90% of my circuits are moving into a new panel right next to my current main panel, for example. Central air, electric stove, and dryer are staying in the main panel, almost everything else is going into the sub (though I'm considering a way to enable inverter power back to the AC or dryer for times when the sun is blasting bright and the batteries are charged, so I can use the extra sun to cool the house or dry some clothes even if the grid is down).

Are these panels and inverter(s) next to each other in the same space or halfway across the house?
 
One thing that needs to be determined is if the inverter has internal dynamic bonding. Original units in the EG4 inverters had dynamic bonding under inverter power. If you have a dynamic bonding inverter, then the transfer switch required would be a 3 pole double throw that allows neutral to be switched. If the inverter doesn't have dynamic bonding, then a simple 2 pole transfer switch can be used that only switches hots. The way to determine whether the unit has dynamic bonding is to check continuity on the inverter output N to G when it is off with no wires connected. If continuity exists, the unit has dynamic bonding. If there isn't continuity, then you may have to install a common neutral (which may not be supported but this may change shortly).
So, what I read was that this unit is internally bonded. But I did what you said, and checked the continuity of neutral to ground at the output, and there is none. I'm including pics of the connections I made (copper posts in neutral and ground of AC Out), as well as the model number (EG4 6000EX) and build date (10/2022).

I read the info you directed me to, but I'm concerned that because I'm running a 6000EX that it won't apply and I'll have to check the different scenarios to see where bonds are being made, unless I can assume that the unit never makes a neutral ground bond based on this test. Thoughts?

And all of this is to determine the correct transfer switch if I understand correctly. I really want an automatic transfer switch. This is because if my power goes out and I'm out of town, I won't be there to manually switch the AC back on. My use case at this time is UPS motivated.
 

Attachments

  • manf_date.jpg
    manf_date.jpg
    94.8 KB · Views: 5
  • model.jpg
    model.jpg
    156.4 KB · Views: 5
  • neutral_ground.jpg
    neutral_ground.jpg
    76.9 KB · Views: 5
  • no_continuity.jpg
    no_continuity.jpg
    97.6 KB · Views: 5
Um, I think just wiring it with all the desired loads in the sub panel would make much more sense than trying to get a bunch of extra transfer switches and whatnots. 90% of my circuits are moving into a new panel right next to my current main panel, for example. Central air, electric stove, and dryer are staying in the main panel, almost everything else is going into the sub (though I'm considering a way to enable inverter power back to the AC or dryer for times when the sun is blasting bright and the batteries are charged, so I can use the extra sun to cool the house or dry some clothes even if the grid is down).

Are these panels and inverter(s) next to each other in the same space or halfway across the house?
The main is attached to the outside of the house, about 30 feet from my garage. My garage houses the Inverter. There's enough room on the wall to add another inverter which I intend to when funds are available. There's a small Sub Panel about 6 feet from the Inverter inside the garage. This is manufactured home, so building a new panel into the side of the house may be tricky.
 
So, what I read was that this unit is internally bonded. But I did what you said, and checked the continuity of neutral to ground at the output, and there is none. I'm including pics of the connections I made (copper posts in neutral and ground of AC Out), as well as the model number (EG4 6000EX) and build date (10/2022).

The unit does not bond N-G.

Build date is correct for shipped without bonding screw.

I read the info you directed me to, but I'm concerned that because I'm running a 6000EX that it won't apply and I'll have to check the different scenarios to see where bonds are being made, unless I can assume that the unit never makes a neutral ground bond based on this test. Thoughts?

It won't make a bond.

And all of this is to determine the correct transfer switch if I understand correctly. I really want an automatic transfer switch. This is because if my power goes out and I'm out of town, I won't be there to manually switch the AC back on. My use case at this time is UPS motivated.
There is some discussion the 6000EX has an isolation transformer internally. One aspect is that if this is true, if you ever want to bypass the inverter entirely for repairs or maintenance, you will need a 3 pole transfer switch with a neutral in order to supply neutral to your subpanel on grid power. On this unit, the input N is not tied to the output N due to the isolation transformer. Common neutral can't be used due to the isolation transformer.

If you had the 6500EX (late model with latest firmware or earlier model with bonding screws removed and latest firmware), it would be more suited for your UPS scenario. If you want an automatic transfer switch, it should have neutral switching so you can supply grid neutral. That requires a 3 pole. Also, when it comes to bonding N-G, you will need to have N-G bonded in the subpanel. This creates a problem on grid backup power, you will have 2 N-G bonds in the system. You will need to have some type of dynamic switching for N-G bond in the subanel.

None of this is really required to some extent as the 6000EX has an isolation transformer. As long as it has power on the 2 hots coming in, it will create the N on output, just like the transformer from the utility grid. That is why N-G bonding is done at the subpanel. You should not need a transfer switch unless you want the ability to bypass the inverter entirely.
 
The unit does not bond N-G.

Build date is correct for shipped without bonding screw.



It won't make a bond.


There is some discussion the 6000EX has an isolation transformer internally. One aspect is that if this is true, if you ever want to bypass the inverter entirely for repairs or maintenance, you will need a 3 pole transfer switch with a neutral in order to supply neutral to your subpanel on grid power. On this unit, the input N is not tied to the output N due to the isolation transformer. Common neutral can't be used due to the isolation transformer.

If you had the 6500EX (late model with latest firmware or earlier model with bonding screws removed and latest firmware), it would be more suited for your UPS scenario. If you want an automatic transfer switch, it should have neutral switching so you can supply grid neutral. That requires a 3 pole. Also, when it comes to bonding N-G, you will need to have N-G bonded in the subpanel. This creates a problem on grid backup power, you will have 2 N-G bonds in the system. You will need to have some type of dynamic switching for N-G bond in the subanel.

None of this is really required to some extent as the 6000EX has an isolation transformer. As long as it has power on the 2 hots coming in, it will create the N on output, just like the transformer from the utility grid. That is why N-G bonding is done at the subpanel. You should not need a transfer switch unless you want the ability to bypass the inverter entirely.
I finally got a reply back from Signature Solar:
"From my understanding the EG4 6kW does have a NGB screw so it is bonded. That being said, it also has a relay that will unbond it if AC power is connected to the input(because it assumes that the AC is bonded at the service panel), and will bond it again when no AC is present."
 
  • Like
Reactions: Zwy
I finally got a reply back from Signature Solar:
"From my understanding the EG4 6kW does have a NGB screw so it is bonded. That being said, it also has a relay that will unbond it if AC power is connected to the input(because it assumes that the AC is bonded at the service panel), and will bond it again when no AC is present."
Good information, thank you for sharing it.
 

diy solar

diy solar
Back
Top