diy solar

diy solar

Help with wiring/grounding

Daniel Powers

New Member
Joined
May 6, 2022
Messages
6
Hello! I'm wondering if someone can help me with my wiring schematic for a hybrid (non-grid-tied) system. Please see the image below.

I have a Growatt inverter outputting 240/120 volt split phase power, and I have a sub panel where I will have my critical load circuits. My questions have to do with proper grounding. In the image you can see 5 points which I've labeled A through E. I suspect that points A and B should both be connected to C (the grounding bar in the main panel), but I'm not sure about that nor how to ground points D or E.

Thanks to everyone for your help!
-Daniel

Solar System Wiring.png
 

Attachments

  • Solar System Wiring.png
    Solar System Wiring.png
    61.5 KB · Views: 4
BTW: Grounding is *not* a simple topic. You may want to review this series of papers from the resource section:
 
BTW: Grounding is *not* a simple topic. You may want to review this series of papers from the resource section:
Awesome, thanks!

I'll definitely give this a read.
 
I'm using the Growatt SPF 8000T DVM-MPV.
Ohhhhhh..... there is another thread where we are trying to figure out the Grounding and bonding for this line of inverters..... and not having a lot of luck. We have figured out what the inverter does, but we are not sure if there is a way to properly ground and bond the system that GW will support. A forum member from EG4 is trying to get an answer from Growatt.... but we have not heard back from them on it.

The thread is now 10 or 11 pages long, but you do not need to follow the whole thing.

Here is a good starting point:


This is the post where @RichardfromEG4 confirmed what the architectural model of the inverter is....
 
While we are waiting to figure out the bonding problem, here is some feedback on your drawing.

* on that inverter there is no neutral-in. It only takes the two hots.
* On that inverter there is a single ground connection where the AC-in and AC-Out is connected. There should be a ground wire from that connection back to the main panel. There should not be another connection to ground or earth ground.
* In your main panel, you almost certainly have a neutral-ground bond and a connection to earth ground.
* The sub-panel should have exactly one ground wire between its grounding bus and the main panel grounding bus.
* The heater Equipment Grounding Conductor should be tied to the Grounding conductor of the Romex going back to the sub-panel... or it can be tied to the ground bar of the main panel. If you tie it to the main panel ground bar, do not use the grounding conductor in the Romex going back to the sub-panel.
* The frames of the solar panels should be tied to the main panel grounding bus bar.

1652153344087.png
 
Yes, I'm still hoping to hear an 'official' answer from our supplier. The real issue when it comes to grounding is there are so many differing opinions on it and so many small things that can affect those opinions. The likely answer will be 'it depends' as in - it depends on the use (on grid or off grid or mix), where in your electrical system it's located (main panel, sub panel, direct to a device), and a number of other small changes that could totally affect the answer. So to say 'you want to do x if you use this product' is a very hard statement to make. The safer answer is that one should research their use case and make the determination based on that, knowing now how the device works. That being said - I'm still waiting for a response to see what they might say.
 
While we are waiting to figure out the bonding problem, here is some feedback on your drawing.

* on that inverter there is no neutral-in. It only takes the two hots.
* On that inverter there is a single ground connection where the AC-in and AC-Out is connected. There should be a ground wire from that connection back to the main panel. There should not be another connection to ground or earth ground.
* In your main panel, you almost certainly have a neutral-ground bond and a connection to earth ground.
* The sub-panel should have exactly one ground wire between its grounding bus and the main panel grounding bus.
* The heater Equipment Grounding Conductor should be tied to the Grounding conductor of the Romex going back to the sub-panel... or it can be tied to the ground bar of the main panel. If you tie it to the main panel ground bar, do not use the grounding conductor in the Romex going back to the sub-panel.
* The frames of the solar panels should be tied to the main panel grounding bus bar.

View attachment 94079

Wow, you're awesome, FilterGuy!


I'll take your points one by one here:

1. On that inverter there is no neutral-in. It only takes the two hots.
-- Yes it's my mistake that I drew a neutral line between the main breaker panel & the inverter. It should've just been the 2 hot lines and a ground. I'll make that change to the drawing.

2. On that inverter there is a single ground connection where the AC-in and AC-Out is connected. There should be a ground wire from that connection back to the main panel. There should not be another connection to ground or earth ground.
-- Points D & E weren't meant to represent an actual earth grounding at those locations, just that there are grounding terminals on those devices. I guess I should've been more clear in my drawing. Sorry about that!

3. In your main panel, you almost certainly have a neutral-ground bond and a connection to earth ground.
-- Yes, there is a neutral/ground bond in my main panel with a connection to earth. I'll add that to my drawing as well.

4. & 5. The sub-panel should have exactly one ground wire between its grounding bus and the main panel grounding bus.
The heater Equipment Grounding Conductor should be tied to the Grounding conductor of the Romex going back to the sub-panel... or it can be tied to the ground bar of the main panel. If you tie it to the main panel ground bar, do not use the grounding conductor in the Romex going back to the sub-panel.
-- Hmm OK. I was thinking I could leave the ground wire for all of my critical load circuits connected to the grounding bus on the main panel, and then also connect the grounds on both ends of the 14/2 Romex to their respective grounding busses as well. That would save me from having to remove them all from the main panel's bus & join them to their respective feeder/extension wires to the sub panel. (i.e. connect points A & B to C. But this would provide multiple paths back to the main panel's grounding bus via all of the other circuits in the panel connected this way, which aren't shown in the diagram). So I agree with you. One properly sized grounding cable connecting the 2 grounding busses is a better solution.

6. The frames of the solar panels should be tied to the main panel grounding bus bar.
-- OK perfect.

I'll make changes to my drawing & upload it here shortly.

Thanks again, FilterGuy. You've been a huge help!
-Daniel
 
Yes, I'm still hoping to hear an 'official' answer from our supplier. The real issue when it comes to grounding is there are so many differing opinions on it and so many small things that can affect those opinions. The likely answer will be 'it depends' as in - it depends on the use (on grid or off grid or mix), where in your electrical system it's located (main panel, sub panel, direct to a device), and a number of other small changes that could totally affect the answer. So to say 'you want to do x if you use this product' is a very hard statement to make. The safer answer is that one should research their use case and make the determination based on that, knowing now how the device works. That being said - I'm still waiting for a response to see what they might say.
Thanks, Richard. I look forward to hearing what they have to say!
 
OK, I've updated my drawing to better reflect the actual scenario & to correct a couple of mistakes I made in the original. I've also included the solution given by FilterGuy and made it an animated PNG so you can see the "before & after".

Any further comments or suggestions would still be welcome. Thanks!

Solar System Wiring (r1_solution_animated).png
 
-- Hmm OK. I was thinking I could leave the ground wire for all of my critical load circuits connected to the grounding bus on the main panel, and then also connect the grounds on both ends of the 14/2 Romex to their respective grounding busses as well. That would save me from having to remove them all from the main panel's bus & join them to their respective feeder/extension wires to the sub panel. (i.e. connect points A & B to C. But this would provide multiple paths back to the main panel's grounding bus via all of the other circuits in the panel connected this way, which aren't shown in the diagram). So I agree with you. One properly sized grounding cable connecting the 2 grounding busses is a better solution.
It is fine to connect the Equipment Grounding conductors for all the critical loads to the ground bus of the main panel..... just don't also connect them to the ground bus of the critical loads panel. Either way, you will still need a single connection to the ground bus of the critical loads panel. Even if the ground bus of the critical loads panel is not used for anything else, it must be tied to the equipment grounding system to properly ground the case of the critical loads panel.

In your new drawing, The connection between C & D should be removed. The case ground screw should be left unused. On all inverters I have ever examined, the AC grounding screws are also tied to the case of the inverter. Consequently, hooking the ground to the AC-in is all you need. Adding the connection between C&D does not improve grounding but it does create a ground loop. (A ground loop through an inverter is just begging for RFI issues).
 
Yes, I'm still hoping to hear an 'official' answer from our supplier. The real issue when it comes to grounding is there are so many differing opinions on it and so many small things that can affect those opinions. The likely answer will be 'it depends' as in - it depends on the use (on grid or off grid or mix), where in your electrical system it's located (main panel, sub panel, direct to a device), and a number of other small changes that could totally affect the answer. So to say 'you want to do x if you use this product' is a very hard statement to make. The safer answer is that one should research their use case and make the determination based on that, knowing now how the device works. That being said - I'm still waiting for a response to see what they might say.
Richard, I really appreciate what you are doing. You are not the manufacturer of the unit but you are using your contacts to try to help all of us out. Thank You!

I agree that there are many different situations, but this situation and the situation on the other thread are the most basic and the most common:


1652198173383.png

Based on the manual, the above diagram is how a reasonable person would wire things. However, the output is left floating when in invert mode. It is not unreasonable to expect a specific answer on how this case should be properly wired.

I *hope* the answer comes back that the correct solution is this:

1652198717270.png
Tieing the neutral of the Critical Load box back to the neutral of the Main Panel would ensure the output is always bonded. We know that some people are using it this way, but is it safe and supported?

1652199040784.png
 
So the Inverter is a 230 volt only in but no Neutral in, and its outputting a Neutral? It can only be a Neutral if its grounded and per the NEC that ground must be in the Main house panel. So the wire that is to be the panel neutral must go back to the Main system ground and be attached there and then back.
 
So the Inverter is a 230 volt only in but no Neutral in, and its outputting a Neutral? It can only be a Neutral if its grounded and per the NEC that ground must be in the Main house panel. So the wire that is to be the panel neutral must go back to the Main system ground and be attached there and then back.
We have confirmation that this is the high-level model of the inverter:
1652229696905.png
While in pass-through, the main bonding jumper in the main breaker panel that is feeding the AC in provides the low impedance path for a ground fault. What we are trying to get info about is bonding when in invert mode. As I said in the previous post, if the unit is wired as the instructions would imply, the output would be floating when in inverter mode. I have studied this a lot and the only reasonable way I see of doing it would be to tie the Neutral output back to the neutral in the main breaker box like I show in my previous post. We know this works in the typical case because we have users that are doing it. What we don't know is if there are corner cases where this would be a problem and if Growwatt will support it. (I guess the alternative is to let the output float while in invert mode. Floating circuits can be made safe, but I really do not like them.)

Note that in Passthrough mode with the neutral tied back to the main breaker box, the output transformer will act as an auto-transformer. This is OK... but what happens if the load balance on the two phases is way off and the output transformer tries to balance them? (This is the corner-case I am concerned about)
 
We have confirmation that this is the high-level model of the inverter:
View attachment 94198
While in pass-through, the main bonding jumper in the main breaker panel that is feeding the AC in provides the low impedance path for a ground fault. What we are trying to get info about is bonding when in invert mode. As I said in the previous post, if the unit is wired as the instructions would imply, the output would be floating when in inverter mode. I have studied this a lot and the only reasonable way I see of doing it would be to tie the Neutral output back to the neutral in the main breaker box like I show in my previous post. We know this works in the typical case because we have users that are doing it. What we don't know is if there are corner cases where this would be a problem and if Growwatt will support it. (I guess the alternative is to let the output float while in invert mode. Floating circuits can be made safe, but I really do not like them.)

Note that in Passthrough mode with the neutral tied back to the main breaker box, the output transformer will act as an auto-transformer. This is OK... but what happens if the load balance on the two phases is way off and the output transformer tries to balance them? (This is the corner-case I am concerned about)
@wmgeorge The GW transformer creats neutral.

" ... but what happens if the load balance on the two phases is way off and the output transformer tries to balance them? (This is the corner-case I am concerned about) ... ".
The part in bold also concerns me. What precautions has GW taken to ensure safe operation if the GW's transformer, or some other part of the GW, is overloaded trying to balance the loads on the utility's street transformer.
 
@wmgeorge The GW transformer creats neutral.

" ... but what happens if the load balance on the two phases is way off and the output transformer tries to balance them? (This is the corner-case I am concerned about) ... ".
The part in bold also concerns me. What precautions has GW taken to ensure safe operation if the GW's transformer, or some other part of the GW, is overloaded trying to balance the loads on the utility's street transformer.
Wrong. The unit creates a center tap to reference the 230 volts to create 115 volts. The wire is only a Neutral when connected to the system Ground at the Main panel. It must not be grounded to a panel otherwise its a Code violation. There is also a functional ground connection that is allowed but I am not sure it applies here.

Can someone post a picture of the UL approval on the unit label?

 
Last edited:
Wrong. The unit creates a center tap to reference the 230 volts to create 115 volts. The wire is only a Neutral when connected to the system Ground at the Main panel. It must not be grounded to a panel otherwise its a Code violation. There is also a functional ground connection that is allowed but I am not sure it applies here.

Can someone post a picture of the UL approval on the unit label?

You are correct, It is not technically a neutral if it is not a grounded conductor. However, I do not understand what point you are trying to make.

We are trying to understand how the manufacturer expects the system to be set up to be properly grounded and bonded.
One solution that we are considering (if GW Supports it) is to tie the center tap back to the neutral of the main panel. This would make the center tap a grounded conductor (Neutral) for both pass-through and inverter modes. The point @Kornbread and I are making is that in this configuration, the output transformer will act as an auto-transformer and try to balance the two phases when in pass-through mode.

Also, there are two purposes for bonding Neutral to Ground.
1) Create a low impedance path that allows a ground fault to be cleared by the breakers
2) Not allow the system to float to an arbitrarily high voltage in reference to earth ground.
Both of the objectives are met in pass-through mode even if the center-tapped conductor is not tied to earth ground.

The two purposes of N-G bonding are not met when the system is in inverter mode.
 
You are correct, It is not technically a neutral if it is not a grounded conductor. However, I do not understand what point you are trying to make.
It can not be called a Neutral nor its it one, if its not grounded. It can be the center tap of some transformers and be neutral in respect to the other outputs but can't be called a true Neutral unless its connected to the system equipment ground. Very simple NEC Requirement and its been Code forever. It is not a manufacturer choice. I asked about UL approval not Chinese but US since that is where its being sold.
 
Last edited:
It can not be called a Neutral nor its it one, if its not grounded. It can be the center tap of some transformers and be neutral in respect to the other outputs but can't be called a true Neutral unless its connected to the system equipment ground. Very simple NEC Requirement and its been Code forever. It is not a manufacturer choice. I asked about UL approval not Chinese but US since that is where its being sold.
It's not UL listed.

Second, this functions as an autotransformer if it is similar to the 12K GW.

Some of the discussion concerning autotransformers takes place in this thread. https://diysolarforum.com/threads/a...-all-in-ones-how-do-they-handle-ground.34429/
 
Back
Top