diy solar

diy solar

Troubleshooting: Neutral return with grid-tie Growatt

ZorbatheGeek

New Member
Joined
Aug 30, 2020
Messages
35
So......... I supervised the installation of a PV-Growatt-Supercapacitor system on Oct 2020 for my parents........ there is grid power as well, so although there is no buy-back it is a solar grid-tie system.
The site consists of a large lodge with a series of separate chalets for accomodation, plus garage, all connected to the main distribution panel which is now fed (via manual changeover switch) by the Growatt SPF5000 TL HMV-P. The system is 240V 50HZ single phase.
In January this 2021 I was zapped by a plug socket in one of the(newly wired) chalets, and discovered the "technician" had wired neutral and ground together in the entire socket circuit !!!! Once I had rectified this the next issue was to find what is/was causing the live neutral, with the following conclusion/questions:
  1. when selecting the manual inverter bypass (so running the household on grid) the neutral return disappears.
  2. isolating the neutral output from the inverter, with the inverter idling but with no load, there is still a live neutral. This does not seem to cause a problem (except when somebody bonds earth and neutral as I discovered)
This led me to conclude the issue lies with the inverter BUT.......
  • I've been given some advice by a professional solar installer acquaintance who says its highly unlikely to be the inverter, and that something else with the neutral circuit is causing the inverter to "make its own neutral"
  • Another local solar installer says that some inverters do produce a volatage on neutral
Has anyone had a similar experience? Are either the above pieces of advice likely to be true?

I have an electrician currently going through all the circuitry, checking grounding, replacing underground cables, connections, separating sockets from lighting circuits etc in order to rule out a problem with consumer side - if nothing resolves the issue I will call in the warranty with the PV installers.
 
So......... I supervised the installation of a PV-Growatt-Supercapacitor system on Oct 2020 for my parents........ there is grid power as well, so although there is no buy-back it is a solar grid-tie system.
The site consists of a large lodge with a series of separate chalets for accomodation, plus garage, all connected to the main distribution panel which is now fed (via manual changeover switch) by the Growatt SPF5000 TL HMV-P. The system is 240V 50HZ single phase.
In January this 2021 I was zapped by a plug socket in one of the(newly wired) chalets, and discovered the "technician" had wired neutral and ground together in the entire socket circuit !!!! Once I had rectified this the next issue was to find what is/was causing the live neutral, with the following conclusion/questions:
  1. when selecting the manual inverter bypass (so running the household on grid) the neutral return disappears.
  2. isolating the neutral output from the inverter, with the inverter idling but with no load, there is still a live neutral. This does not seem to cause a problem (except when somebody bonds earth and neutral as I discovered)
This led me to conclude the issue lies with the inverter BUT.......
  • I've been given some advice by a professional solar installer acquaintance who says its highly unlikely to be the inverter, and that something else with the neutral circuit is causing the inverter to "make its own neutral"
  • Another local solar installer says that some inverters do produce a volatage on neutral
Has anyone had a similar experience? Are either the above pieces of advice likely to be true?

I have an electrician currently going through all the circuitry, checking grounding, replacing underground cables, connections, separating sockets from lighting circuits etc in order to rule out a problem with consumer side - if nothing resolves the issue I will call in the warranty with the PV installers.
What is setting #24?
 
So......... I supervised the installation of a PV-Growatt-Supercapacitor system on Oct 2020 for my parents........ there is grid power as well, so although there is no buy-back it is a solar grid-tie system.
The site consists of a large lodge with a series of separate chalets for accomodation, plus garage, all connected to the main distribution panel which is now fed (via manual changeover switch) by the Growatt SPF5000 TL HMV-P. The system is 240V 50HZ single phase.
In January this 2021 I was zapped by a plug socket in one of the(newly wired) chalets, and discovered the "technician" had wired neutral and ground together in the entire socket circuit !!!! Once I had rectified this the next issue was to find what is/was causing the live neutral, with the following conclusion/questions:
  1. when selecting the manual inverter bypass (so running the household on grid) the neutral return disappears.
  2. isolating the neutral output from the inverter, with the inverter idling but with no load, there is still a live neutral. This does not seem to cause a problem (except when somebody bonds earth and neutral as I discovered)
This led me to conclude the issue lies with the inverter BUT.......
  • I've been given some advice by a professional solar installer acquaintance who says its highly unlikely to be the inverter, and that something else with the neutral circuit is causing the inverter to "make its own neutral"
  • Another local solar installer says that some inverters do produce a volatage on neutral
Has anyone had a similar experience? Are either the above pieces of advice likely to be true?

I have an electrician currently going through all the circuitry, checking grounding, replacing underground cables, connections, separating sockets from lighting circuits etc in order to rule out a problem with consumer side - if nothing resolves the issue I will call in the warranty with the PV installers.
From Wats247. https://i0.wp.com/watts247.com/wp-content/uploads/2020/08/PanelWiring2of2.jpg?resize=1024,574&ssl=1
 
Disabled. It says the setting will only work if if the inverter is installed with a grounding box?
Should I enable?
If you watch this video, which he didn't cover in detail setting #24 Ground Neutral Relay in the newer video, you will see how this works. It also applies to utility feed if you think about it. It appears that a switching relay for bonding needs to be installed in the breaker panel. The relay would need to close, thus engaging a contactor in the breaker panel, thus enabling bonding when under battery power but be open on generator or utility feed.
 
Why do you have a manual inverter bypass?
The growatt has an automatic transfer switch and from reading above it can switch the neutral ground bond appropriately.
 
Why do you have a manual inverter bypass?
The growatt has an automatic transfer switch and from reading above it can switch the neutral ground bond appropriately.
The manual bypass allows grid use in case of the inverter going u/s
 
If you watch this video, which he didn't cover in detail setting #24 Ground Neutral Relay in the newer video, you will see how this works. It also applies to utility feed if you think about it. It appears that a switching relay for bonding needs to be installed in the breaker panel. The relay would need to close, thus engaging a contactor in the breaker panel, thus enabling bonding when under battery power but be open on generator or utility feed.
Great video for first time Growatt users (like myself :) ) - thanks a stack @Zwy. I taught myself how to do the settings but left #24 on default(disabled) - I've contacted the installers to pose the question ...........
 
What is "u/s"?
Unserviceable.
So, if our inverter breaks or misbehaves we can changeover manually to an alternative power input (grid or genset for us) whilst the inverter is repaired.
This way the freezers stay frozen.... :)......we hope
 
Why do you have a manual inverter bypass?
The growatt has an automatic transfer switch and from reading above it can switch the neutral ground bond appropriately.
If you read the manual for setting #24, it clearly states, from my understanding, it does not bond neutral and ground when in battery mode.
1615850445219.png
 
That would be backwards to what I would expect.
I know, I sent a message to Watts247 and will see what they say. I would think the inverter would be bonded to neutral under battery power such as you think, but I'm not certain it is after reading the manual and the video. Considering they refer to a contactor in the breaker box being activated for bonding neutral and ground, then it must not be.
 
I know, I sent a message to Watts247 and will see what they say. I would think the inverter would be bonded to neutral under battery power such as you think, but I'm not certain it is after reading the manual and the video. Considering they refer to a contactor in the breaker box being activated for bonding neutral and ground, then it must not be.
I watched the video but I would like to see a demonstration of all 8 permutations.
24_enabled,utility_power,NC
24_enabled,utility_power,NO
24_enabled,battery,NC
24_enabled,battery,NO
24_disabled,utility_power,NC
24_disabled,utility_power,NO
24_disabled,battery,NC
24_disabled,battery,NO
 
I watched the video but I would like to see a demonstration of all 8 permutations.
24_enabled,utility_power,NC
24_enabled,utility_power,NO
24_enabled,battery,NC
24_enabled,battery,NO
24_disabled,utility_power,NC
24_disabled,utility_power,NO
24_disabled,battery,NC
24_disabled,battery,NO
I'm currently discussing this with Ian. His response was bonding should only occur in one place, which is true. However, in the video he discusses under genny power and how bonding occurs at the generator. This is the same as my utility feed on my property, all bonding is done at the service entrance. A separate ground is ran to each building and tied into the service panel with bond occurring there. This is in accordance with current NEC.

So, if instead of showing a generator, we show a utility line coming to the Growatt. No difference as far as bonding, the bonding is at the utility line entrance. If the transfer switch disconnects under battery power, there isn't any bond if bond passes thru the transfer switch. This actually would be correct as under battery power could backfeed thru the utility line on ground or neutral.

I'm intending to look inside my panel later today and see how the transfer switch is wired for neutral and ground and if this is just a pass thru. Under battery power, ground should be at the source and not earth.
 
I'm currently discussing this with Ian. His response was bonding should only occur in one place, which is true. However, in the video he discusses under genny power and how bonding occurs at the generator. This is the same as my utility feed on my property, all bonding is done at the service entrance. A separate ground is ran to each building and tied into the service panel with bond occurring there. This is in accordance with current NEC.

So, if instead of showing a generator, we show a utility line coming to the Growatt. No difference as far as bonding, the bonding is at the utility line entrance. If the transfer switch disconnects under battery power, there isn't any bond if bond passes thru the transfer switch. This actually would be correct as under battery power could backfeed thru the utility line on ground or neutral.

I'm intending to look inside my panel later today and see how the transfer switch is wired for neutral and ground and if this is just a pass thru. Under battery power, ground should be at the source and not earth.
I'm thinking of an interlock setup just to validate the concept.
The interlock switch makes the the mains breaker and the generator breaker mutually exclusive.
The generator should be floating neutral and the generator breaker will only interrupt the hots and not the neutral.
I'm not sure and can't check right now but I think the master breaker only interrupts the hots and not the neutral and quite sure it doesn't interrupt the ground.
Does that sound right?
If think it is and that it is NEC compliant.
 
I'm thinking of an interlock setup just to validate the concept.
The interlock switch makes the the mains breaker and the generator breaker mutually exclusive.
The generator should be floating neutral and the generator breaker will only interrupt the hots and not the neutral.
I'm not sure and can't check right now but I think the master breaker only interrupts the hots and not the neutral and quite sure it doesn't interrupt the ground.
Does that sound right?
If think it is and that it is NEC compliant.
I ran across this by @FilterGuy https://diysolarforum.com/attachments/1614907573554-png.39595/
 
I was notified of this thread when @Zwy mentioned me so I have only read a few posts back.... Forgive me if I am repeating what others have said.

As mentioned by others, the key is to only have one local N-G Bond. In a standard US grid-tie house, this is done at the main breaker panel and is called the Main Bonding Jumper. The primary (only?) purpose of this bond is to provide a very low impedance path for a ground fault (short to the Equipment grounding system) to pop the breaker and clear a fault.

For the vast majority of grid tie solar installations without storage, the inverter does *not* insert another N-G bond. Therefor the 'one-local-bond' rule is maintained.

With inverters that insert an N-G bond, things can get complicated. If they are feeding a Main breaker at the house, the Main Bonding Jumper should be removed in order to maintain the one-local-bond' rule. However, if the inverter dynamically inserts and removes the N-G bond that would leave the system with no N-G bond at times and that is even more dangerous (and against NEC code). When used in a home situation with a Main Bonding Jumper in place, the LV2424 needs to have the dynamic N-G disabled and never provide a bond. (I do not know if the LV2424 has a way of disabling it).

If the LV2424 is used off grid with a generator, then the Generator should have an N-G bond and there should not be a Main Bonding Jumper at the main breaker panel. That way, when on battery the LV2424 is supplying the N-G bond and when on Generator, the Generator is providing the N-G bond.

Note: The dynamic N-G bond is really intended for a mobile system with shore power. When on battery, the inverter supplys the N-G bond and when on shore power, the shore power supplies the N-G bond.
 
I was notified of this thread when @Zwy mentioned me so I have only read a few posts back.... Forgive me if I am repeating what others have said.

As mentioned by others, the key is to only have one local N-G Bond. In a standard US grid-tie house, this is done at the main breaker panel and is called the Main Bonding Jumper. The primary (only?) purpose of this bond is to provide a very low impedance path for a ground fault (short to the Equipment grounding system) to pop the breaker and clear a fault.

For the vast majority of grid tie solar installations without storage, the inverter does *not* insert another N-G bond. Therefor the 'one-local-bond' rule is maintained.

With inverters that insert an N-G bond, things can get complicated. If they are feeding a Main breaker at the house, the Main Bonding Jumper should be removed in order to maintain the one-local-bond' rule. However, if the inverter dynamically inserts and removes the N-G bond that would leave the system with no N-G bond at times and that is even more dangerous (and against NEC code). When used in a home situation with a Main Bonding Jumper in place, the LV2424 needs to have the dynamic N-G disabled and never provide a bond. (I do not know if the LV2424 has a way of disabling it).

If the LV2424 is used off grid with a generator, then the Generator should have an N-G bond and there should not be a Main Bonding Jumper at the main breaker panel. That way, when on battery the LV2424 is supplying the N-G bond and when on Generator, the Generator is providing the N-G bond.

Note: The dynamic N-G bond is really intended for a mobile system with shore power. When on battery, the inverter supplys the N-G bond and when on shore power, the shore power supplies the N-G bond.
I believe this is why Growatt has the relay control concerning bonding built in and it is disabled/enabled under setting #24. Under battery power, the N-G is enabled automatically by the Growatt at the inverter and disabled when under utility power thru the transfer switch which is tied to the main service panel where bonding occurs.

In the case of using a generator where the N-G is bonded, then following the one place N-G bonding rule, we have 2 choices. Either remove the bonding of the generator and ground it separately to earth or have a contactor in service panel to open the N-G bond. This contactor can be controlled by the relay control on the Growatt. However, this also means the Growatt will not have grounding pass thru when under utility power unless one determines a wiring solution to enable pass thru. The solution appears as bond at the service panel transfer switch and disable bond at the generator for permanent non mobile situations. Power should be wired from the generator to service panel transfer switch and not generator to Growatt.

In the case of mobile situations, then bonding should be enabled at the generator and power fed directly to the Growatt. As long as the transfer switch internal to the Growatt is grounding pass thru and power not run thru the inverter, then there would only be one N-G bond. However, if the power is run thru the inverter in a case where generator and battery power are used at the same time, there would be 2 N-G bonds. I'm not certain how that should be handled. I believe the setting under #24 plays into it. The best policy in such a situation (although rare) might be bonding at a service panel with a contactor that disables/enables bonding.
 
Back
Top