Safe Grid Use of the 5000ES and transformer

[Cheap 4-life]

As shown, if a fault develops in a GroWatt (or AC In panel, or AC Out panel, or Load panel),
and "L1" shorts to chassis, chassis of all those devices goes to 120VAC.
The only path back to utility meter is through the earth. Your added ground rod is at 120VAC, ground rod by utility meter is zero volts. Current flows, but not enough to trip a breaker feeding GroWatt (Faults downstream of GroWatt might trip a breaker or shut off GroWatt, but faults through AC Input of GroWatt will not.

The earthworms will frantically wriggle to the surface.

I believe this is wrong. A ground wire from utility meter providing continuity to case of GroWatt case is required, so fault causes breaker to clear.

I agree, but if a ground from the utility meter goes to the main panel and to the growatts then we have the same issue that Ian’s video showed with the neutral circling amps due to varying 120v loads.
what is suggested to do this right. Was it an isolation transformer?

 

[Hedges]

Isolation transformer before or after GroWatt should work.

What is wrong with running ground and neutral all the way from utility entrance, to GroWatt, to loads?

I understand balancing transformer can't perfectly balance a single-phase load onto two phases (driven by grid) due to it's winding resistance. But breakers protect against that. You just have to make sure transformer breaker (25A in the drawing?) can't shut off without cutting power to the panel. Not handled as shown in schematic, but we could fix that.

 

[Cheap 4-life]

What is wrong with running ground and neutral all the way from utility entrance, to GroWatt, to loads?

Wouldn’t the extra ground rod in sigs diagram above have to be removed? and the ground/neutral bond in the load panel from the diagram would have to be removed? And then aren’t we right back to what Ian’s videos were describing about the unbalanced 120v loads current looping thru the growatts? Or did I miss something

 

[Solar Guy Richard]

Why does it have AC input connectors and show a possible grid connection in the manual if it is an off grid unit ???

Off grid means you are not selling to the grid, not that you can’t use utility power as a source. The reason this comment is relevant is the underlying issue of having a ground from a utility source as opposed to isolating the ground on your system.

 

[Hedges]

Wouldn’t the extra ground rod in sigs diagram above have to be removed? and the ground/neutral bond in the load panel from the diagram would have to be removed? And then aren’t we right back to what Ian’s videos were describing about the unbalanced 120v loads current looping thru the growatts? Or did I miss something

I'm not sure that multiple ground rods is a problem. Some scheme of multiple rods, plates, other things in the dirt is sometimes recommended. Mike Holt doesn't like it, at least if lighting applies a potential between them. Probably OK if connected in a star, not carrying through ground wires of circuits.

Now that I've watched Ian's video, I see he says imbalanced voltage on the pole (could be neighbor's heavy 120V load) makes your auto-transformer try to balance it. OK, that's a potential problem if we pull neutral through (or just ground, and tie ground to neutral thereby making ground a current-carrying conductor.)

Damned if you do, Damned if you don't.

Ian did mention a neutral bonding relay. When operating off-grid, neutral bonds to ground locally. When operating on-grid, local bond is opened because grid provides neutral-ground bond. In the US, that is UL-xxxx for mobile units. I think GroWatt is fixed and hard-wired, not mobile. But might get fed by a generator. Does GroWatt have a relay that bonds neutral to ground whenever it disconnects from grid? And it has a 2-pole relay to isolate "Line" and "Neutral" (which we're going to hook to 120/240V split phase "Line1" and "Line2")?

In that case, what could work if you could access that neutral bonding relay (having removed a screw so GroWatt "Neutral", US DIY "Line2" isn't grounded), is use that relay to bond center-tap of auto-transformer to ground. When on-grid, only L1 and L2 of autotransformer are connected. It is an inductor, nothing more. Any 120V loads you have are fed by Line1 and Neutral of grid, or Line2 and Neutral of grid, all connected through what might be a 3PDT relay to the grid. When off-grid, relay switches and L1, L2, N are open. Neutral goes from Neutral of grid to Ground inside GroWatt. Assuming these nets are suitably accessible. But I think the GroWatt PCB only knows Line, "Neutral", and ground, so terminals of relay probably wired to something other than what I want. I could probably tap off that neutral-bonding circuit to add a separate bonding relay.

So Surgery is the more elegant and more difficult approach. Isolation transformer (either input or output) is KISS.

Good I finally watched Ian's video. I've contemplated an autotransformer, thinking it would balance current through two legs. Apparently only in some situations; others it tries to balance the sins of the utility grid.

 

[Solar Guy Richard]

I'm not sure that multiple ground rods is a problem. Some scheme of multiple rods, plates, other things in the dirt is sometimes recommended. Mike Holt doesn't like it, at least if lighting applies a potential between them. Probably OK if connected in a star, not carrying through ground wires of circuits.

Now that I've watched Ian's video, I see he says imbalanced voltage on the pole (could be neighbor's heavy 120V load) makes your auto-transformer try to balance it. OK, that's a potential problem if we pull neutral through (or just ground, and tie ground to neutral thereby making ground a current-carrying conductor.)

Damned if you do, Damned if you don't.

Ian did mention a neutral bonding relay. When operating off-grid, neutral bonds to ground locally. When operating on-grid, local bond is opened because grid provides neutral-ground bond. In the US, that is UL-xxxx for mobile units. I think GroWatt is fixed and hard-wired, not mobile. But might get fed by a generator. Does GroWatt have a relay that bonds neutral to ground whenever it disconnects from grid? And it has a 2-pole relay to isolate "Line" and "Neutral" (which we're going to hook to 120/240V split phase "Line1" and "Line2")?

In that case, what could work if you could access that neutral bonding relay (having removed a screw so GroWatt "Neutral", US DIY "Line2" isn't grounded), is use that relay to bond center-tap of auto-transformer to ground. When on-grid, only L1 and L2 of autotransformer are connected. It is an inductor, nothing more. Any 120V loads you have are fed by Line1 and Neutral of grid, or Line2 and Neutral of grid, all connected through what might be a 3PDT relay to the grid. When off-grid, relay switches and L1, L2, N are open. Neutral goes from Neutral of grid to Ground inside GroWatt. Assuming these nets are suitably accessible. But I think the GroWatt PCB only knows Line, "Neutral", and ground, so terminals of relay probably wired to something other than what I want. I could probably tap off that neutral-bonding circuit to add a separate bonding relay.

So Surgery is the more elegant and more difficult approach. Isolation transformer (either input or output) is KISS.

Good I finally watched Ian's video. I've contemplated an autotransformer, thinking it would balance current through two legs. Apparently only in some situations; others it tries to balance the sins of the utility grid.

Exactly. It’s a mixed bag of what problems you want to deal with for whatever results you want. We just want to dispel any fears from a safety standpoint, but being on a DIY forum typically means each person has to come up with interesting solutions as long as they stay safe. As someone who puts together and takes apart these systems daily for a multitude of reasons, I’m not too interested in being hurt by them anymore than anyone else so we do a ton of testing to ensure the safety and integrity of our products if used correctly.

 

[Solar Guy Richard]

Just a quick update, we are still generating additional schematics for different set ups (such as without the transfer switch) for everyone to utilize. As always, I welcome any feedback (publicly or through PM) on ways we can improve any aspect of your interaction with Signature Solar.

 

[Jon Hamilton]

We drive ground rods because we are electrical engineers who give a crap, Joe blow turning all green to save the earth on the cheap ain't going to. Nor the guy living on rock ledge lol.
Single phase loads will not get power and be subject to possible loss of neutral and hence unbalanced 120v legs that will blow them up if the circuit breaker kills the voltage to the primary of the transformer.

I am no electrical engineer, but I drove one.

 

[Hedges]

Exactly. It’s a mixed bag of what problems you want to deal with for whatever results you want. We just want to dispel any fears from a safety standpoint, but being on a DIY forum typically means each person has to come up with interesting solutions as long as they stay safe. As someone who puts together and takes apart these systems daily for a multitude of reasons, I’m not too interested in being hurt by them anymore than anyone else so we do a ton of testing to ensure the safety and integrity of our products if used correctly.

Your schematic posted has a problem I described earlier, which needs to be fixed.

If a fault develops shorting L1 or L2 from grid to chassis of GroWatt or to the electrical box before it, entire "ground" network of your system goes to 120V.
There is no ground wire back from GroWatt to utility connection. The equipment will sit at 120V relative to (some parts) of earth. Current will flow through the dirt, because one ground rod is at zero volts and one is at 120V. But current won't be high enough to trip a circuit breaker.
That is a shock hazard. Someone touching a grounded appliance like washing machine and also touching a water pipe will get shocked. Also shocked if touching case of GroWatt and standing on concrete garage floor.
It also causes "step potential", someone walking across the dirt will have a voltage difference between their feet.

Please fix it.

 

[LeRoyK]

Your schematic posted has a problem I described earlier, which needs to be fixed.

If a fault develops shorting L1 or L2 from grid to chassis of GroWatt or to the electrical box before it, entire "ground" network of your system goes to 120V.
There is no ground wire back from GroWatt to utility connection. The equipment will sit at 120V relative to (some parts) of earth. Current will flow through the dirt, because one ground rod is at zero volts and one is at 120V. But current won't be high enough to trip a circuit breaker.
That is a shock hazard. Someone touching a grounded appliance like washing machine and also touching a water pipe will get shocked. Also shocked if touching case of GroWatt and standing on concrete garage floor.
It also causes "step potential", someone walking across the dirt will have a voltage difference between their feet.

Please fix it.

I believe if the fault to chassis or to the electrical box happens after the breakers in "AC in panel" then these breakers will protect it.
But if the fault happens on the wires leading from meter to AC in panel then there is no ground fault protection.

I was wrong, see @Hedges post#52 below

 

[Solar Guy Richard]

Your schematic posted has a problem I described earlier, which needs to be fixed.

If a fault develops shorting L1 or L2 from grid to chassis of GroWatt or to the electrical box before it, entire "ground" network of your system goes to 120V.
There is no ground wire back from GroWatt to utility connection. The equipment will sit at 120V relative to (some parts) of earth. Current will flow through the dirt, because one ground rod is at zero volts and one is at 120V. But current won't be high enough to trip a circuit breaker.
That is a shock hazard. Someone touching a grounded appliance like washing machine and also touching a water pipe will get shocked. Also shocked if touching case of GroWatt and standing on concrete garage floor.
It also causes "step potential", someone walking across the dirt will have a voltage difference between their feet.

Please fix it.

Researching. Thanks!

 

[Hedges]

I believe if the fault to chassis or to the electrical box happens after the breakers in "AC in panel" then these breakers will protect it.
But if the fault happens on the wires leading from meter to AC in panel then there is no ground fault protection.

The devil is in the details.
"AC in panel" is fed with 100A breaker from utility meter.
Panel feeds GroWatts with 30A breakers.

If a fault occurs before 100A breaker (wire leading into box shorts), before 30A breakers (contact between busbar and box), or after 30A breaker (including to chassis of GroWatt), ground net is drive to 120V, or at least tries to. This causes a short only to center-tap of autotransformer, which is protected by 25A breaker.

I think that short only occurs by the magic of the transformer. A transformer's ability to drive currents and voltage is limited by two things. Winding resistance causes a voltage sag, usually 5% or 10% of spec at full current. Current coupling is limited by magnetic saturation of core. This transformer is meant to transfer maximum 2500W between phases (20A at 120V). It likely can't reach 40A even if one phase is short circuited. Likely something between 25A and 40A. Even if greater than 25A, breaker won't trip immediately. At 40A, a 25A thermal/magnetic breaker takes about 10 minutes.

The 100A breaker won't trip. 30A breaker might trip (which clears the fault.) If 30A breaker doesn't trip, voltage on ground is likely reduced below 120V by some amount until 25A breaker trips. Once 25A breaker trips, "ground" of system goes to full 120V from utility feed. (Also, 120V devices plugged in are unhappy.)

 

[Hedges]

Another issue, important but not as urgent because not human safety related.
When you introduced this transformer, you noted,

"Unlike other transformers on the market, this unit makes it virtually impossible to allow an open ground within your load center. By providing a single phase inverter the autotransformer as a ‘pass through’ we are completely removing any chance of the transformer tripping prior to an inverter in any non-standard event. By separating the incoming inverter power from the outgoing split phase power from the autotransformer, we have eliminated the possibility of independent transformer trips due to overload. This also allows for the addition of a breaker between the inverter and transformer that will completely disable any output in the event of any non-standard occurrence."

However, you have left out this protection (meant to keep 120V loads from experiencing surge or brownout if breaker disconnects autotransformer.) By branching off to the autotransformer from a breaker instead of feeding through it, your circuit goes back to how "other transformers on the market" do it.

New Split Phase Transformer from Signature Solar

Growatt Split Phase Transformer Signature Solar now offers the latest US Split Phase Auto Transformer by Growatt. This autotransformer takes Single Phase 240Vac (L/N) electricity output and creates true Split Phase 120Vac/120Vac (L/L/N). This addition to our linesheet increases safety and ease...

diysolarforum.com

 

[Solar Guy Richard]

Another issue, important but not as urgent because not human safety related.
When you introduced this transformer, you noted,

"Unlike other transformers on the market, this unit makes it virtually impossible to allow an open ground within your load center. By providing a single phase inverter the autotransformer as a ‘pass through’ we are completely removing any chance of the transformer tripping prior to an inverter in any non-standard event. By separating the incoming inverter power from the outgoing split phase power from the autotransformer, we have eliminated the possibility of independent transformer trips due to overload. This also allows for the addition of a breaker between the inverter and transformer that will completely disable any output in the event of any non-standard occurrence."

However, you have left out this protection (meant to keep 120V loads from experiencing surge or brownout if breaker disconnects autotransformer.) By branching off to the autotransformer from a breaker instead of feeding through it, your circuit goes back to how "other transformers on the market" do it.

New Split Phase Transformer from Signature Solar

Growatt Split Phase Transformer Signature Solar now offers the latest US Split Phase Auto Transformer by Growatt. This autotransformer takes Single Phase 240Vac (L/N) electricity output and creates true Split Phase 120Vac/120Vac (L/L/N). This addition to our linesheet increases safety and ease...

diysolarforum.com

This schematic is specifically for the SolarEdge transformer. We will likely have a different set of schematics for this new product.

 

[Hedges]

This schematic is specifically for the SolarEdge transformer. We will likely have a different set of schematics for this new product.

Your "5000 ES split phase connection" schematic:

Safe Grid Use of the 5000ES and transformer

diysolarforum.com

Has two GroWatt each feeding 20A breaker of AC Out panel. That panel could drive 240V loads, with no 120V loads connected, if desired. Capable of 40A, or whatever inverters will actually drive.
Then that panel drives Load Panel. Load panel has 25A breaker to "Split Phase Transformer" (apparently auto-transformer, considering just L1, L2, and N connections.) (Suggest you add an explicit ground connection to transformer.)

Problem here is you can trip the 25A breaker, causing 120V loads on that panel to receive incorrect voltage, damaging them.
You could modify the design so if transformer gets disconnected by breaker, all power to the panel gets disconnected too.
If there is a spec for maximum imbalance (2500W you quoted for one transformer, or 1500W I saw in one datasheet), may be best that breakers protect to that level - it further limits what can be powered even if balanced, but any schematic you publish should have protection so it can't get overloaded by anything connected downstream.

 

[LeRoyK]

The devil is in the details.
"AC in panel" is fed with 100A breaker from utility meter.
Panel feeds GroWatts with 30A breakers.

If a fault occurs before 100A breaker (wire leading into box shorts), before 30A breakers (contact between busbar and box), or after 30A breaker (including to chassis of GroWatt), ground net is drive to 120V, or at least tries to. This causes a short only to center-tap of autotransformer, which is protected by 25A breaker.

I think that short only occurs by the magic of the transformer. A transformer's ability to drive currents and voltage is limited by two things. Winding resistance causes a voltage sag, usually 5% of 10% of spec at full current. Current coupling is limited by magnetic saturation of core. This transformer is meant to transfer maximum 2500W between phases (20A at 120V). It likely can't reach 40A even if one phase is short circuited. Likely something between 25A and 40A. Even if greater than 25A, breaker won't trip immediately. At 40A, a 25A thermal/magnetic breaker takes about 10 minutes.

The 100A breaker won't trip. 30A breaker might trip (which clears the fault.) If 30A breaker doesn't trip, voltage on ground is likely reduced below 120V by some amount until 25A breaker trips. Once 25A breaker trips, "ground" of system goes to full 120V from utility feed. (Also, 120V devices plugged in are unhappy.)

You are right, I did not consider that the current needs to flow through the transformer to trip the breaker.
I edited my post above.

 

[Quattrohead]

If that 25A breaker trips you loose your neutral and can fry all the 120v equipment.
And it will try to use the ground as the neutral, you NEVER want current on the ground wire, that is not what it is for. Newer 240v appliances specifically separated the neutral and ground wire to remove this safety issue.
Still a cluster and you won't listen to us.

 

[Solar Guy Richard]

Your "5000 ES split phase connection" schematic:

Safe Grid Use of the 5000ES and transformer

diysolarforum.com

Has two GroWatt each feeding 20A breaker of AC Out panel. That panel could drive 240V loads, with no 120V loads connected, if desired. Capable of 40A, or whatever inverters will actually drive.
Then that panel drives Load Panel. Load panel has 25A breaker to "Split Phase Transformer" (apparently auto-transformer, considering just L1, L2, and N connections.) (Suggest you add an explicit ground connection to transformer.)

Problem here is you can trip the 25A breaker, causing 120V loads on that panel to receive incorrect voltage, damaging them.
You could modify the design so if transformer gets disconnected by breaker, all power to the panel gets disconnected too.
If there is a spec for maximum imbalance (2500W you quoted for one transformer, or 1500W I saw in one datasheet), may be best that breakers protect to that level - it further limits what can be powered even if balanced, but any schematic you publish should have protection so it can't get overloaded by anything connected downstream.

Agreed on the ground wire to the transformer.

 

[Bud Martin]

So is the wiring diagram the finale version as shown on this thread and in the thread below?

Growatt SPF 5000 ES Wiring for 0 Export / Need help

I have my growatt 5000es setup like you describe. Check out my diagram I posted. I am trying to get it verified, so use at your own risk. Any questions on inverter settings, let me know. I can send you what mine is set to, and what settings do. This is how these units are to be set up, you can...

diysolarforum.com

 

[Bud Martin]

Your schematic posted has a problem I described earlier, which needs to be fixed.

If a fault develops shorting L1 or L2 from grid to chassis of GroWatt or to the electrical box before it, entire "ground" network of your system goes to 120V.
There is no ground wire back from GroWatt to utility connection. The equipment will sit at 120V relative to (some parts) of earth. Current will flow through the dirt, because one ground rod is at zero volts and one is at 120V. But current won't be high enough to trip a circuit breaker.
That is a shock hazard. Someone touching a grounded appliance like washing machine and also touching a water pipe will get shocked. Also shocked if touching case of GroWatt and standing on concrete garage floor.
It also causes "step potential", someone walking across the dirt will have a voltage difference between their feet.

Please fix it.

Yep! That diagram needs to be fixed, it is being used as ref in the other threads now too. Some one may get hurt and it also has SS notes on it too.
For example:

Growatt SPF 5000 ES Wiring for 0 Export / Need help

I have my growatt 5000es setup like you describe. Check out my diagram I posted. I am trying to get it verified, so use at your own risk. Any questions on inverter settings, let me know. I can send you what mine is set to, and what settings do. This is how these units are to be set up, you can...

diysolarforum.com