Isolation Transformer Neutral Ground Connections

[SeaBoy]

Hello TF,

I’m in the works of rewiring my boat and installing a new panel I’m making to fit my needs, I have 2 30amp shore power lines, an 8.5kW generator, and a 2500w inverter. I’m ditching the “Ship-Shore” transfer switches and replacing them with double pole breakers with “lock out slides” to allow only 1 AC source to supply the loads. I will be using an inverter sub panel to prevent back-feeding the inverter as well

I’ve done an incredible amount of research online and in forums, but what I’m having issues with is my AC neutral to ground connections and there seems to be no clear answers online.

I understand the neutral and ground need to be connected at the source (gen and inverter do this automatically). But for my isolation transformer, there is no ground connection available on the secondary winding side. So how/where do I connect my neutral and ground?

See below images

 

[Iggy]

You don't connect them. Thats the point of isolation.

Keep in mind, that a transformer does not have a physical connection to the hot OR neutral on the power station side. As your pic shows. The power is transferred though magnetism to keep it simple.

With that in mind, and general speaking. Most transformers with metal cases are grounded from the power source/input side. On the output side, a transformer from a Victron is marked Neutral. and Hot terminals. So your ground would now come off the Neutral terminal on the output side or a terminal marked ground. But the input and output grounds are never connected, just to make that clear.

 

[SeaBoy]

Iggy- understood on how isolation transformers work. I installed it because I was tripping a 30 mA shore power breaker that I later found was my water heater. My isolation transformer does not have a metal casing. It’s a plastic case with the 1 to 1 windings encased in resin.

So if neutral and ground never get connected, why is it the Gen and Inverter connect them when running? What’s the difference? All 3 are AC power sources

 

[Bmarler]

Your system is very similar to mine. Almost identical. I used Victron transformers and they do in fact have designated terminals for incoming ground and isolated output ground. There’s a removable jumper to isolate the ground/neutral connection if needed.
I suspect the incoming ground wire terminal may be connected to the transformer shield, but I’m not sure.
If you’re using some other brand of transformer, if there’s a shield accessible, I’d hook the incoming ground there. If no shield connection is available, drop the ground at the shore power inlet.
I wrangled ground connections just like you. I have 2 shore inlets, 2 transformers, genset, inverter. Shore 1 goes to non-inverter loads, shore 2 to inverter loads. Genset connects like yours, blue seas lockout selectors. It’s been flawless since we did the re wire.

 

[twistedtree]

For a project of this scope, I think it's essential to become an ABYC member and get a copy of E-11 which is the AC and DC wiring standard. It spells out how to bring shore power onto the boat, where to ground the shore side, where neutral & ground are bonded on the boat, where you need over current protections (OCP), etc. There is quite a bit in your diagram that isn't correct, and would get flagged by a thorough surveyor. Here are a few things, in no particular order:


- You need a breaker between the shore inlet. You are using one of your interlocked breakers, but need a separate one for the input. The transformer output should go to one of the interlocked breaker's input.


- Depending on the cable length between shore inlet, isolation transformer, and interlocked main breaker, you may need other breakers.


- The outputs of the interlocked breakers should be wired in parallel, and provide power to the loads. Because the breakers are interlocked, only one input source will power the outputs.


- I think your "Inverter Loads" drawing is intended to show a blend of the inverter and some sort of load panel? I'm not quite following it.


With isolation transformers, you have two options for bonding neutral and ground.


1) You can bond the isolation transformers, generator, and inverter output, and have switching that ensures that only on bonding point is active at any given time. You will need to check carefully in this system to be be sure the two isolation transformers can't both be bonding at the same time, and also that the inverter bonds or doesn't bond as needed.


2) Because you are using isolation transformers, you can simply tie all the neutrals together on the boat, and bond to ground at a single central point, typically at the load panel. This can often be a lot easier than trying to ensure only one bonding point is active at any given time.


All this will just give you a flavor of what's needed, but you really need to consult E-11 for complete guidance.

 

[Bmarler]

Iggy- understood on how isolation transformers work. I installed it because I was tripping a 30 mA shore power breaker that I later found was my water heater. My isolation transformer does not have a metal casing. It’s a plastic case with the 1 to 1 windings encased in resin.

So if neutral and ground never get connected, why is it the Gen and Inverter connect them when running? What’s the difference? All 3 are AC power sources

With the non metallic case, ( I don’t think abyc approved, but is euro spec)
You create the neutral to ground at the transformer.
I did a drawing of my system, but I’m away from the boat at the moment.

 

[SeaBoy]

Bmarler - do you have ELCI main breakers or standard double pole breakers?

I do not have the ground shielding available. My shore inlet ground is cut and capped. See below iso transformer wiring schematic

 

[twistedtree]

Be cautious with the Victron Isolation Transformer. They instruct you to install it incorrectly compared to ABYC. I'm not sure what ISO says, but it's more likely to match ABYC.


The shore side ground needs to connect to the transformer's internal ground shield. The boat-side ground connects to any metal chassis.


If your transformer doesn't include an internal ground shield, then it's not currently suitable for use as a isolation transformer. That said, the most recent draft of E-11 that was open to public review this past fall adds language for toroidal transformers that inherently do not include an internal ground shield. In that case you must use a RCD breaker between the shore power inlet and the transformer. The shore ground is not brought past the shore inlet. But none of this has been approved, and there were quite a few comments on it that are not yet resolved. So proceed at your own risk.

 

[SeaBoy]

Twisted Tree. Thank you for you replay. Respectfully. I don’t understand some of your comments:

1. Why would I need another breaker? If you look at my schematic, I have the transformer output to the line side of the interlocked breaker. Are you saying there needs to be a breaker on the primary side of the isolation transformer to protect the 2 feet of wiring to the primary side?

“ you need a breaker between the shore inlet. You are using one of your interlocked breakers, but need a separate one for the input. The transformer output should go to one of the interlocked breaker's input.”

2. My schematic allows shore 1 loads to be separate from shore 2 loads, but when I want to use my Gen, the transfer breaker will connect both shore 1 and shore 2 loads

 

[twistedtree]

Bmarler - do you have ELCI main breakers or standard double pole breakers?

I do not have the ground shielding available. My shore inlet ground is cut and capped. See below iso transformer wiring schematic

If you used an RCD breaker at the xfmr input, and bonded the xfmr output neutral with the boat's ground, then I think it would agree with the E-11 draft. But again, all this language in E-11 has yet to be approved, and has a lot of outstanding and unresolved public comments.

 

[Bmarler]

Bmarler - do you have ELCI main breakers or standard double pole breakers?

I do not have the ground shielding available. My shore inlet ground is cut and capped. See below iso transformer wiring schematic

I do not have elci inlet breakers as I’m within the required length of wire to the transformer.
I had the same toroidal transformers you have on my last boat, and did use the elci, only as there was no metal case or shield.
That company claims superior double isolation, and is euro approved, but not abyc. (Yet)

 

[twistedtree]

Twisted Tree. Thank you for you replay. Respectfully. I don’t understand some of your comments:

1. Why would I need another breaker? If you look at my schematic, I have the transformer output to the line side of the interlocked breaker. Are you saying there needs to be a breaker on the primary side of the isolation transformer to protect the 2 feet of wiring to the primary side?

“ you need a breaker between the shore inlet. You are using one of your interlocked breakers, but need a separate one for the input. The transformer output should go to one of the interlocked breaker's input.”

Yes, that's what's called for. And in this case it needs to be an RCD breaker.

2. My schematic allows shore 1 loads to be separate from shore 2 loads, but when I want to use my Gen, the transfer breaker will connect both shore 1 and shore 2 loads

Yes, and a very reasonable approach. I just think the power sources should each go to a dedicated breaker input, with the outputs tied together and serving the loads.


One of the issues as drawn is that when you are using your gen you are also back-powering one or both of the transformers, and their N-G bond will be in circuit rather than out of circuit. This assumes you bond at each power source.

 

[Bmarler]

Twisted Tree. Thank you for you replay. Respectfully. I don’t understand some of your comments:

1. Why would I need another breaker? If you look at my schematic, I have the transformer output to the line side of the interlocked breaker. Are you saying there needs to be a breaker on the primary side of the isolation transformer to protect the 2 feet of wiring to the primary side?

“ you need a breaker between the shore inlet. You are using one of your interlocked breakers, but need a separate one for the input. The transformer output should go to one of the interlocked breaker's input.”

2. My schematic allows shore 1 loads to be separate from shore 2 loads, but when I want to use my Gen, the transfer breaker will connect both shore 1 and shore 2 loads

Tomorrow, when I’m at the boat, I’ll try to find my schematic. It’s similar to your approach, but has some differences. I worked through it with help from an abyc electrician so I could get it right according to them.
I’m certainly capable of putting together a safe system, but if abyc compliance is desired, you need the information because there are some nuances that will be missed. Just like nec for land based work.

 

[SeaBoy]

Twisted Tree - Okay okay I gotcha. Makes sense

Bmarler - can you send me your schematic when you get back to the boat?

 

[Bmarler]

Twisted Tree - Okay okay I gotcha. Makes sense

Bmarler - can you send me your schematic when you get back to the boat?

Will do. You can use any or none of it as you see fit.

 

[SeaBoy]

Thank you sir!!

 

[Bud]

Hello TF,

I’m in the works of rewiring my boat and installing a new panel I’m making to fit my needs, I have 2 30amp shore power lines, an 8.5kW generator, and a 2500w inverter. I’m ditching the “Ship-Shore” transfer switches and replacing them with double pole breakers with “lock out slides” to allow only 1 AC source to supply the loads. I will be using an inverter sub panel to prevent back-feeding the inverter as well

I’ve done an incredible amount of research online and in forums, but what I’m having issues with is my AC neutral to ground connections and there seems to be no clear answers online.

I understand the neutral and ground need to be connected at the source (gen and inverter do this automatically). But for my isolation transformer, there is no ground connection available on the secondary winding side. So how/where do I connect my neutral and ground?

See below images

This is an interesting subject that as you point out needs more clarity. I can tell you what I did and why, which might be helpful in understanding the neutral ground issues.
I did not want neutral connected to ground while plugged into shore power because it will trip gfci breakers and will/can draw current through my underwater grounded gear causing electrolosis and if shore power loses or has a high resistant neutral then other vessels on your dock can draw current through the grounded gear(in the water). Gfci breakers compare the current going out and compare it to the current coming back and if it is not within 30ma it will trip. By drawing current through the underwater gear that trips the gfci breaker. My fix was not an isolation transformer ( which is a good fix) it was making sure the two places in my electrical system where the neutral was tied to ground i used an interlocking breaker that also disconnected the neutral.
thanks,
Bud

 

[twistedtree]

This is an interesting subject that as you point out needs more clarity. I can tell you what I did and why, which might be helpful in understanding the neutral ground issues.
I did not want neutral connected to ground while plugged into shore power because it will trip gfci breakers and will/can draw current through my underwater grounded gear causing electrolosis and if shore power loses or has a high resistant neutral then other vessels on your dock can draw current through the grounded gear(in the water). Gfci breakers compare the current going out and compare it to the current coming back and if it is not within 30ma it will trip. By drawing current through the underwater gear that trips the gfci breaker. My fix was not an isolation transformer ( which is a good fix) it was making sure the two places in my electrical system where the neutral was tied to ground i used an interlocking breaker that also disconnected the neutral.
thanks,
Bud

ABYC calls for doing exactly as you did when there is no isolation transformer. In that case the shore power neutral-ground bond is in the shore infrastructure, and must not be replicated on the boat. RCD shore breakers will now alert you if your boat is not wired accordingly.

 

[twistedtree]

For SeaBoy, something for your consideration....


When your shore connections run through isolation transformers, as yours do, ABYC allows for tying all the neutrals together, and bonding them to ground in one central location. I think I mentioned this in an earlier note too. This is my own preferred approach, and I think is much easier to implement, and much less prone to error vs bonding at each power source and ensuring that those bonds are always mutually exclusive regardless of what's powering what.

 

[Bud]

ABYC calls for doing exactly as you did when there is no isolation transformer. In that case the shore power neutral-ground bond is in the shore infrastructure, and must not be replicated on the boat. RCD shore breakers will now alert you if your boat is not wired accordingly.

Roger that.
Bud

 

[marcs]

You have gotten a lot of good advice, but if you haven't, you may want to read through Steve D'Antonio's article on the subject:


https://stevedmarineconsulting.com/wp-content/uploads/2019/12/ShorePowerTransformers181-04.pdf


It includes a few diagrams from E-11, but I agree with twistedtree that reading the whole of E-11 is very valuable. You can find PDFs of (usually) slightly older versions online if you search.

 

[PierreR]

I am reading this with high interest. I have my own issues with shore power but I will start my own thread. Meantime, some of my questions might be answered here.

 

[SeaBoy]

You have gotten a lot of good advice, but if you haven't, you may want to read through Steve D'Antonio's article on the subject:


https://stevedmarineconsulting.com/wp-content/uploads/2019/12/ShorePowerTransformers181-04.pdf


It includes a few diagrams from E-11, but I agree with twistedtree that reading the whole of E-11 is very valuable. You can find PDFs of (usually) slightly older versions online if you search.

In reading this article, because my isolation transformer is installed within 10’ of wire. It looks like I can just use a convention circuit breaker. But I will confirm with ABYC standards.

However, the article does not address if or when I connect my neutral to ground. I peruse the ABUC E-11 and report my findings

 

[twistedtree]

Here’s a summary of the n-g requirements.

Option 1; what I call distributed bonding:

With your power source selector switches all open, the loads should not experience a bonded neutral.

You should bond the neutral at the generator, and at each iso transformer.

Now when you turn on one of those power source breakers, the power source’s bonding is carried through to the loads.

Then there are two other things you will need to do.

When your inverter is on and there is no incoming power source, the output will not be bonded. This is typically solved by an internal bonding relay in the inverter. I think you said your inverter has this. However, it requires that you inverter output neutral bus be separate from the other neutral buses in the boat so the bonding relay only bonds the inverter loads neutral, and not the other loads which might be powered from the gen or isos, and bonded through them

Second, I think you need to similarly separate out the neutrals for the second iso transformer. The switching should be such that if those loads are powered by the gen or the first iso, their respective binding is carried through to the second iso’s loads. And if the second iso is directly powered from shore, its bonding is carried through to its loads

Option 2; what I call central bonding.

Because you have iso transformers, you can take advantage of an exception in E-11 that allows for a single, central bonding point, and a common neutral through the boat. In my opinion this is a huge simplification. In this case you would not bond at the gen, isos, or inverter output, would have a common neutral bus throughout the boat (what you currently show), and you would bond at a single point, typically at the main electric panel. Now all the neutrals are always bonded, and only at one point as required. There is no keeping track of different neutral buses and which loads connect to which, and there are no worries about power source switching of bonding.

One thing you will need to ensure is that it’s possible to disable the bonding relay in your inverter.

 

[SeaBoy]

Here’s a summary of the n-g requirements.

Option 1; what I call distributed bonding:

With your power source selector switches all open, the loads should not experience a bonded neutral.

You should bond the neutral at the generator, and at each iso transformer.

Now when you turn on one of those power source breakers, the power source’s bonding is carried through to the loads.

Then there are two other things you will need to do.

When your inverter is on and there is no incoming power source, the output will not be bonded. This is typically solved by an internal bonding relay in the inverter. I think you said your inverter has this. However, it requires that you inverter output neutral bus be separate from the other neutral buses in the boat so the bonding relay only bonds the inverter loads neutral, and not the other loads which might be powered from the gen or isos, and bonded through them

Second, I think you need to similarly separate out the neutrals for the second iso transformer. The switching should be such that if those loads are powered by the gen or the first iso, their respective binding is carried through to the second iso’s loads. And if the second iso is directly powered from shore, its bonding is carried through to its loads

Option 2; what I call central bonding.

Because you have iso transformers, you can take advantage of an exception in E-11 that allows for a single, central bonding point, and a common neutral through the boat. In my opinion this is a huge simplification. In this case you would not bond at the gen, isos, or inverter output, would have a common neutral bus throughout the boat (what you currently show), and you would bond at a single point, typically at the main electric panel. Now all the neutrals are always bonded, and only at one point as required. There is no keeping track of different neutral buses and which loads connect to which, and there are no worries about power source switching of bonding.

One thing you will need to ensure is that it’s possible to disable the bonding relay in your inverter.

In regards to option 2, I would need a double pole breaker on the inverter sub panel to prevent current feeding back to it if using the other power sources since all neutrals would be tied together, correct?

Side note. In reading E-11, it speaks on grounding on the metal case of the iso transformer. Because mine has a plastic housing, I wouldn’t need to ground there and my bonding would only be a jumper wire from my main ground bus to my main neutral bus?

 

[sunchaser]

For SeaBoy, something for your consideration....


When your shore connections run through isolation transformers, as yours do, ABYC allows for tying all the neutrals together, and bonding them to ground in one central location. I think I mentioned this in an earlier note too. This is my own preferred approach, and I think is much easier to implement, and much less prone to error vs bonding at each power source and ensuring that those bonds are always mutually exclusive regardless of what's powering what.

Peter:
We did this on our vessel even with no isolation transformer. Doing this cured our trips on the newer GFIC shore pedestals.

 

[sunchaser]

Marcs:
Did you consider one 50 amp/240V vs the two 30 amp /110V shore power inlets?

 

[twistedtree]

In regards to option 2, I would need a double pole breaker on the inverter sub panel to prevent current feeding back to it if using the other power sources since all neutrals would be tied together, correct?

I don't think so, but a sketch would be helpful to be sure we are talking about the same thing. And knowing the exact model inverter would be good. I assume it's an inverter/charger, not just an inverter?

Side note. In reading E-11, it speaks on grounding on the metal case of the iso transformer. Because mine has a plastic housing, I wouldn’t need to ground there and my bonding would only be a jumper wire from my main ground bus to my main neutral bus?

That's correct, but note my previous comments about an RCD input breaker, and the limbo status of this type of transformer WRT ABYC, just so you know what risks you are taking. The Bridgeport wiring diagram doesn't show and n-g bonding, and it is required. And the RCD breaker will almost certainly remain required.

 

[SeaBoy]

I’ve got a dinosaur
Xantrex Freedom Marine 25
2500 Watts
130 Amp Charger
Model: 81-2511-12(202)
SN: 281071
Date: 1999

 

[twistedtree]

Peter:
We did this on our vessel even with no isolation transformer. Doing this cured our trips on the newer GFIC shore pedestals.

That's curious. The challenge with all this stuff is that the devil is in the details, and the slightest detail can change everything.


Without an isolation transformer, when you are plugged into shore power, the n-g bond is on-shore in the dock power infrastructure. In this case it's critical that there not be an on-board bonding or some portion of the neutral current will flow through the ground and trip a shore-side RCD.


For something that is so simple in concept, it can be very difficult to implement it correctly since there are so many ways things can be powered between multiple shore cords, generator(s), and inverters.