Heads Up - Victron Isolation Transformer

[Nor'Sea 27]

Shore neutral most certainly does not continue beyond the primary side of the transformer.

I imagine we will bore others nevertheless -

If shore safety green is bonded to the transformer case and boat side safety green is bonded to the transformer case (and that is all anyone says should happen) what "panoply of ground" threats exist? Return paths for faults on the secondary side still return to the transformer; a fault on the primary side (rare but possible) still returns to shore ground.

What threat occurs if safety green is continuous or is severed? (apart from safety green providing a galvanic path to the dock and neighbor boats)

Charles

[goboatingnow]

Originally Posted by Chas Erwin

Shore neutral most certainly does not continue beyond the primary side of the transformer.

I imagine we will bore others nevertheless -

If shore safety green is bonded to the transformer case and boat side safety green is bonded to the transformer case (and that is all anyone says should happen) what "panoply of ground" threats exist? Return paths for faults on the secondary side still return to the transformer; a fault on the primary side (rare but possible) still returns to shore ground.

What threat occurs if safety green is continuous or is severed? (apart from safety green providing a galvanic path to the dock and neighbor boats)

Charles

Dave Rifken argues that the shore AC earth is connected to boat earth in a isolating transformer application. Since for safety the boat earth has to be connected to the neutral of the transformer to allow the fuses to blow in the event of a appliance going hot . See his Diagram 1.

While you could allow the shore AC earth to remain continuous without connecting it to the transformers neutral , it removes a major safety feature , as I said above if a hot to case fault develops , you will not get a shock , like ground referenced AC , BUT you expose yourself to being inserted into the AC circuit if you then happen to touch both be hot appliance and a neutral. Hence most safety code, where a AC protective wire exist, require it to be returned to the neutral of the generator.

So Rifkens argument is that to protect against one , obscure fault , ie secondary breakdown to shield , he compromises the whole safety of isolating transformers by ground referencing their output , albeit using a galvanic isolator.

Yet his obscure fault is easily solved by using a RCD on the shore power inlet , such a shield secondary failure would cause it to trip. All this without compromising the isolation feature

His thinking is wrong , outdated and ignorant of other codes and practices , NEC is not the only standard , nor is ABYC.

Dave

[Nor'Sea 27]

An RCD is very worthwhile - a defacto requirement for new boats. Also, every source requires that green and neutral be bonded at that active source. Moreover - safety green - no matter how it is arranged - offers no protection whatsoever if one touches hot and neutral.

So far I can't think of any safety threat - nor has anybody listed one - that arises because safety green is made continuous.

Charles

[goboatingnow]

An RCD is very worthwhile - a defacto requirement for new boats. Also, every source requires that green and neutral be bonded at that active source. Moreover - safety green - no matter how it is arranged - offers no protection whatsoever if one touches hot and neutral.

So far I can't think of any safety threat - nor has anybody listed one - that arises because safety green is made continuous.

Jeepers , how many times , if you bond the on board green wire to the transformer neutral and onto the shore earth , you effectively destroy the purpose of an isolating transformer. Fault currents can now run from the transformer hot side , thru you to earth and back via the earth connection to the neutral of the transformer.

What that does in ABYC speak is convert a perfectly safe isolating transformer into a less safe polarising one. It exposes onboard recipients to all the fault modes that earth referenced AC systems exhibit and essentially defeats the purpose of isolation.

I've just listed several threats above. The whole point of magnetic isolation is that fault currents can only flow back to the source via specific paths that do not include earth. Hence I can hold the live wire of an isolating output and stand in the sea or a puddle of water and nothing happens.

Dave

[CharlieJ]

A lot of this has been stated in various posts above so I will summarize:

1. The ABYC requires bonding of neutral to safety ground at all onboard derived sources; generators, inverter/chargers, and isolation transformers. This is similar to the NEC requirement for shore based systems.
2. The shore side safety ground is terminated at the electrostatic shield of the isolation transformer. Of course this requirement deals with a conventional core based transformer and not the new switch mode isolation transformers.
3. In a 240VAC/50A split phase application with an isolation transformer, the shore side neutral is not brought aboard. The neutral is derived at the secondary of the isolation transformer.
4. The vessel side safety ground is bonded to neutral and to the case of the isolation transformer.

The results of this are:
1. The vessel is galvanically isolated from the rest of the world.
2. The shore side safety ground provides a redundant low impedance path back to the source (the grid transformer) should a fault occur within the windings of the isolation transformer and to ensure that a fault has a path back to the source so that the protective circuit breaker will trip.
3. The bonding of N>G at the secondary of the isolation transformer provides a redundant low impedance path back to the source to ensure that a fault has a path back to the source (the isolation transformer) so that the protective circuit breaker will trip.

My advice to the OP:
1. Don't remove the IT, correct the installation problem.
2. Determine why your hull potential is in the low end of the protected range.
3. Determine if you have a stray current issue on your vessel that is accelerating anode wastage.

Hope this helps clarify this issue.

[goboatingnow]

Originally Posted by CharlieJ

A lot of this has been stated in various posts above so I will summarize:

1. The ABYC requires bonding of neutral to safety ground at all onboard derived sources; generators, inverter/chargers, and isolation transformers. This is similar to the NEC requirement for shore based systems.
2. The shore side safety ground is terminated at the electrostatic shield of the isolation transformer. Of course this requirement deals with a conventional core based transformer and not the new switch mode isolation transformers.
3. In a 240VAC/50A split phase application with an isolation transformer, the shore side neutral is not brought aboard. The neutral is derived at the secondary of the isolation transformer.
4. The vessel side safety ground is bonded to neutral and to the case of the isolation transformer.

The results of this are:
1. The vessel is galvanically isolated from the rest of the world.
2. The shore side safety ground provides a redundant low impedance path back to the source (the grid transformer) should a fault occur within the windings of the isolation transformer and to ensure that a fault has a path back to the source so that the protective circuit breaker will trip.
3. The bonding of N>G at the secondary of the isolation transformer provides a redundant low impedance path back to the source to ensure that a fault has a path back to the source (the isolation transformer) so that the protective circuit breaker will trip.

My advice to the OP:
1. Don't remove the IT, correct the installation problem.
2. Determine why your hull potential is in the low end of the protected range.
3. Determine if you have a stray current issue on your vessel that is accelerating anode wastage.

Hope this helps clarify this issue.

I would argue that 4. Isn't correct electrically. The case of the isolating transformer should be connected to shore side earth. This is the only device onboard that has such a connection to shore power and hence is technically part of the earth referenced supply

The output is floating , so no advantage in connecting case ground to output neutral.

5. With an isolating transformer vessel protective earth wire should not be connected to shore power protective earth.

Nothing to stop a switched mode isolating unit from having a shield either. !

Dave

[transmitterdan]

Originally Posted by goboatingnow

I would argue that 4. Isn't correct electrically. The case of the isolating transformer should be connected to shore side earth. This is the only device onboard that has such a connection to shore power and hence is technically part of the earth referenced supply

The output is floating , so no advantage in connecting case ground to output neutral.

5. With an isolating transformer vessel protective earth wire should not be connected to shore power protective earth.

Nothing to stop a switched mode isolating unit from having a shield either. !

Dave

I would not want the iso transformer case connected to shore earth. I think ABYC requires the case to be connected to boat safety ground and boat neutral just like a generator or inverter. If someone knows ABYC well please let us know what it says. I really want to know this for sure.

[goboatingnow]

Originally Posted by transmitterdan

I would not want the iso transformer case connected to shore earth. I think ABYC requires the case to be connected to boat safety ground and boat neutral just like a generator or inverter. If someone knows ABYC well please let us know what it says. I really want to know this for sure.

ABYC does require such an arrangement , but it never made any electrical sense to me. If the case goes hot from the shore power side ( primary) , there is a path to ground , not protected by the incoming shore earth. Bad news. The alternative , hot from secondary , will not generate a shock hazard.

Dave

[DotDun]

Originally Posted by transmitterdan

I would not want the iso transformer case connected to shore earth. I think ABYC requires the case to be connected to boat safety ground and boat neutral just like a generator or inverter. If someone knows ABYC well please let us know what it says. I really want to know this for sure.

Solution is to use a non-metallic case.

[transmitterdan]

Originally Posted by goboatingnow

ABYC does require such an arrangement , but it never made any electrical sense to me. If the case goes hot from the shore power side ( primary) , there is a path to ground , not protected by the incoming shore earth. Bad news. The alternative , hot from secondary , will not generate a shock hazard.

Dave

This makes sense. The protected person is in the boat. If the potential between neutral and case is near zero then all is well. But if case is connected to shore ground and not boat neutral/ground then the potential from boat ground to case could be anything up to deadly voltages. I always do it like ABYC says as they have thought long and hard.

Even if there were a short from case to the shore hot or neutral the person in the boat would be ok but a nearby freshwater swimmer would not. That requires a GFCI at shoreside to prevent in any case.

[goboatingnow]

Originally Posted by transmitterdan

This makes sense. The protected person is in the boat. If the potential between neutral and case is near zero then all is well. But if case is connected to shore ground and not boat neutral/ground then the potential from boat ground to case could be anything up to deadly voltages. I always do it like ABYC says as they have thought long and hard.

Even if there were a short from case to the shore hot or neutral the person in the boat would be ok but a nearby freshwater swimmer would not. That requires a GFCI at shoreside to prevent in any case.

Sorry the output has no path nor is referenced to earth, hence the comment about potential from ground to case is irrelevant

If the case isn't connected to shore earth , any shore hot fault creates a unprotected shock fault by touching the case. That's just bad practice

In my lab, there are several transformer based power supplies , all have isolated outputs. The case is always connected to earth not the output. Neutral.

Abyc makes no sense in this regard and they haven't thought long and hard ( for example they only recently discovered all boat RCD. )

[CharlieJ]

goboatingnow
You are more than welcome to disagree but your facts regarding the deliberations by the ABYC Project Technical Committee are just wrong. If you want to facilitate change to the E-11 Standard, which I paraphrased in #95, you are more than welcome to submit your comments during the next review cycle. Contact me by IM and I will provide you with the procedure for becoming actively involved with the ABYC Standards.

Regarding RCD (ELCI) implementation: this was discussed for several years to align better with the EU. Implementation was delayed because the manufacturers had a difficult time developing an ELCI that was cost effective and that was not subject to nuisance tripping when a VHF radio was keyed.

If the case isn't connected to shore earth , any shore hot fault creates a unprotected shock fault by touching the case. That's just bad practice

I do not understand this as written. Are you postulating that a fault to ground on the shore side of an ABYC compliant system creates a shock hazard by touching the isolation transformer's case? I do not understand how as the case is isolated from both sets of windings and a "shore fault" will trip the upstream circuit breaker.

[goboatingnow]

Originally Posted by CharlieJ

goboatingnow
You are more than welcome to disagree but your facts regarding the deliberations by the ABYC Project Technical Committee are just wrong. If you want to facilitate change to the E-11 Standard, which I paraphrased in #95, you are more than welcome to submit your comments during the next review cycle. Contact me by IM and I will provide you with the procedure for becoming actively involved with the ABYC Standards.

Regarding RCD (ELCI) implementation: this was discussed for several years to align better with the EU. Implementation was delayed because the manufacturers had a difficult time developing an ELCI that was cost effective and that was not subject to nuisance tripping when a VHF radio was keyed.

I do not understand this as written. Are you postulating that a fault to ground on the shore side of an ABYC compliant system creates a shock hazard by touching the isolation transformer's case? I do not understand how as the case is isolated from both sets of windings and a "shore fault" will trip the upstream circuit breaker.

Charlie , the fact is the ABYC connection strategy is the subject of much debate see for example a very good treatise http://www.smartgauge.co.uk/iso_wire.html. Which convincingly argues that case and shield are connected to shore earth.

Note that RCDs ha e been on European boats for over 20 years no such difficulty exists as you describe and ordinary domestic units have functioned correctly. ABYC simply has a blind spot as NEC on 30ma whole boat ( house) RCDs

What I am arguing is that the outer case should be connected to shore earth and not boat earth, connected to boat earth , for example a fault on the shore hot inside, making the case live has no fault path . For the shore breakers to work , it would rely on the existence of a boat earth to water connection which may or may not exist.

To summarise the technical debates ( which rage throughout the electrical industry , not just boats )

In regards isolating transformers

(A) which is better , floating AC output no RCD on boat side ( well assume RCDs on shore power inlet as standard) no neutral

advantage that earth or seawater is not in the return path , disadvantage potential for common mode high voltage ( mind you in a damp marine situation!) , disadvantage hot faults to appliances or hull , cause raised potential and if there is a path to the other secondary , potential for shock ( rare )

(B) floating AC , but with protective fault wire , and RCDs , but still no local boat earth

Can be difficult to achieve , as the bonding of the protective wire to appliances can in effect create a local neutral , ie earth referencing the output. , advantages RCDs work , but by design or otherwise , earth return now exists and the major satey isolation feature is defeated

(c) earth referenced output. , by establishment of a seawater ground

Advantageous , looks like conventional supply , RCDs can be added to output to protect boat circuits ( if fact they should be added) , disadvantages , no isolation , galvanic currents have a path via seawater from shore ground to boat ground , galvanic isolators may be needed , difficult to know what advantage is actually achieved. This is the ABYC circuit


The fact is like the Op found , an isolating transformer may be nothing of the sort.

Dave

[CharlieJ]

no such difficulty exists as you describe and ordinary domestic units have functioned correctly. ABYC simply has a blind spot as NEC on 30ma whole boat ( house) RCDs

Dave, this is simply not true. I was involved in discussions regarding implementing RCD (ELCIs) on USA vessels and the facts, as I stated in #102 are, well, the facts. The major circuit breaker manufacturers had significant difficulty in producing an acceptable RCD for the US market. The requirement was finalized in E-11 about four years ago with an implementation date one year hence to give boat builders and the breaker manufacturers time to develop and install the equipment. When the breaker manufacturers could not meet the technical requirements, a one year postponement was initiated. When the ELCIs were being deployed about two years ago, it was found that a handheld VHF, transmitting on the dock, would sometimes cause a nuisance trip. The ELCIs had to be hardened against RFI/EMI to survive in the real world.

The fact is like the Op found , an isolating transformer may be nothing of the sort.

The fact is, any piece of equipment that is not installed in accordance with the manufacturer's installation directions has an excellent probability of not performing to specification. With the limited info we have from the OP, this is what I think occurred in this instance.

And I want to extend my previous invitation to you again:

If you want to facilitate change to the E-11 Standard, which I paraphrased in #95, you are more than welcome to submit your comments during the next review cycle. Contact me by IM and I will provide you with the procedure for becoming actively involved with the ABYC Standards.

To quote T. Roosevelt: "It's not the critic who counts, not the man who points out how the strong man stumbled, or when the doer of deeds could have done better...."

[transmitterdan]

Originally Posted by goboatingnow

Sorry the output has no path nor is referenced to earth, hence the comment about potential from ground to case is irrelevant

If the case isn't connected to shore earth , any shore hot fault creates a unprotected shock fault by touching the case. That's just bad practice

I strongly disagree. In fact it is the best practice. If the case is not connected to boat ground then there is the possibility of shock within the boat which is the most likely danger. The case must not be allowed to be at a different potential than the boat ground and neutral.

Even if the shore hot comes in contact with the case you will not be shocked so long as you do not touch the case and shore ground at the same time. You would need really long arms to do that. The probability of shore hot contacting the case is remote anyway.

BTW, this fault condition is what can cause drownings in fresh water marinas and why many marinas are installing GFCI or whatever the acronym of the week is. It can also cause corrosion to the boat. But this type of fault is many times less likely than the marina sparky criss-crossing hot and neutral. The iso transformer makes that mistake a non event which is why they are a great safety device.

Originally Posted by goboatingnow

In my lab, there are several transformer based power supplies , all have isolated outputs. The case is always connected to earth not the output. Neutral.

In your lab it is easy to touch the case and earth ground at the same time. Thus the case must be earthed. In a boat it is nearly impossible to come in contact with both at the same time.