Isolating Transformers, The Earth wire connection conundrum

[colemj]

Our previous Outback, Trace and (shudder) Xantrex inverter/chargers also had different options for connecting/not connecting the ground to neutral. Our auxiliary Exeltec inverter has this choice as a removable hard clip on the terminals.

Mark

[transmitterdan]

Originally Posted by s/v Jedi

Inverter/chargers often do that when they switch to invert mode. On the Victron unit I have, you can re-program this. The inverter is a power source, so it needs to create a grounded neutral for installations that require one.

To comply with ABYC recommendations most inverters and generators connect AC ground to AC neutral when there is no shore power. This mimics the shore power AC supply configuration in the US. On some units (e.g. Mastervolt) this feature is defeatable. But do any units internally connect DC minus to AC ground? I am not aware of any current production unit that does so. So far as I know the only way to connect AC ground to DC minus is to install a wire for this purpose. ABYC recommends this wire as I understand it.

[colemj]

I meant the AC ground to AC neutral. I know of no unit that connects AC ground to DC minus (although ours did during our lightning strike!). I missed that point about your question. This is done at the main panel if wanted or required.

Mark

[Oceanride007]

Meaured my boats potentials yesterday, unfortunately my DC Neg and AC earth are connected. My Victron charger (phoenix I think) has pass thru AC, I think in this mode, its dum. Our choice is shore and gen its done thru a selector breakers. If there is a wire between AC earth and DC Neg, how the hell can you find it.

[Oceanride007]

Sorry dum question, it only makes sense if we went for a IT and we wished to disconnect the link between Earth Ac and Battery neg.

[transmitterdan]

Originally Posted by Oceanride007

Meaured my boats potentials yesterday, unfortunately my DC Neg and AC earth are connected. My Victron charger (phoenix I think) has pass thru AC, I think in this mode, its dum. Our choice is shore and gen its done thru a selector breakers. If there is a wire between AC earth and DC Neg, how the hell can you find it.

ABYC recommends this wire. Some will connect it at the AC panel however some recommend connecting the AC/DC ground wire to the engine frame. It could be pretty much anywhere so you may want to consult your boat's wiring diagram.

Before you rip this wire out be sure you know why you want it gone. If you don't have an IT there are some safety considerations for leaving the wire intact. And if you have an IT or galvanic isolator there is no reason to remove it for corrosion reasons.

If you have an IT this AC/DC ground wire serves no purpose in my view. This thread is (or at least was) a discussion about this issue.

[goboatingnow]

Originally Posted by Oceanride007

.....

I like the idea of IT , but with the qualifier that testing should occur to ensure insulation has not broken down and do it before a second fault occurs. I also mentioned double pole switching on each circuit, nobody apart from Jedi picked up on the importance of that. (I suspect my other idea earth lamps won’t fly, but you may consider it.).

Note that double pole switching has been mentioned by several contributors

Note that insulation testing is a test in relation to common mode voltage not live to appliance faults. And to do insulation testing for example in medical IT system requires a good earth ground.

Our best solutions to problems were done at a time when there were little regs to get around, best solution is to encourage experiments outside the norm but done safely and knowledgably and get pear review somewhere otherwise the evolutionary process will not deliver the best system.

There are very few regs dealing with ITs on boats. And there are significant differences on approach better bodies like the eU and us on such matters. The literature is full of discussions about ITs on boats and arguments that ABYC is wrong etc ( see Calder , smart gauge etc )

At sea on a vessel usually with isolated neutral (therefore no hull/earth current), a fault is allowable to be created, so long as you have a method of detecting it. Isolated neutral and floating IT share a lot of advantages in that with a single fault it is still safe and no voltage differences around the hull to cause you the hull/fitting corrosion. The problem occurs cose you tie up and connect with a shore supply and have to comply with a reg/code that is rigid. Hence, I think the purpose of this thread.

iTs have no role except when connected to shore power. The regs we talk about are those in relation to shore power connectivity. But which regs should apply. If I home build an boat in the EU, I am not required to met any regs for example.

Could we have more talk on solution to satisfy the neigh sayers, especially around the area of AC equipment like chargers inverters that may inadvertently connect AC earth to DC neg and my prop shaft.

That's the beauty of an IT , inadvertent connection to DC is not an issue. Anyway this issue is totally overstated. invertors are Self contained power sources and has s nothing to do with this issue.

Battery chargers in my experience generally indicate such faults by emitting large amounts of smoke as a precautionary warning.

Dave

[Oceanride007]

Note that insulation testing is a test in relation to common mode voltage not live to appliance faults. And to do insulation testing for example in medical IT system requires a good earth ground.

Sorry don't get you, I'm talking insulation test either of the hot wires to the metal frame of what you are testing. With equipment isolated of course. But there is no reason why you can't do the whole system test. If you have bond wire.

There are very few regs dealing with ITs on boats. And there are significant differences on approach better bodies like the eU and us on such matters.
The regs we talk about are those in relation to shore power connectivity. But which regs should apply

I thought the deal was that if you pull into a US Marina, your system would/should comply with local authority.

That's the beauty of an IT , inadvertent connection to DC is not an issue.

Sorry, still catching up.

Common mode? I thought that two hots would be opposite in phase.

Thanks Gobo, good topic.

[goboatingnow]

Originally Posted by Oceanride007

Note that insulation testing is a test in relation to common mode voltage not live to appliance faults. And to do insulation testing for example in medical IT system requires a good earth ground.

Sorry don't get you, I'm talking insulation test either of the hot wires to the metal frame of what you are testing. With equipment isolated of course. But there is no reason why you can't do the whole system test. If you have bond wire.

There are very few regs dealing with ITs on boats. And there are significant differences on approach better bodies like the eU and us on such matters.
The regs we talk about are those in relation to shore power connectivity. But which regs should apply

I thought the deal was that if you pull into a US Marina, your system would/should comply with local authority.

That's the beauty of an IT , inadvertent connection to DC is not an issue.

Sorry, still catching up.

Common mode? I thought that two hots would be opposite in phase.

Thanks Gobo, good topic.

Insulation testing as the term is understood , is a test at very high voltage to ensure the insulation around a conductor is holding to specification.

What you are suggesting is different

A boat in a us marina would merely comply with its flag state construction rules. Equally I've yet to see where the species of on board wiring are actually US code as opposed to ABYC suggestions of good practice.

Common mode voltage is the potential the output of an floating IT can be raised against a standard ground , usually earth , even though no current can flow. All transformers have a common mode rating

Dave

[transmitterdan]

Someone asked for a list of acceptable recommendations when a boat has an isolation transformer (IT). What say we try to do that? Here are some requirements I think we should try to meet:

1) Risk of electro-shock drowning due to an IT equipped boat should virtually nil.

2) However rare you think the risk of on-board shock we should not be afraid to reduce that risk but balance against other hazards.

3) Nothing about the system should interfere with the boat's corrosion prevention system whether passive or active.

4) The system should not be prone to nuisance shut down. After all, food poisoning and sinking are real risks too.

5) Recognize there is no such thing as fool proof safety systems.

Here is a start, please read them all before writing a response:

1) The IT primary shall be connected to the shore power through a dual pole circuit breaker and 30mA RCD (or combination device having both functions). (This handles the case where a fault in the trafo or wiring might injure a swimmer.) The trafo shall have an electrostatic shield connected to the shore ground wire. The electrostatic shield shall have no other connection (i.e. it is insulated from primary, secondary and frame of IT).

2) The secondary L1 (hot) and L2 (neutral) is fitted with dual pole circuit breaker within 1M of the trafo and routed to the AC panel where all branch circuits shall use double pole circuit breakers (fuses are not permitted). Optionally the output of the secondary circuit breaker may be first routed to a manual or automatic transfer switch for inverter and/or generator. If so, only L1 and L2 should be switched.

3) All convenience outlets no matter where on the boat shall have 5mA RCD protection. The in-built test feature of each RCD device shall function normally.

4) All other branch circuits may optionally have up to 30mA RCD protection with functioning in-built test feature.

5) The third so-called ground wire shall be connected to L2 at the AC panel. This is so RCD in-built test feature will function.

6) The third so-called ground wire may optionally be connected to boat DC minus at a single point (e.g. propulsion engine frame). In this case, DC minus shall not have any fuses or circuit breakers.

7) Generators and inverters with changeover switches (auto or manual) shall not connect L2 to so-called ground.

I intentionally put in some that might be controversial and hope we can debate them. Please add anything I forgot (e.g. wire color scheme). I am not saying this is the only "right" way to do it. Also, let us not jump to saying our ideas violate one standard or another. Once we have a list we can see how it might fit into current or future standards. Let us pretend we are going to submit our recommendations to various standards bodies with a significant chance they will be accepted.

[DotDun]

If in #2 you wire the transfer switch on the primary side of the IT, #7 becomes a mute point.

[chala]

Originally Posted by pbiJim

If that 3'd connection on the outlet can be used to improve safety in some way

Yes it does. If an appliance is faulty, why waiting for a person to trip the RCD? Why having a person subjected to a fault current? The 3’ connection will make the RCD to trip as soon that the appliance become faulty.

[goboatingnow]

Originally Posted by transmitterdan

Someone asked for a list of acceptable recommendations when a boat has an isolation transformer (IT). What say we try to do that? Here are some requirements I think we should try to meet:

1) Risk of electro-shock drowning due to an IT equipped boat should virtually nil.

2) However rare you think the risk of on-board shock we should not be afraid to reduce that risk but balance against other hazards.

3) Nothing about the system should interfere with the boat's corrosion prevention system whether passive or active.

4) The system should not be prone to nuisance shut down. After all, food poisoning and sinking are real risks too.

5) Recognize there is no such thing as fool proof safety systems.

Here is a start, please read them all before writing a response:

1) The IT primary shall be connected to the shore power through a dual pole circuit breaker and 30mA RCD (or combination device having both functions). (This handles the case where a fault in the trafo or wiring might injure a swimmer.) The trafo shall have an electrostatic shield connected to the shore ground wire. The electrostatic shield shall have no other connection (i.e. it is insulated from primary, secondary and frame of IT).

2) The secondary L1 (hot) and L2 (neutral) is fitted with dual pole circuit breaker within 1M of the trafo and routed to the AC panel where all branch circuits shall use double pole circuit breakers (fuses are not permitted). Optionally the output of the secondary circuit breaker may be first routed to a manual or automatic transfer switch for inverter and/or generator. If so, only L1 and L2 should be switched.

3) All convenience outlets no matter where on the boat shall have 5mA RCD protection. The in-built test feature of each RCD device shall function normally.

4) All other branch circuits may optionally have up to 30mA RCD protection with functioning in-built test feature.

5) The third so-called ground wire shall be connected to L2 at the AC panel. This is so RCD in-built test feature will function.

6) The third so-called ground wire may optionally be connected to boat DC minus at a single point (e.g. propulsion engine frame). In this case, DC minus shall not have any fuses or circuit breakers.

7) Generators and inverters with changeover switches (auto or manual) shall not connect L2 to so-called ground.

I intentionally put in some that might be controversial and hope we can debate them. Please add anything I forgot (e.g. wire color scheme). I am not saying this is the only "right" way to do it. Also, let us not jump to saying our ideas violate one standard or another. Once we have a list we can see how it might fit into current or future standards. Let us pretend we are going to submit our recommendations to various standards bodies with a significant chance they will be accepted.

Would agree with most , however the outlet GFCIs are a bit over kill.

Note that running a ground wire to L2 without establishing a local seawater ground is tely of very little use and floats the case of every appliance up to. The IT float voltage

AC to DC connection is virtually useless, since earth ground is not in the circuit

Really it boils down to ABYC where all on board power sources including ITs and invertor generators are earth referenced OR fully floating outputs where ground wiring is essentially superfluous

Dave

[Lake-Effect]

Just a reminder on electro-shock drownings (ESD). The hazard for ESD is created when the following two fault conditions have occurred:

1 - AC hot (from shore) has somehow contacted boat ground, and
2 - AC ground is not connected to boat ground

Without condition 2, the fault in #1 would cause breakers and/or GFIs to blow.

So whether or not there's an isolation transformer on the boat, it's still possible for the shorepower hot to get shorted to boat ground somehow (eg chafe at a metal pass-thru, or a screw through the AC lead-in, etc)

[Saltyhog]

Something that's been siting in the back of my mind during this entire discussion is safety considerations of the appliances themselves. When I say appliance, I'm speaking of anything with a power cord which may be powered via a mains outlet. Generally, appliances must meet safety regulations which require two faults to render an unsafe condition. This two fault scenario is generally met by designing the appliance to be either safety ground wire protected, or double insulated. Safety ground is the third wire (green). I'm calling it safety green wire here to differentiate it from ground. It may or may not be connected to earth or ground on a boat. (That's the crux of the discussion here.) In an appliance using safety green wire design, normally non-conducting metal is connected to the safety green wire. In the event of a failure of the insulation or other inadvertent connecting of voltage (hot) to the green safety wire protected components, full rated current of the OCD (panel breaker) is exceeded and the circuit opens (breaker trips). Generally, metal chassis's, enclosures, etc. are connected to the safety green wire lead. Two faults are needed to make the appliance unsafe: 1) The insulation failure or inadvertent hot to safety green wired parts, and 2) Failure of the safety green wire connections themselves. The other route for double fault protected appliances is double insulation. These appliances have no safety green wire in their power cord. Double fault protection is provided in the design. Often, there are no accessible metal components. Often, the mechanical design of the appliance precludes a single failure resulting in an unsafe condition. With all that said, now to the the crux of my concern. I believe the suggestion is being made to not bother with the safety ground wire and thus not connect it to L2, as it provides little protection in an isolated full floating mains supply (via IT). So just wire your boat with 2 conductor cable, double pole breakers, and bob's your uncle. This is fine for double insulated appliances. For safety green wire appliances, one fault is (by intent) pre-existing; ie: the safety green wire is broken (non-existent). This is argued as not mattering as there is no return path since L2 is not connected to anything locally (un-grounded). OK, fine, but I'd still like to know if my appliance has faulted in some way, and there is no (passive) way to tell here. Another, more practical consideration is that many electronics will use the safety ground wire as a high frequency dump lead. This is not really good design on the part of the manufacturer, but it's generally a cheap and dirty way to clean up EMC issues with cheap consumer grade electronics. If you have a three wire power cord on a computer or TV or other electronics with cheap switch-mode power supplies, disconnecting the third wire (safety green wire) will often result in them emitting more crap than they should, and they may then interfere with other stuff. So, if I were going to wire my boat with an IT, and fully float the output with no safety green wire, I'd be sure to only permit double insulated appliances aboard. It seems many appliances manufactures are moving this way, so it may be a reasonable approach. Just my (admittedly lengthy) 2 cents.