Ship's electrical 1

[pamungkas]

all,
I am working on safety audit on our shiping contractor. I didn't find any ELCB/RCD to protect their seaman. They use 3 phase generator with 3 cables only to the main busbar.Is it possible to install ELCB/RCD for 1 phase outlet/receptacle (hand tools purposed)on this installation? All ELCB I ever known require neutral and grounding which I coudn't find in the ship.

Genset spesification : ship 1 :3 phase 225VAC/110VAC
ship 2 :3 phase 380VAC/220VAC

Thank you.
rgds,

 

[ScottyUK]

An RCD does not require a neutral - it will work perfectly well with a single phase supply taken across two lines, or with a three phase, three wire supply. Both require that the source is grounded so that leakage to earth is possible.

Ship 1 sounds like it is either using a centre-tapped single phase supply or has a separate transformer for the 110V system.

In both cases, how is the neutral grounded to the hull? This has enormous significance in determining how well an RCD can protect you.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[lananh]

I agree with ScottyUK. An RCD does not require a neutral. i think we can use the hull to ground for this system
Regards.

 

[waross]

Many shipboard systems isolate the electrical systems from ground. A grounded electrical system on board ship may lead to serious hull corrosion when running on shore power.
I would not assume the liability of grounding an ungrounded ships electrical power system.
Nothing will protect a person from contact with two circuit conductors of differing potential.
However a ground fault circuit breaker will give protection against lethal currents through the body to ground even on an ungrounded system.
Ship grounding has been discussed on Eng-Tips in greater detail.
respectfully

 

[ScottyUK]

Hi Bill,

On an ungrounded system how would current return to the source via an earth leakage path so as to cause imbalance within the RCD / GFCI and trip it? I'm no expert on marine wiring and although the use of ungrounded installations on a ship doesn't surprise me, I am curious how an RCD would provide protection in this case.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[waross]

Hi ScottyUK;
It provides almost the same level of protection as it does on a grounded circuit. There is always almost enough capacitive current in a system of any size to give you any thing from a good tingle to serious injury, depending on the size of the system, the capacitance of the cables and the presence of any high resistance paths to ground.
This current is distributed throughout all the wiring in the system.
If you contact a hot conductor and ground, only a small amount of the ground current will return through the circuit you have contacted. Most of the current will return to the other hot lines through the impedance to ground of the other circuits.
The Ground Fault breaker will see this unbalance and will trip if the current is higher than the breaker set point.

I said almost as good protection.
On a conventional grounded system, the breaker will not protect you from contact with two hot wires of different phases. However, if your body is grounded there is a chance that enough current will bypass to ground to operate the ground fault feature of the breaker. You can't count on this, it comes under the heading of "sometimes you get lucky".
You can get a serious shock to ground on an ungrounded system. A ground fault breaker will usually "SEE" most of this "shock" current, but you can't gaurantee it.

Re: grounding of ships systems;
consider this scenario;
A ship with an ungrounded system ties up and connects to shore power. For some reason their system is drawing a heavy neutral current. There is a long shore power cable back to the distribution center and the grounding grid of the shore system. In a large dockyard, the grounding may be an extensive grid. That is, a lot lower resistance to ground than if a few ground rods were driven. The proximity to the ocean also tends to favour good grounding conditions. The neutral current is causing a voltage drop on the neutral and neutral potential out on the dock is not at true ground potential.
This is no problem as the neutral is not grounded to the hull.
Now, another ship ties up alongside and goes on shore power.
They are tied onto the same cable as is feeding the other ship.
This second ship has a grounded neutral system. You have now connected the neutral to a very large ground electrode. The hull of the ship. In some cases most of the neutral current will flow through the hull, through the sea water and back to the ground grid. This may cause hull corrosion and will definitely shorten the life of the anti-corrosion "Zincs".
By the way, there will probqably be no corrosion issues with the ship whose unbalanced load and heavy neutral current is causing the corrosion.
yours

 

[ScottyUK]

The dockyard scenario is interesting - hadn't thought of that!

UK (shore-based) regs don't allow ungrounded installations except in very specific installations. Grounded installations with permanent low impedance neutral earthing is one of the fundamentals our wiring regulations are based around. Reading between the lines I sense that the use of an RCD on an ungrounded system is an additional level of protection which in a practical installation offers some benefit but which isn't recognised by the standards writers, not over here at least.


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[pamungkas]

The scottyUK's question at 27 Jan is also question. I've tried install 1 phase ELCB (2 pole, Line-neutral), but since there is no neutral, both poles are life wire. I test the installation using 2 methods:
(1) Using trip button built in the ELCB
(2) Using ELCB tester, by connecting life wire to ground in receptacles (both life sockets have been tested, but all is failed).
By first method, the ELCB was trip successfully, but it failed during second test. I assume the ELCB did not detect any unbalance (30 mA) in this circuit. It seem
I am afraid that the accident which reported in this hyperlink was happened (

http://www.mardep.gov.hk/en/publication/pdf/mai050526_f.pdf).

The tug boats have certification, which state that everything in tug boat is safe, even without ELCB in their receptacles. The boat’s owner said that it means the boat has complied marine standard.
thanks

 

[waross]

Several points;
One thing that I strongly suggest that you do before we discuss these issues much more is to measure the current to ground that will flow in both these systems.
Step one; Make sure the system is free from grounds.
Stepm two; Use a fused switch and connect each phase in turn to ground.
Step three: Measure the current that flows to ground from each phase. It may be an amp or more, it may be milliamps.
Now you will have an idea of the current that a worker will be subjected to if he contacts a hot wire and ground.
If it is near or less than 30 ma. you will not benifit from a 30 ma. GFI.
Comments
One; Don't use a single circuit, line-neutral GFI for two hot leads. It is not safe as it will only disconnect one lead and leave the other "Live".
Use a proper two pole GFI breaker which is based on a two pole breaker rather than one pole and a neutral pass-through.
Two; You can't test it on the bench. It must be connected to the system under considferation as the ground leakage and the ground capacitance of the system will affect the operation.
I assume that the GFI testor is basically a resistor that will pass enough current at rated voltage to cause the GFI to operate. On an ungrounded system the testor will be in series with the system impedance to ground and will probably not draw enough current to cause operation of the breaker.
The point is; the GFI will operate when the difference between the current in one line and the other line is greater than 30 ma. or whatever the setpoint is.
The GFI breaker will operate on an ungrounded circuit if the current passing to ground is more than the setpoint (30 ma.)
You must test it with a milliameter and a variable resistor. If the system is small enough that the ground leakage is less than 30 ma. the hazard is slight.

RE; the accident.
We are discussing normally ungrounded ships systems. The accident was on a grounded system. Had the system been an ungrounded system, it is possible that there would not have been a reportable incident much less a fatality. The size of the generator is not given nor is there any indication of the extent of the wiring installed onboard. I suspect that it is a small generator and the use of extension cords to run almost the length of the vessel indicates that there is not much onboard wiring. If so, then the leakage current could easily have been little enough to be imperceptible or cause only a painfull but not lethal shock.

No matter how good a system is, you can't garantee that someone will not connect a hot wire to an equipment grounding terminal.
No matter how much care is given to the proper selection and installation of GFIs if they are damaged and not repaired or if they are not checked regularly and/or not repaired when found to be faulty you have no protection.
The incident was regretable, all the more so because it was the result of a combination of human error and neglected maintenance.
In regards to your boat. In North America we have extension cords with a GFI device built into the receptacle end. If such equipment is available in your area you may consider using these type of cords onboard. To test them onboard, you must causew 30 ma. to flow to ground. A normal testor may not work properly.
These can't be garanteed but they will give a much higher level of safety to workers using portable tools and pumps. These activities are often the most prone to shock hazard for general workers.
I suggest that you test the ground leakage and if it is much over 30 ma., try some tests with a protected cord. Based on the results you find, you may or may not recommend GFI cords.
The usefullness of GFI will depend on the system and each system is different.
respectfully

 

[pamungkas]

Thanks a lot warross for the input. I got your point of testing.

thanks all.

respectfully.