ABYC Presentation on Electric Shock Drowning

[kenn]

Recent fatalities have brought increased attention to the issue of injury and death which result from paralysis or electrocution caused by AC wiring faults in marinas or boats that energize the water.

In response, the ABYC prepared and presented a 2-hour webinar on the issue. The presenter Kevin Ritz started right in with the details and analysis of the ESD death of his own son, as well as some other typical occurrences. This was followed by a detailed technical analysis of the causes, and included techniques for detecting potentially lethal shore-power faults.

Mr Ritz also outlined the regulatory steps that are being proposed in the US to prevent AC faults from energizing boats and the surrounding water. He pointed out that in Europe, for almost 30 years, ground fault protection has been mandatory for all shore-power AC connections, and ESD is now virtually unknown there.

I caught the presentation on Wednesday, and it was riveting, and very informative. Demand for this information has been very strong, and the ABYC has responded by making the recorded presentation publicly available. Below are the details as posted on the ABYC website:

SAFETY ALERT - ELECTRIC SHOCK DROWNING
In response to the recent Electric Shock Drowning tragedies and overwhelming need for information, the webinar Electric Shock Drowning...The Invisible Killer will be available until August 19, 2012. It is presented by ABYC certification instructor and Master Marine Technician Kevin Ritz and first aired July 19, 2012 to alert the unsuspecting to the dangers of this Invisible Killer. This is an essential and educational webinar covering the basics of electric shock drowning unfamiliar to many marine professionals.

Please view the presentation and help spread the word.

 

[walt]

Something Ive always wondered about. In the US, likely most marina's have 110 VAC (at different amp service but its usually 110).

The best place for ground fault detection would be at the dock where you plug an extention cord to the boat. So no matter what bad wiring is done on the boat, the dock ground fault protection keeps bad things from happening.

But in the US with the way the wiring is set up where one current carrying wire is at 110 and the other is near ground, this creates an big imbalance in "leakage" current from parasitic capacitance in long extension cords (which are often used at docks). That imbalance probably causes a lot of false ground fault trggers and the hassle of this could cause people to just disable them. I dont know if this really is an issue but it seems that it could be.

The figure hopefully illustrates this. In Europe where they use a symetrical 220 supply, they would not have this issue and I would guess ground fault false trigger is almost no problem.

 

[tommays]

What i see is the same stuff i deal with in a work wash-down environment

It starts with using a cheep twist-lock plug and and no matter how many "waterproof" boots they try and put on them there just NOT safe in a wet area.

And then we do not bother with GFI protection shore-side and its just downhill from there

 

[kenn]

The best place for ground fault detection would be at the dock where you plug an extention cord to the boat. So no matter what bad wiring is done on the boat, the dock ground fault protection keeps bad things from happening.

But in the US with the way the wiring is set up where one current carrying wire is at 110 and the other is near ground, this creates an big imbalance in "leakage" current from parasitic capacitance in long extension cords (which are often used at docks). That imbalance probably causes a lot of false ground fault trggers and the hassle of this could cause people to just disable them. I dont know if this really is an issue but it seems that it could be.

Actually it's a simpler situation than that. It's covered in excellent detail in the presentation.

In a properly wired shore-power situation, consumed current is only carried in the black (hot) and white (neutral) , and the current in each wire is equal. The ground is there for protection and there should be NO current on it.

So, if you put a loop current ammeter around the whole shore-power cable, you should see NO current reading, because the currents in white and black wires cancel out, and there's no current in the ground wire.

When a ESD-capable fault exists, the water has somehow become a return path for current, so when you put the loop current ammeter around the shore-power cable, you will see a current reading equivalent to the current being dumped into the water. This same principle (detection of net current) is what's used in the proposed breakers, and as mentioned, there's 30 years of success in Europe to draw on. The trip threshold will be above the miniscule currents resulting from cable capacitance.

Do view the presentation, it will completely answer your questions.

 

[walt]

Ken, I think my parasitic capacitance example is correct (but I could be wrong).

As you said, the current in the hot and neutral should be the same when there is no problem. If the hot wire were to come in contact with the water, some of the current from the hot wire would go to ground - and not return through the neutral causing the imbalance.

Note that in a three wire extension cord, there will be a capacitance from each of the two current carrying wires to the ground wire simply because the wires run close to each other for a long distance.

The parasitic leakage current of a capacitance will be proportional to the drive voltage and the capacitance (and frequency but that is constant here). The capacitance is about the same for both hot and neutral to ground in the cord. However, the drive voltage is WAY higher for the hot wire with respect to ground in the US 110 VAC system. Ie, the hot wire has close to 110 VAC to ground but the neutral wire has close to zero. So the hot wire will have also way higher leakage current to ground. A ground fault circuit back at the source would then measure a different current between hot and neutral. Even if this imbalance doesnt trigger the groud fault, it lowers the threshold.

I didn’t do this (I might) but I think google can find some references of long extension cords - or long wire runs - tripping GFI.

As I mentioned before, Europe does not have this issue since both current carrying wires have the same potential to ground.

I would like to view the presentation. Is the link presentation 2 hours (a little long..)

 

[kenn]

Walt:
the proposed NEC protective device is NOT your garden-variety GFI. It senses NET current across ALL conductors. Leakage/parasitics cancel out. Any measured current must be getting back to ground via another path (the water)

The presentation is 2 hours, and it's one of the better ways to spend 2 hours, other than sailing. Do watch it.

 

[tkanzler]

Thank you for that. I've been looking for something like this, but while everyone 'knows' about electric shock drowning (and actual electrocution), there is relatively little information out there written in lay terms. This presentation was very well done, though sad in that the presenter's son losing his life through electrocution was the impetus for it.

 

[Don12364]

.

 

[Don12364]

Walt - Twisted Cable Shunt Self Capacitance.

Walt,
Your diagram is more suited to the capacitors normally found on the input filter of battery chargers and microwaves etc, though they often include inductors as well. These capacitors are often in pi filter arrangement and need to be added together to represent the extent of out of balance leakage current.
You are quite correct in stating the capacitance on the hot side passes vastly more current to ground than the neutral side does. For 110v @60Hz it would need 0.7µF to bleed 30mA to ground and trip a 30mA RCD (GFCI) so this is often the case.
Twin twisted cable shunt capacitance can be calculated in this link:- http://www.mogami.com/e/cad/electrical.html

In the case of a twisted mains cable the shunt capacitance is in the region of 20pF per foot between conductors so the mains cable would need to be approx 35,000 feet long in order to have a shunt capacitance of 0.7µF.
For information I am a Brit and when I first read of electrocution incidents to swimmers in American marinas I quite frankly did not believe it happened. I guess it is that we have 230 volts that makes us a darned sight more cautious. Hence the legal requirement right across Europe that every marina outled is protected by a ground fault device.

 

[walt]

Donalex, cant argue with your numbers which would say nuisance GFI tripping shouldn’t be an issue in the US from long extension cords.

But.. I still wonder if actual experience from people who have boats at Marina's in the US with GFI dockside experience false or "nuisance" tripping. Don’t know.

FYI, I believe this is the same presentation but no password required http://www.youtube.com/watch?v=O7-s_mdEPb0&feature=youtu.be

What I saw in that presentation was measuring dock extension cords with a hand held current meter as a way to tell if the boat was leaking current to the water. The current meter would measure zero if the net current in the three wires going out was the same as the net current coming back. This will find a condition where the boat is leaking current to the water which of course is the key safety issue here and a great way to spot check a marina. However, we could still have a ground fault condition internal to the boat where for example the hot current was 5 amps, neutral was 3 amps and ground was 2 amps. The meter placed around the extension cord would show zero since 5 amps = 3 amps plus 2 amps. This is a true ground fault condition that the meter on the extension cord didn’t find because the leakage wasn’t to the water. But a GFI dockside socket would trip for this condition since the GFI is comparing hot to neutral and in this case would find 5 amps out, 3 amps coming back - and would trip. And of course the GFI would also trip for a condition where current was leaking into the water.

I’ve never been in a marina where I used power so don’t know if they are all GFI dockside protected in the US (trailer sailor budget). Seems a no- brainer to put these in as dockside GFI would have prevented pretty much every example in that presentation and it doesnt rely on the boat to be wired correctly. The only reason I could think of that GFI is not in every marina would be that they make themselves a pain in the axx by nuisance tripping???

 

[kenn]

Walt - the video you linked to is an earlier presentation, but I imagine the technical content is the same.

 

[walt]

FYI
This morning I measured the capacitance from hot or neutral to ground of a 50 foot 14 gauge hardware store extension cord..

.0012 uf.

Using a simple model (probably the same as Donelex used), a 100 foot extension cord leakage between hot and ground would be about 100 ua.

Even if the model is way off, 100 ua is so far down from where the GFI trips (a factor of 250 for a 25 ma trip) that the imbalanced capacitance problem I brought up likely is NOT a real issue.

I.e., the issue I first brought up about false GFI tripping because of long extension cords in the US - is likely not an issue at all for at least the capacitance model.