In article ,
David Billington wrote:

Joseph Gwinn wrote:
I saw this article in The New England Journal of Medicine. Here is the
URL, unfortunately behind a paywall.

http://content.nejm.org/cgi/content/short/360/13/1363

The gist is that a danish doctor had a patient who was getting
inappropriate cardiac shocks from an ICD (implantable cardioverter-
defibrillator) while in the shower, which turned out to be due to bad
house wiring causing a 50 Hz leakage current from showerhead to floor
drain.

A current of 3.5 to 4.0 milliamps is not usually felt, but can spoof the
ICD, which looks for such signals lasting about 300 milliseconds.

This would be easy to check with a voltmeter. And to prevent by
running a ground wire from showerhead to drain pipe.

Joe Gwinn

IIRC in the UK all runs of conductive pipe, typically copper in my
experience, have to be earthed to prevent this sort of thing. Having
said that I do know a guy that did some DIY and had a section of plastic
pipe joining 2 copper piping systems in his house which left one section
unearthed and this situation arose where the unearthed section became
live, luckily no one was injured just s few minor shocks.

I've run into a parallel in the US, in the 1970s. Friends complained
that the sink in the bathroom off the kitchen generated sparks when
used. I didn't believe it, but they persisted, so I came over one day,
voltmeter in hand. Turns out there was 110 volts between water valve
and drain, with plenty of power. No idea why nobody was hurt. This
should have cooked the grease out of some luckless soul. Turned out
that the plumbers had broken a kitchen-light wire while installing the
2nd-floor bathroom drain, and the broken wire was resting on the copper
drainpipe, which was isolated from ground by an oakum&lead connection to
cast iron pipe.

I forgot to mention, but there is a RCM issue here as well. Old
machines are often leaky electrically, and so the frame of all machines
should be firmly grounded. There was a thread about this, but I don't
recall date of title (which may be unrelated). Anyway, I'd use the
voltmeter between cold-water ground and machine frames, with machines
running, to verify that the ground works.

Joe Gwinn