After doing some electrical work I now get a mild shock when using the
bathroom tap. It's a TT system. I have a couple of questions.
1. Am I looking for a live/earth fault only.
2. What's the best way of checking the taps etc. to see if the fault
persists after I find the fault(s).
3. The RCD didn't trip since I was working on lighting circuits. What can
I fit to my consumer unit to prevent/alert me to such faults in the future.
If it's only a mild shock, it sounds as if live is connected to the tap
through a high resistance (e.g. brickwork) and that the tap isn't properly
Have you measured the (ac) potential between the tap and a proper earth? If
so, re-do this after you have fixed the fault, and it should be zero.
I had a problem a few years ago where a builder had screwed into a live wire
when fitting a skirting board - making the whole wall slightly live. This
only came to light when we were wallpapering the wall - because the dampness
of the paste made the wall conductive, and we were getting tingles off it.
Maybe your problem is similar?
That's not a safe assumption -- it might be that Niel was not well
earthed, and if he was, then a full mains belt is readily available.
This situation is not something to treat lightly -- the potential for
a fatality might already be there. If you are not fully familiar with
electrical fault finding and you have a problem of this nature, I would
suggest getting an electrician in and keep well away from any areas you
have found with this problem until it's been fixed.
This reminds me of the chap electrocuted in his bath a few weeks ago
when the showhead fell into the water.
It might have to be on, it only needs other circuits in the house to be
used, to create the back feeding effect along that broken/loose conductor
and give the tingle.
Just a few questions to start. Is your consumer unit a spilt board with
sockets, cooker, shower RCD protected and the other ccts non RCD protected.
When working on your lights even with the lighting mcb off did your rcd trip
when you touched the neutral to any other bare wires Are your water and gas
pipes earthed from your earth block in cu where they enter your property ? And
do you know if any plastic fittings ( elbows t pieces etc ) have been added to
your copper ?installation at any time ?
Firstly, seeing that you're posting via caltech.edu, can you confirm that
you're in the UK?
Secondly (assuming that you are in the UK) it's most unusual and potentially
VERY DANGEROUS to have non-RCD protected circuits in a domestic TT
installation (or ELCB protected in the case of older installations). It's
not actually banned by BS 7671, but in the event of an earth fault on a
lighting circuit you're relying on having sufficient current flow to earth
to blow the fuse or trip the MCB. This means that your earth electrode
resistance has to be very low indeed, requiring an extensive earth system.
Referring to Table 2D in the OSG you'll see that the measured earth fault
loop impedance for a circuit protected by a 6A Type B MCB must be 6.4 ohms
or less. Given that the external (supplier's) component of this, Ze, is
allowed to be as high as 21 ohms you can see the problem! Even with a
typically low suppliers Ze of an ohm or two, your earth system would need to
be under about 4 ohms, and you will not achieve this with the two rods that
you mentioned in another thread.
I conclude that your installation may be inadequately protected and
therefore very dangerous. Whoever has caused the installation to be like
this, by modification or whatever, is seriously negligent in a way that
makes me think that Part P might be a good idea after all. The 'quickest
fix' would be the insertion of a 100 mA Type S (time delayed) RCD in a
separate enclosure in the meter tails, between the meter and exisiting
There isn't sufficient information to diagnose the fault symptoms that you
described, but _one_possibility_ is that in modifying the lighting circuit
you've introduced a phase-earth fault (e.g. trapped a wire and nicked its
insulation in a junction box) which is not being cleared due to the lack of
RCD protection combined with the earth fault loop impedance being too high
to operate the overcurrent device. Such a fault would make all your
'earths' live (and your earth rods hot, and your electricity bill ...). If
you have neighbours who are also using TT earthing then its possible for the
fault described to make their earths live too, via connection through bonded
metal water pipes.
This needs sorting urgently. A simple test to start with would be to
isolate the lighting circuit in question (pull fuse or turn off MCB) and see
if the shocks from the tap stop.
First it might be an electrostatic shock, in which case no worry. But
if it isn't:
1. it is likely to be life threatening
2. its the result of multiple faults, not just one
3. it needs sorting very quickly.
4. it only takes one bath to kill.
this wont fix it actually, the pipework would still be unearthed, the
earth leakage still present, and the level of RCD protection
inadequate to protect against this fault.
How do you know the pipework is "unearthed" (what ever that might mean)?
The OP mentioned having added supplementary bonding.
Have you got the faintest clue what you're talking about? We know that the
OP has an earthing system consisting of two earth rods. What grounds do you
have for thinking that the earth electrode resistance is greater than the
500 ohm upper limit required to provide adequate protection against indirect
contact when a 100mA RCD is in use? The chances are that it will be less
than the 200 ohm guideline figure in the OSG, which includes a large factor
of safety. The RCD will trip well within the permitted disconnection time
(5 seconds in this case, since we're considering a fixed equipment circuit).
Actually there _might_ be a simpler way to add RCD protection to the
lighting circuit(s) in question and that's to change the MCB(s) in the
existing consumer unit - if it is an MCB type, that is - to RCBO(s), space
I guess you and I are thinking of 2 different scenarios, either of
which could be true. We know the OP has earth rod/s, and we know he's
getting zapped off the bathroom piping.
Option 1: there is no RCD/ELCB protection on some of wiring - that
would surprise me, as its a no-no with local earth rods, and has been
for quite a long time.
Option 2: the pipework isnt earthed.
I think so. Both the above look like viable options from the info I've
got so far. And even if they weren't due to some oversight, I think
I'd still have [at least] a faint clue what I'm talkin bout.
That comment is correct for the option of the pipework being
unearthed, but not if its earthed but no RCD/ELCB in circuit.
I'm not saying that. Whichever scenario is correct, fitting a whole
system RCD isn't going to solve the problem.
That would be rather better, tho again it wont be a solution. The only
solution is to
a) fix the earth leakage problem
b) fix the RCD protection
We also know that the shocks started when Neil made alterations to a
lighting circuit which he has told us is not RCD protected.
It shouldn't surprise you because he has told us that there is no RCD
protection on the lights. Yes, that's a "no-no" - that's whole point!
Can I remind you that this pipework doesn't need to be earthed. There has
to be main bonding to incoming services, and local supplementary bonding in
bathrooms. The latter may quite legitimately end up being electrically
floating if there is plastic piping involved and no link back to the main
earth terminal (as would happen if the local bonding only involved
Well, of those options, No. 1 is not in itself a reason for shocks, but it
would allow all his earthing to sit at (almost) full mains voltage
indefinitely if an earth fault were to occur on the non-RCD protected
circuit. That the latter had occurred was my supposition. Again, option
No. 2 is not a reason for shocks from taps - although a lack of local
supplementary bonding combined with some other fault might be, but I don't
think that's relevant in this case.
Which comment? You've lost me.
Now I have to disagree. I think that the problem is an earth fault (or a
light wired between phase and earth, or something like that) which is not
being cleared because there's no RCD on this circuit and Zs isn't low enough
to cause enough current to flow to trip the MCB. Therefore his earthing is
live, therefore his taps are 'tingly' - the shocks are from current flowing
to earth via body capacitance. If the local bonding is doing its job it
will prevent a really serious or lethal shock.
The lighting circuit(s) need to have RCD protection added. A "whole system
RCD" as you call it is one valid option, and it needs to be a 100mA Type S
one in order (a) to avoid nuisance tripping of the lights and (b) to permit
discriminative operation of the other (30mA, fast) one downstream.
Yes, I was taking that as read.
Precisely - that's what I said in the first place :-)
Well Andy, I disagree on a number of the details you brought up, but
the OP should first sort out the earth leakage fault, and ensure all
circuits are RCD protected. That's the real point.
About the earth bonding. I think that this will not reduce a shock, but
only reduce it to an acceptable (less than 50V) level. In fact, if I
hadn't put supp bonding in the bath then I wouldn't have been shocked.
The loose wire touching the light box must have caused a current too the
earth bar of the CU and hence through the earth and supp bond system.
Option 3, there is an electric fence system with an earth common with
the mains earth (yes I know it's against regulations). I suffered
from this until I realised the error of my ways and separated the
electric fence earth. It gave me several quite noticeable shocks of
the pipework under the sink one day when I was working there.
HomeOwnersHub.com is a website for homeowners and building and maintenance pros. It is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.