Voltage in shower waste!

[purplemaf]

Hi All,

I am at my wits end here so here goes...

On Monday I was called to a domestic premises with reported shocks occuring while the client was taking a shower. This was apparently first noticed over 2 weeks ago!

We proceeded to do all the relevant checks on Earth Bonding and various testing including IR on all circuits. Most circuits proved to be fine apart from a high Zs (3.06) on the upstairs ring. All bonding was proved to be sound. Install is a PME with a split board with only 1 RCD. Main incomer has a Zs of 0.26. I moved all circuits over to RCD protection to see if any tripping would occur....to no avail.

Upon further investigation we ran a lead out from the board to provide a means of testing in the shower...what is clear is that a voltage of 6v can be meaured between Earth and the actual water in the shower tray but ONLY WHEN IT IS DRAINING THROUGH THE WASTE! This voltage increases to around 18v at least as load is increased around the house.

Upon FURTHER investigation I have now discovered this to be the case on all 3 showers and all basins upstairs. With the basins I actually filled them up, released the plug then could measure the voltage as the water drained...NOT BEFORE. With the showers I ran the water and turned off then measured the voltage again as the water drained. All showers are hot and cold fed with no local pumps.

I have had every circuit apart and individually connected 1 circuit at a time in the board and found that 4 out of the 6 circuits seemed to bring about the voltage. I have also bizarrely placed a lead between the mixer taps and into the water in the plug hole while running the water. This always removes the voltage.

I am left with the only answer that somehow a voltage is finding its way through the waste water and once someone gives it a path then the shock occurs!!

I have had SSE out today checking everything their end....no problems. Contacted the NICEIC...no advice could be given! I've consulted 2 of my so called guru electricians who are stumped. I am hoping that tomorrows investigation of the external wiring will reveal something but I'm not holding my breath!

Somebody please tell me they have come accross this before!!

 

[SirKit Breaker]

I have not read the full thread, so i may be on the wrong track here, but you never know.

Is all the supplementary bonding in place where is should be? Irrespective of whether or not BS7671 says it should or doesnt have to be.

I have never had this fault, but i have had my fair share of head scratchers, and when i have eventually got to the bottom of it, the problem was usually staring me in the face. I am not trying to imply that you have overlooked anything, but i thought it worth a mention, we are all human, and frustration can fog the issue at times.

Cheers..........Howard

 

[Rockingit]

As the DNO have been, the seals will undoubtedly have fallen off the bullet and the main fuse has fallen out. This is a good thing as it means you can test to see whether you still have a voltage present with the installation 'dead' but the neutral still in place, something you can't do using the DP main switch. If you DO still have a voltage present then it's back-feed from somewhere outside of your installation, if you don't then it has to be yours, and start with everything off at the board then turn it all back on a circuit at a time until your problem appears.

 

[Lucien Nunes]

Umm, did anybody notice those IR readings in post #55:

The following IR readings were taken with each circuit completely disconnected on 500v - 20mohm. I literally had a lead poked into the waste with he water running.

Live to water
Lights Down - 0.62
Lights Up - Clear
Skts Down - 3.07
Skts Up - 0.63
Smokes - Clear
Cooker - 13.55

Earth to water
Lights Down - 0.45
Lights Up - 0.3
Skts Down - 0.07
Skts Up - 0.07
Smokes - Clear
Cooker - 0.08

Water to Met - 0.07
Water to Neutral Bar - 2.45
Water to Bond - 0.08

Now 620kΩ from line won't kill you and won't trip an RCD, and I'm not sure it's the direct cause of the shocks, but it's unusually low and a failure anyway. The fact that two unrelated circuits are coming in with similar low insulation is rather striking. Upstairs sockets and downstairs lights both have a tale to tell, I think.

 

[Rockingit]

Umm, did anybody notice those IR readings in post #55:



Now 620kΩ from line won't kill you and won't trip an RCD, and I'm not sure it's the direct cause of the shocks, but it's unusually low and a failure anyway. The fact that two unrelated circuits are coming in with similar low insulation is rather striking. Upstairs sockets and downstairs lights both have a tale to tell, I think.

I looked at those and they scream that something is very very not quite right about this whole affair. It's PME, right? So, the Neutral and MET are effectively one and the same, agreed? So, how come there's 2M difference in a reading? (assumes no RCD in cct). And if we're now looking at lights AND sockets....exactly how many screws do we think have pierced these three pipes which magically stop conducting before they all join together??

If this is a genuine thread, then I think maybe a fresh pair of eyes may be needed on the job.

 

[ArseyDee]

If suspect circuits are downstairs lighting and upstairs sockets, it seems fair to assume that at least some of the cabling shares a common route.

Could it be possible that nail or screw damage has occurred in one place to the cables and a waste pipe, causing a small water leak which not only is causing voltage to appear on that waste, but the water leak is also tracking down the outside of pipe to the external discharge point where it comes into contact with the waste water from other two pipes, thus causing voltage to appear in those too.

May not even be nail or screw penetration to cables or pipe. Could just be a combination of cable damage (rodents?) with a poorly connected and leaking waste from shower, basin, etc. This is tracking on outside of waste pipe as above.

Just thinking out loud. May be way off the mark with this thought process.

 

[Tridentsystems]

Think it's time to get some boards up. Test results are likely to change hourly and will tie you in knots. You'll end up chasing your tail

 

[Rockingit]

OP - what is your background / experience / testing knowledge?

 

[sendandreturn]

Hi, I'm Mark, the homeowner, just wanted to say thank you to all of you for your time here and input and of course to my sparky. Really humbled and grateful for helping us sort the problem. If any of you want to hook up via Google+/Skype video so I can show you around the house (if this is of any use), I happy to do that.

 

[alanjw]

how close it this property to the next one the fault my not be in the one your testing it might be an overlapping earth path from the next property.

 

[Rockingit]

How thoroughly checked have you for polarity reversal?

 

[LittleHelper]

Could it not be an induced voltage with cables running near pipework?

 

[Tridentsystems]

Looks like purple abandoned us for the IET forum

 

[KAS1]

I agree, it's also in full flow on the IET forum lol

 

[snowhead]

Could it not be an induced voltage with cables running near pipework?

Causing an induced shock?

 

[ArseyDee]

Looks like purple abandoned us for the IET forum

What? A defection by the OP?

The voltage is not always leaner on the other side you know!

Sorry, that's a pathetic attempt at humour on my part.

 

[purplemaf]

Lol sorry people! I've not abandoned you honest! Just getting as much info as pos. If you've not seen yet then I'll enlighten you.

Today I found 25v on the metal studwork in 2 seperate places oon the 1st floor. Appeared to be directly energized by the lighting circuit. However once the lighting circuit was completely removed the other circuits combined brought back 7v. Af least now I have metal to test to instead of water! To experiment I put a bond to the metal work in one room which removed the 7v however it remained in another room. So this does explain why problem is upstairs only as the 2 floors are not connected with metal stud. Hunting on...

 

[purplemaf]

Very grateful for all the assistance so far.

 

[Phoenix]

Cable run through metalwork with no grommets? Only a thought...

 

[brman]

so let me get this straight.....

we have random voltages to metalwork and pipework throughtout the building. Bonding some bits doesn't affect other bits and it all depends on the load on multiple circuits?

Do I recall right this is a PME system?


Have we really ruled out the MET floating relative to the true mass of the eath?
Was it ever tested against a earth spike (not connected to anything else) with load on all the circuits? I got the impression it was but was confused what the results were........

 

[Lucien Nunes]

It was implied that the DNO had done such a test but I did re-ask the question and so far no explicit answers have been given. I was trying to rule it out but agree that we can't yet do so. However we do have numerical answers for insulation resistance that include some strange failures.

 

[brman]

It was implied that the DNO had done such a test but I did re-ask the question and so far no explicit answers have been given. I was trying to rule it out but agree that we can't yet do so. However we do have numerical answers for insulation resistance that include some strange failures.

yes, it was your post I was picking up on really.
As to IR readings, again - I agree it must be significant that there are so many low results. Even the cooker I would class as low for the type of circuit!

I don't suppose the house is on a flood plain is it?

 

[guest119]

I don't suppose the house is on a flood plain is it?

Will be this week, dunno about last week :smile5:

Boydy

 

[moospark]

I agree, it's also in full flow on the IET forum lol

It's only on waste flow here.

 

[Tridentsystems]

Guess that's the end of that. We will never know what problem was found

 

[ackbarthestar]

Guess that's the end of that. We will never know what problem was found

I think we're getting to the bottom of it.
The new build, as it was then, was ruffed in by a sparks, through the metal stud work, without protective grommets. The cables have now worn through and the phase is making fortuitous contact with the metalwork. Give it a few more months and we'll have a S/C.

Now how many circuits have been affected so far?
Upstairs lighting
RFC

Looks like a bit of a rewire.

 

[Rockingit]

I'm mightily cynical about this whole business: You've conducting water in a plastic pipe, attached to a plastic trap, attached to a plastic shower tray or porcelain bowl, sat on a wooden floor which is pulling enough current from some chaffed cable in a stud wall somewhere meters away that's sufficient to produce a tingle yet doesn't upset an RCD......

OP - I apologise, genuinely, for being this cynical, but it's just all a little too improbable! However, I do also hold a belief in the scientific approach which states that in any given problem, when ALL other possibilities have been removed, whatever remains - however implausible - has to be the truth.

 

[ackbarthestar]

However, I do also hold a belief in the scientific approach which states that in any given problem, when ALL other possibilities have been removed, whatever remains - however implausible - has to be the truth.]

Lol, cyncical? I should coco .........

Picard: Come back! Make a difference!
Kirk: I take it the odds are against us and the situation’s grim.
Picard: You could say that.
Kirk: If Spock were here, he’d say that I was an irrational, illlogical human being for going on a mission like this... Sounds like fun!

He is !

 

[telectrix]

it's life, jim. but not as we know it.

 

[chrish_1977]

Just going to be right out there and chuck an idea in, is it any water or is it hot water, maybe a cable in contact with a water pipe an when temp increases the insulation has broken down enough to give a small voltage, of course the longer this problem the higher the voltage (possibly) just thinking outside the box really as it's seems other than tearing the house apart

 

[Rampantchilli]

I had a very similar problem, which I did post up on here.
i had found some voltage inthe running water under the electric shower.
my testing concluded there was low IR on one lighting circuit, very long story short, I found the 28w 2d fitting was connected L-E, and although some supplementary bonding was present I doubled up in the old fashioned places, retested after, and it was gone, I suspect just doing the light repair would have solved it.
the IR readings were still fairly good, nearly 2M ohm, even with the light connected wrong