Had a call from a customer today to say his wife had a shock from the outside tap. The grass was wet and she was in bare feet.

Wandered over there to find he had turned off the consumer unit "just in case" (obviously he didn't have her well insured)

17th edition board, tncs supply.


Anyway, turned the mainswitch on and checked the tap, just over 20v to earth.

Done the usual tests, IR ok, then earth loop at the mainswitch to find 0.80 ohms which was a bit higher than I would have expected knowing where the substation is in relation to the property and how it's fed.

Anyway, turned the mainswitch on and checked the tap, just over 20v to earth.

Reported it to UKPN at 16:00, Had an engineer on site by 17:30 who confirmed my findings and was turning out the on call linesman as I left site.

Nice when it works properly.

Stu

---what can you say?
an iconic DNO----UKPN!

Regards.

--by the way, thanks for your one post demolition of the "its ok to give
PME to outside situations" brigade"

outside taps should of course have an insulated insert.

Regards.

It was one of those almost classic "I know what that's going to be calls"

We didn't rewirel the farmhouse as it was done by the "builders electrician" who happened to be one of the big players in Chelmsford.
I knew it was TNCS as one of my guys stuck a couple of extra sockets in a few months ago.

As soon as the farmer told me the symptoms I was already mentally sticking a screwdriver in the grass to measure before I'd even left the previous job.

Don't know the full outcome but I suspect tomorrow will bring the story as it was 50/50 for a term on the eaves of the house or a joint somewhere under the farmyard.

Stu

But of course if the tap was insulated, the DNO wouldn't have known they had a problem


Stu

--its not necessarily a problem, high loops are common in rural areas,
we get the queries of course, but 0.80 is not uncommon. there is no guarantee of the usual figures.

Regards.

Where should we draw the line? Class 1 light fittings on the outside of a house? Indoor sockets that could foreseeably be used for equipment outdoors?


Agreed that 0.35 isn't guaranteed, but wouldn't 0.8 be a problem for any supply rated at over 46A? - as it's TN-C-S the 'earth loop' is also the L-N loop - and if the lines had such as high resistance the voltage drop with a decent load would be outside permitted (statutory) limits.
- Andy.

It was a 23KVA service and given that it's about 200 yards from the sub I think 0.8 was a tad on the high side even in a rural area.

Andy, It would have probably gone un-noticed for years other than the customer was playing with her kids in the garden and everywhere around the tap was wet (including her shoes) which probably made the 20v noticeable.

When I stuck my carefully calibrated finger on it (having metered it first) I couldn't feel a thing, but then I did have big rubber soled boots on.

Stu

A fault in the CU would have been interesting - about 4.5 minutes clearance time on a 100A BS1361!

Regards,

Alan.

Your heading for this thread is TN-C-S problem, "what problem", cannot be 20V to ground earth on the TN-C-S connection with a reading of 0.8 ohms Ze. Why go to the throat of the DNO right away, most cases problem like that is internal with no cpc or poor earthing to conductive parts. Due to the lack of written down marshalling of evidence, which may have lead to your conclusion, it is difficult for the reader to follow. Have to go back and read it again to see if missed any point. Remember: First tell your readers what you are going to tell them. Tell it them. And then tell them what you have told them.

I'm going to tell you that because the slant of this thread is toward toward the supply, and not to the usually assumed internal cpc problems I find it very informative.

Because the slant of this thread is toward toward the supply, and not to the usually assumed internal cpc problems I find it very informative.

I just told you that because the slant of this thread is toward toward the supply, and not to the usually assumed internal cpc problems that I find it very informative.

Let us know the outcome Stu.

Zs

Surely no one, who had tested a higher than expected reading of Ze on TNCS, would assume that the problem was with an internal cpc ?

-------------------------
Empty barrels make the most noise.

No Ricicle, but the 'my wife got an electric shock' scenario usually recieves the CPC/look for a dodgy connection/check bonding advice for a few posts first, then it will ramble on to something else. Stu went straight for the jugular, called UKPN and went home. Nice. I like Andy's calculation too, working it backwards while I wait for the other computer here to react between presses.

Zs

Well when I spoke to the farmer this morning to check what happened overnight. The linesman duly turned up and checked the terminations on the eaves, all ok and still 0.8 towards the supply.

This morning, a team returned and dug a sizeable hole in the footpath to investigate the underground service. I popped in about lunchtime to find a jointer "in the hole" opening up the main and after testing muttered several four letter words in a highly technical manner.. The fault was still towards the sub.

I left them wandering down the street debating where to dig the next hole.



I didn't add last night that my 20v to earth was with a probe stuck half inch into the grass and not to a proper "screwdriver in the garden" test.

Stu

So I might be missing something here but why can't I have 20v to earth on a TNCS connection with a loop of 0.8 ohms ?. The problem made apparent due to good bonding to the water service, the house has wooden floors and the equipotential zone was working a treat indoors.


Stu

Why not?

If we guess that the line conductor has an impedance of let's say 0.1 Ohms, then 0.7 Ohms is due to the CNE. To generate 20V across that we'd only need a N current of about 28.6 A (possibly totalled across several properties) - which doesn't seem entirely unreasonable. Any multiple electrodes are likely to be several Ohms, if not tens of Ohms, to earth so won't make a significant difference to the numbers.

- Andy.

Well that's what I thought Andy, but unless I read JCM's post wrong (and it's possible as it's been a long week), I started to think I was going mad.

We've got a farm, some street lighting and I believe 5 properties (including the one I'm interested in) sitting on the same main so at 4 pm in the afternoon I'd say there's a good chance your maths is spot on, bearing in mind the kettle went on as soon as I turned the mainswitch on (for tea)


Stu

Ok, I am aware that PME is not suitable for use in potentially explosive atmospheres because connection of the neutral to true earth is at the transformer (and should be at intervals along its length). Thus, any neutral current will introduce a volt drop in this conductor, which could raise the voltage at the MET above true earth. However, 20v is high. The last time I have seen and measured about 30 volts between neutral and earth was because it was floating (3P), (the neutral earth link left out in a pillar box) . The MET is not normally connected to true earth at the consumer's installation in a TN-C-S system. (Maybe it was with the water pipe). Surely if one house had 20volts neutral to earth all the houses would be the same. If your sparks were in that house wiring additional sockets, why did they not pick up this fault, how did the added RCD circuit protection react to this voltage? . No doubt, your good self will keep us informed of the outcome, the cause and solution to solve the problem.

jcm,

My sparks was there about 6 months ago and having looked back at the cert everything was in order then.

The RCD protection didn't appear to care and to be honest I wouldn't really expect it to as what's coming in is going out with the exception of the few milliamps (possibly microamps) following through the farmers wife,

There's a good chance the other houses are the same but given that I never felt a thing when I touched the tap I wonder if anyone else would have.

The incoming water main is plastic, so the only connection to true earth was through the customer's wife when she touched the tap.


The jointers were fairly confident there was a fault in a joint somewhere between the sub and the hole we were looking in at lunchtime and I left them to decide where to dig next (not really my problem). One possible solution they came up (with was subject to approval from an engineer) was to add several electrodes to hold the neutral back down but as they said it wasn't really a cure just a way of hiding the symptom

The other option of course is for us to TT it on our side, but then there could still be issues on other properties in the road so doesn't really provide a fix.

Stu

Agreed - but in this case things are abnormal - seemingly due to a problem with the supply CNE having a much higher resistance than it should (e.g. loose/corroded connection somewhere).


Probably didn't care one bit. Simple RCDs won't even able able to notice (as they only see L-N), ones with a cream functional-earth connection might be able to notice, but I doubt they'd object as N can differ from PE in TT systems quite considerably (N might be higher due to v.d., PE might be higher (up to 50V) due to leakage) and the same models of RCD are used there without problems.

- Andy.