A real moan tonight...

[SUNRAY]

It is one of those 'several people have had a look so can you go in and sort it out' jobs, I popped in to aquaint myself with it with the expectation I'd get the call when required.

This is an outbuilding [happens to be an ex telephone exchange] currently used as changing room and pump house for a swimming pool and a koi pool. There is a foolscap log book for all of the history of the complex including all repairs etc, dosing, feeding, etc.

Problem: Intermittent RCD tripping, but when it happens it seems to occur for a number of days then fades away again.

A big old house in a country lane, overhead pole route feeding mostly farms.

The call came in tonight about 10pm.

The outbuilding installation consists of a 2p C50 MCB off the Henley block in the house, 80A RCD, 70m of 3C 16mm² SWA, outbuilding CU:
Main isolator,
2x C32 RFC total of 9DSSO, basically one RFC for swimming pool plant and one for outbuilding use
C32 for 10.5KW shower but one element disconnected making it ~6 to 7KW,
B6 for lighting ring,
B6 feeding 2 DSSO for koi plant customer is convinced this is the circuit causing the problem and ushered me directly to it:

Customer has generator and flood lamps so I was able to turn off main isolator to remove face plates.

Apologies for the poor quality pictures.

 

[SUNRAY]

So I removed the RCD socket [which was installed only a few months ago by a previous electrician to overcome the tripping issue] and found 2 red wires then removed the screws for the metal clad and nearly got thrown across the room as I pulled it off the back box

A bit later I rigged this to see if the volts between CPC and back box are real.

I've connected the L & N of the socket to CPC and earth terminal in back box to power a 60W bulb to double check this isn't a high impedance meter issue.
That voltage was changing as rapidly as a fiddlers elbow.

 

[SUNRAY]

The horizontal tube on the right feeds those sockets and the same voltage appears between the CU and the termination box as appears at the sockets.

I need to make a confession here, having tripped the RCD in the house twice I swapped over the 2p isolator in the CU and the RCD in the house for convenience and it could have been me [but I don't know] who got the black a brown feed round the wrong way.

Customer states the white plastic fitting between the box & CU was another of the things done to sort the problem.


So this is where I've got to so far:
The T&E feeding the 2 sockets links the CPC with the backbox & conduit effectively bridging the gap between the box & CU, the SWA is the DNO earth and the CU is a TT earth, being the original P.O. telephone exchange earth it is a very effective TT earth - thats the lighter G/Y wire at the bottom, the water feed into the building is galv and bonded using the dark green wire. I didn't have decent test gear with me but the cheap meter indicates 1.6Ω between the wires [1.1Ω with probes shorted together]. So there seems to be a low resistance path between water pipe and TT earth.

DNO arrived within 20 minutes, did some quick tests and went to the sub, shortly after they shut off the sub.



My moan takes the form of 2 questions:

Why do people insist on terminating the SWA into a silly add on box rather than the destination device?

and the second and more salient, When will people realise there is more than one way to terminate a SWA? They do not always have to have a brass gland and on this ocassion it created a dangerous situation.


My solution:

After removing the brass gland I put a few turns of a strand of 2.5mm² 7x round the armour to get it back snugg to the inner, covered with self amalgamating tape then heatshrink. that is inside the 32mm stuffing gland and coupler. I would have preferred to take it directly into the CU lower down but there was no slack anywhere. Even the original coupler between the box & CU had been left on top of the CU [as stated by customer] for me to refit.


I left at 01:30 but left the CU powered down as DNO had not restored power, going back later to finish off.

I have advised customer this building really needs upgrading [as have others before me], they seem keen to get to the bottom of the problem rather than just trusting it will dissappear with a rewire.

 

[bernardgreen]

Why do people insist on terminating the SWA into a silly add on box rather than the destination device?

Using an intermediate ( silly add on ) box to gland the SWA to with the cores and the inner sheath then going on into the consumer unit does make the process of glanding much easier. Fit the box to the cable, then fit the box to the wall and extend the cores and sheath into the consumer unit.

If as in this case the SWA "PME Earth" has to be isolated from the local TT "Earth" then cores and inner sheath run in plastic conduit from intermediate box to the CU.

When will people realise there is more than one way to terminate a SWA? They do not always have to have a brass gland and on this ocassion it created a dangerous situation.

I see the dangerous situation as being the result of a defective Neutral in the local network and/or a serious in-balance pf loads on the 3 phases in the local network giving rise to high currents in the Neutral.

Maybe install a lamp between the TT "Earth" and the PME "Earth" as a indicator of future problems with the Network Neutral.

 

[JohnW2]

Using an intermediate ( silly add on ) box to gland the SWA to with the cores and the inner sheath then going on into the consumer unit does make the process of glanding much easier. Fit the box to the cable, then fit the box to the wall and extend the cores and sheath into the consumer unit.

Indeed, but only if 'glanding' (in the SWA gland sense) is required ...

If as in this case the SWA "PME Earth" has to be isolated from the local TT "Earth" then cores and inner sheath run in plastic conduit from intermediate box to the CU.

As I think SUNRAY was implying, in this situation, the intermediate (or 'silly add-on') box is (unless space in the CU is very limited) unnecessary and, if metal, a potential hazard, since the last thing one wants in this situation is for the SWA armour to be 'glanded' to a touchable metal box - so, if one has a metal box and an 'SWA gland', one has to devise some method of using the gland which is different from how it is meant to be used.

Maybe install a lamp between the TT "Earth" and the PME "Earth" as a indicator of future problems with the Network Neutral.

I suppose that rather depends upon the nature of the 'lamp'. A neon (and very large resistor) would be safe, but may well give false indications, but I think I would (rightly or wrongly) be a buit nervous connecting the relatively low impedance of an incandescent bulb/lamp between a TN-C-S "Earth" and a TT one which was connected to exposed-c-ps.

Kind Regards, John

 

[winston1]

B6 for lighting ring,

Lights are not wired on rings. I’m surprised you don’t know that!

 

[JohnW2]

Lights are not wired on rings.

As you've often been told, that assertion is not inevitably true - and there is certainly no regulation that would prevent it.

 

[SUNRAY]

Lights are not wired on rings. I’m surprised you don’t know that!

Do you know anything at all Winston? It certainly appears to me that you don't.

I'd have though by now you would have learnt that I have a lot of experience and can identify and test most circuits with a reasonable degree of accuracy and

I’m surprised you don’t know that!

by now.

As I mentioned earlier there is an excellent log book for the building which the customer encourages the use of as a note book and there are several sets of dead tests showing ~0.35Ω of L & N end to end resistance and my first thought at 11pm on a callout in the cold and damp was Winstons repeated and incorrect bleating so I just had to do this:

So once and for all will you PLEASE keep your factually incorrect and rude/arrogant statements to yourself.

I find it so sad that while dealing with potentially life threatening situations one of the first thing that goes through my head is the comment from a silly little know nothing.

 

[SUNRAY]

Using an intermediate ( silly add on ) box to gland the SWA to with the cores and the inner sheath then going on into the consumer unit does make the process of glanding much easier. Fit the box to the cable, then fit the box to the wall and extend the cores and sheath into the consumer unit.

I come across these things fairly regularly and on most ocassions I simply treat them as the botch that they are [in my opinion]. This particular installation apparently had a smaller feed going directly into the bottom of the CU. The box was fitted when the new harmonised cable was installed, to allow for a shower to be installed, and as mentioned originally using a galve coupling, the additional steel tube was added for the 2 DSSO's [despite being red/black T&E]

I have never even had a thought about installing an external box and certainly do not see the need for the additional work they generate unless the access doesn't allow direct entry as in the case of the Besa box top left of CU. Glanding into a trunking is a different issue.

If as in this case the SWA "PME Earth" has to be isolated from the local TT "Earth" then cores and inner sheath run in plastic conduit from intermediate box to the CU.

Isn't this always the case with a TT earth? The white plastic conduit was a more recent alteration in a failed attempt to make that isolation.

I see the dangerous situation as being the result of a defective Neutral in the local network and/or a serious in-balance pf loads on the 3 phases in the local network giving rise to high currents in the Neutral.

Yes it is something of that ilk, power was restored to the road around 4am, I may know more shortly when I visit to reinstate the outbuilding and look for the RCD tripping problem.

Maybe install a lamp between the TT "Earth" and the PME "Earth" as a indicator of future problems with the Network Neutral.

Why? For what purpose?
Maybe install some sort of voltage controlled isolating units... oh no we have done away with those.


Surely all that was needed was some commonsense and not use the unneccessary and in this case [like so many] totally inappropriate brass gland. As can be seen in the pics a stuffing gland is far more appropriate in this case and would have been so much better for cable access and far neater 6" lower on the side of the CU.

 

[JohnW2]

Why? For what purpose? ... Maybe install some sort of voltage controlled isolating units... oh no we have done away with those.

In the context of bernards oft-voiced concerns about supply side neutral faults on TN-C-S supplies, it has sometimes been suggested that (given some 'true earth' reference, like a 'TT' electrode) it would be easy enough to engineer a relay/contactor which would disconnect the supply in the event of a high pd arising between the DNO's 'earth' (i.e. neutral) and true earth - and I think I've even seen/heard suggestions that some such arrangement should perhaps be required with TN-C-S supplies.

However, a DP relay/contactor (switching just L and N within the installation) would not achieve what was needed - the PEN (i.e. before the installation's 'earth branched off from the neutral) would have to be switched - and that is certainly not allowed within an installation by current regs!

Surely all that was needed was some commonsense and not use the unneccessary and in this case [like so many] totally inappropriate brass gland. As can be seen in the pics a stuffing gland is far more appropriate in this case and would have been so much better for cable access and far neater 6" lower on the side of the CU.

Very much so, in this case. Even if the 'add-on box' were not metal, to use a metal gland for the SWA would be introducing unnecessarily (potentially 'safety') issues, and that risk becomes magnified if one also has a metal box.

Kind Regards, John

 

[bernardgreen]

Isn't this always the case with a TT earth?

This confuses me ......

At the base of a pole there is a ground electrode that is connected to the Neutral at the top of the pole to be one of the Protective MULTIPLE Earths of the network. Is this any different to a connection of the Neutral to a ground electrode in the customers premises ?

The only difference I can think of is that a fire can be caused when a failed Neutral causes a high current to flow along the cable from Neutral to ground electrode and the cable overheats. Better that this fire is outside on a pole and not inside a home.

It was initially said that this fire was due to a fault in the telephone junction box, ? ? or was it a overloaded / defective Neutral to Ground electrode bonding cable. OpenReach and Western Power at the time denied responsibility. The burnt pole was removed before it could be properly looked at by independent eyes to see if there had been a bond cable.

In the context of bernards oft-voiced concerns about supply side neutral faults on TN-C-S supplies,

Having "survived" a few Neutral bounces ( over head open wire supply along the street ) I fear they are not in-frequent events.

 

[JohnW2]

This confuses me ......

Yes, I'm also a bit uncertain as to what SUNRAY meant. As you say, the 'M' of the 'PME' (which is required with TN-C-S) means that there are 'multiple' connections between the supply neutral and earth electrodes (even though, if what westie used to tell us is correct, that 'multiple' often/usually meant only one 'earth' in addition to the one at the transformer).

Furthermore, it was only at the 11th hour that the requirement for all TN-C-S installations to have a local TT rod (connected to the TN-C-S 'earth') was removed from the draft of the current version of BS7671 - so I presume that this was not considered to be a 'hazard' (which sounds reasonable, since hardly any credible domestic TT electrodes would have a low enough impedance to result in enough current to flow to damage, let alone 'set alight' any credible cable).

In practice, main bonding cables from TN-C-S 'earths' to low-impedance extraneous-c-ps, are far more likely to carry very high currents than is a connection to a domestic TT electrode.

Kind Regards, John

 

[JohnW2]

Having "survived" a few Neutral bounces ( over head open wire supply along the street ) I fear they are not in-frequent events.

I would certainly think that it's most likely to happen with overhead supplies using singles, particularly given that such supplies usually have the neutral conductor below the L one(s), hence more likely to be damaged - I'm actually a little surprised that TN-C-S is allowed with such a supply (which is becomingly increasingly uncommon).

However, with overhead ABC, or underground, cables I would imagine that selective loss of the neutral conductor is extremely rare.

Kind Regards, John

 

[bernardgreen]

Investigations revealed the blast was caused after a piece of cable was cut from an overhead line. This affected the earthing of the electrical network in the area and resulted in some unusual electrical activity, which affected some of the gas pipes within the properties. This resulted in the fires and explosion.

Yorkshire Post Saturday, 11th February 2012,

Initial fires were caused by overloaded Main Bonds to metal water mains,

 

[SUNRAY]

In the context of bernards oft-voiced concerns about supply side neutral faults on TN-C-S supplies, it has sometimes been suggested that (given some 'true earth' reference, like a 'TT' electrode) it would be easy enough to engineer a relay/contactor which would disconnect the supply in the event of a high pd arising between the DNO's 'earth' (i.e. neutral) and true earth - and I think I've even seen/heard suggestions that some such arrangement should perhaps be required with TN-C-S supplies.

However, a DP relay/contactor (switching just L and N within the installation) would not achieve what was needed - the PEN (i.e. before the installation's 'earth branched off from the neutral) would have to be switched - and that is certainly not allowed within an installation by current regs!
Very much so, in this case. Even if the 'add-on box' were not metal, to use a metal gland for the SWA would be introducing unnecessarily (potentially 'safety') issues, and that risk becomes magnified if one also has a metal box.

Kind Regards, John

Quite right, it may have been noticed all the breakers in the CU are off and aqequate testing for dead was performed.
Not importing the DNO's earth into the building would have removed so much danger, as would not exposing within the building. This would effectively negate any percieved requirement to use a device to disconnect anything.
Excuse the patchwork but a basic version of a nicer way to put this SWA in:

Woud have been just as easy to the right of the pic but 10% of the difficulty at the CU. Unthreading and rethreading those 16mm² wires innthat quite restricted gap by the earth bar was a PITA.