Termination of SWA on distant end of TT supply

[shedbuilder]

I have a job to change out a consumer unit in a small workshop that is currently fed by a three core 10mm SWA cable. one conductor is earthed, as is the armouring at the supply end CU. The circuit is protected by a 40amp S type 100mA RCD and 32amp MCB on a PME supply.
At the workshop end the installation has been made into a TT system. The earth from the rod has been taken directly into the consumer unit and from there to the water pipes etc. The supply is protected by a 30mA RCD and appropriate MCBs for the final circuits. The SWA has been terminated on a gland that has been fitted into a metal box. The live and neutral pass straight through the box to the CU and the spare earth conductor has been insulated with heat shrink sleeving. The metal box earth potential is naturally at the potential of the supply end rather than the local earth. Should this box be changed to plastic and should the SWA gland also be enclosed so that no one can touch both earths at the same time? Indeed could the gland be removed and aplastic one used with the armouring tidied up with heat shrink sleeving.

 

[JohnW2]

... a PME supply. .... At the workshop end the installation has been made into a TT system. .... The metal box earth potential is naturally at the potential of the supply end rather than the local earth. Should this box be changed to plastic and should the SWA gland also be enclosed so that no one can touch both earths at the same time? Indeed could the gland be removed and aplastic one used with the armouring tidied up with heat shrink sleeving.

I think the short answer has to be YES. To have the PME earth accessible in an environment earthed to the local TT electrode is potentially dangerous. By whatever means, you should make it impossible to touch the SWA armour or anything in continuity with the SWA armour (and, of course, not connect the earth core of the SWA to anything - as is the case).

Kind Regards, John.

 

[JohnW2]

...I should perhaps have added that an alternative strategy would probably be to bond the SWA earth to the shed's TT earth. However, the earth rod would then become an extraneous-conductive part as far as the PME installation was concerned, so the SWA's effective CSA (armour + core) (and the CSA of the CPC of any cable supplying the SWA, if it didn't go all the way to the CU) would have to be adequate as a PME Main Protective Bonding conductor.

Kind Regards, John.

 

[Taylortwocities]

And (IIRC) your earth electrode has a very interesting role should the PME's neutral become disonnected.

 

[JohnW2]

And (IIRC) your earth electrode has a very interesting role should the PME's neutral become disonnected.

Exactly. That's the reason for what I've said. In the event of loss of the supply neutral, the potential of the installation's 'earth' can (depending on where the neutral break occurs) rise to close to the phase/line voltage - which is obvioulsy very dangerous if one can touch that and also a true earth (like that provided by the earth rod). One therefore either has to make sure that the PME earth is in no way accessible in the shed, or else bond it to the rod so as to create an equipotential zone within the shed (with caveats above regarding SWA sizing etc.)

Kind Regards, John.

 

[riveralt]

I found this a while ago but the info is still relevant

http://www.theiet.org/publishing/wiring-regulations/mag/2005/

Issue 16 Autumn 2005 article on earthing a shed

 

[sparkticus]

Is it not permisable to remove the TT (take out the earth rod) then bond any garage extraneous conductive parts back to the main dwelling MET using a minimum of 10mm earth cable. Remove the heat shrink covering the CPC and connect it as normal. Essentially making the garage part of the main dwelling PME system (exporting the PME)?

 

[RF Lighting]

Yes you could keep the garage as a PME supply providing there is at least a 10.0mm² earth and it is of low enough impedence to be used as a main protective bonding conductor, and bond all extraneous-conductive-parts within the garage. This would be my personal preferred method.

If the supply remains as a PME supply with a local TT conversion, then the TD RCD needs removing from the supply end of the cable, the SWA needs to be terminated into something insulated such as a nylon stuffin gland, or a gewiss box etc, and then carry out the installation from here as a normal TT supply with all the normal extra measures that this supply type requires.

 

[sparkticus]

This would be my personal preferred method.

My preferred method also. You know where you are with that.

Delayed RCD, two earth potentials, CPCs disconnected - very messy and easy for a DIYer to mess it up in the future and kill themselves!

 

[JohnW2]

Is it not permisable to remove the TT (take out the earth rod) then bond any garage extraneous conductive parts back to the main dwelling MET using a minimum of 10mm earth cable. Remove the heat shrink covering the CPC and connect it as normal. Essentially making the garage part of the main dwelling PME system (exporting the PME)?

Two potential problems:

• 1...the supply comes via (I assume buried) SWA, so the OP would probably not want to have to run a separate G/Y cable. All therefore depends upon whether the copper-equivalent of the SWA (armour +CPC) is adequate (which it may well be, with 10mm - haven't looked it up).
  
  2...bonding 'traditional' extraneous-conductive-parts (incoming pipes, structural metalwork etc.) in the garage is straighforward enough, but if it has potentially conductive (damp and in contact with earth) walls and/or floor, there could be an essentially unbondable potential hazard.

Probably better - as I said in an earlier message, one can probably do as you suggest except for leaving the local earth rod in place, and simply treating it as an extraneous-conductive-part as far as the installation is concerened. That local connection to 'local earth' ought to help to reduce the walls/floor problem mentioned above, although it won't completely eliminate that problem.

That's how I see it, anyway.

Kind Regards, John

 

[RF Lighting]

The OP mentions a 3 core 10.0mm² in their post.

Providing the run isn't too long, the 3rd core combined with the SWA will have sufficient CSA and probably low enough impedence to be used.

You should really remove the rod as although unlikely, there can be big problems in the open circuit CNE conductor scenario.

 

[sparkticus]

the supply comes via (I assume buried) SWA, so the OP would probably not want to have to run a separate G/Y cable. All therefore depends upon whether the copper-equivalent of the SWA (armour +CPC) is adequate (which it may well be, with 10mm - haven't looked it up).

Yes, I'm not sure either about the copper equivalent for 10mm SWA but (and I have not looked up the regs so may be wrong) is the 10mm (minimum CSA) not mandatory as a discrete protective conductor (just like within the main dwelling) since the workshop now becomes an extension of the main dwelling from the perspective of exporting PME?

 

[RF Lighting]

Yes you have to use a minimum of 10.0mm² as bonding conductors may be subjected to network fault currents if things go very wrong.

 

[JohnW2]

The OP mentions a 3 core 10.0mm² in their post. Providing the run isn't too long, the 3rd core combined with the SWA will have sufficient CSA and probably low enough impedence to be used.

Indeed, as I said the 10mm² SWSA's core+armour may well be adequate.

You should really remove the rod as although unlikely, there can be big problems in the open circuit CNE conductor scenario.

It's not the first time I've seen this suggested, but what 'big problems' are you thinking of? As far as I can see, in the event of the infamous 'open CNA conductor scenario', an additional (there will usually be plenty of others via the house's main bonding) local connection to true earth can but improve the situation slightly, particularly in the garage. What am I missing? People sometimes talk about 'massive currents' flowing into the rod but the reality is that a riode Ze <50 is pretty unlikely, so we're probably talking about 5A tops.

Kind Regards, John.

 

[westie101]

Yes the 10mm2 is the minimum size, I seem to recall we discussed this recently and 6mm2 with a core and the SWA in parallel is greater than 10mm2 copper equivalent, so 10mm2 will be more than adequate