Separating PME & TT earthing for outbuildings

[Msitekkie]

I would appreciate peoples thoughts on fixing this earthing setup to correctly separate PME & TT:

Setup: Workshop at the bottom of a long garden connected to supply via 3 core armoured cable. Earth rod fitted at workshop, but the earth has also been connected to the PME from the supply & the armouring is earthed throughout.
Just to complicate matters further, a No2 Pratley box (Metal) has been used to T off the armoured cable in a garage located between supply & workshop.

More detail: Both outbuildings have individual consumer units & have concrete floors & block walls. Floor in the garage tends to be damp. SWA is partly buried, partly clipped & connected to a consumer unit in the meter box with 40A RCD (not S type unfortunately, but that doesn't affect the earthing & DNO seem happy with CU in the meter box).

Questions/thoughts:
I am aware TT & PME should not be joined, so would begin by disconnecting the SWA earth core at the supply end. Probably also in the Pratley box on the supply side. Leaving the TT earthing connected between workshop & garage.
I am not so sure about the armouring... If it was just the workshop I would just isolate the armouring from the consumer unit there. Would you just do the same with the garage consumer unit and leave it at that?

Additional thoughts:
Would you expect the garage to need it's own earth rod?
Pratley have told me they can supply "non metallic glands" for their box so maybe for the garage consumer unit I could isolate the armouring where it comes out of the Pratley box rather than at the consumer unit end of that short leg (on the wall). Otherwise at the CU end like the workshop.

Summary: So hopefully I would end up with the circuits in both outbuildings TT earthed, but the armouring PME earthed (& isolated from TT) except the short leg on the wall from Pratley box to CU which would be TT earthed.

 

[marcuswareham]

In theory both buildings could use the PME earth instead of TT, if you remove the rod at the workshop then there is no extraneous conductive parts at either building so no need for 10mm bonding conductor back to the house.

Although a lot of people would favour the option I just mentioned, I myself due to the fact of the wet concrete floor having good conductivity to true earth I would ditch the PME earth and make both outbuilding TT as you are suggesting,

What I have done in my own outbuildings is use a plastic 2 module enclosure and terminated the SWA onto the bottom of this enclosure using an insulated gland so this isolates the PME earth connected to the amour, within this enclosure I have a 100mA delay RCD for fault protection and then T&E coming out of the top of the enclosure with the CPC connected to the TT earth going to a small garage-style consumer unit which contains a 30mA RCD as the main switch for additional protection.

I opted for doing it this way rather than using a plastic wiska box for isolation because: it gives me a plastic enclosure to fit an RCD for fault protection and means that the supply cables going into the metal garage consumer unit are RCD protect, if I used a wiska box for isolation in the unlikely event that one of the supply cables feeding the garage consumer unit shorted out to the metal case of the consumer unit, due to the high resistance of the earth electrode the MCB feed the SWA would not disconnect, make sense?

 

[Rockingit]

In theory both buildings could use the PME earth instead of TT, if you remove the rod at the workshop then there is no extraneous conductive parts at either building so no need for 10mm bonding conductor back to the house.

Be interesting to actually do this by calculation rather than assume the tabular values?

 

[marcuswareham]

Be interesting to actually do this by calculation rather than assume the tabular values?

There could be other extraneous conductive parts we don't know about, do you mean do the calculation for the earth rod and cable coming into the building or for the conductivity of the floor

A cable coming to a building connected to an earth rod in the ground surely would obviously bring an earth potential into the building

 

[telectrix]

I'd use the PME and dry th floor. ever heard of Trench Foot. That's worse than the occasional electric shock.

 

[Msitekkie]

I'd use the PME and dry th floor. ever heard of Trench Foot. That's worse than the occasional electric shock.

I have tried hard to get the damp problem fixed without success unfortunately, although with this type of outbuilding where you are in contact with the ground, my understanding is TT is the correct choice anyway even without the damp because of the risks from a broken PEN conductor. Different scenario if you have a wooden building. There are some metal accessories too.

 

[Msitekkie]

Be interesting to actually do this by calculation rather than assume the tabular values?

? Sorry, I am not sure what calculation you have in mind? I have a mix of TT & PME which needs to be separated, are you suggesting there is some calculation that could in theory suggest having both is OK?

 

[Rockingit]

? Sorry, I am not sure what calculation you have in mind? I have a mix of TT & PME which needs to be separated, are you suggesting there is some calculation that could in theory suggest having both is OK?

It's (theoretically) possible that by applying the calculations for minimum supplementary bonding csa that if it comes back in small enough (min 4mm) that your SWA armoured might be up to the job, in which case just extend the PME. Have a read of GN8.

 

[marcuswareham]

I have tried hard to get the damp problem fixed without success unfortunately, although with this type of outbuilding where you are in contact with the ground, my understanding is TT is the correct choice anyway even without the damp because of the risks from a broken PEN conductor. Different scenario if you have a wooden building. There are some metal accessories too.

It is up to the designer to assess the risks associated with a broken PEN conductor, and then put mitigation in place to reduce the risk, if the risks are unacceptable.

One way of mitigating against a broken PEN conductor risk is by making the outbuilding TT, although this is not as simple as a fix-all solution sometimes using TT can increase the risk and careful consideration should be taken into account when setting up TT systems very close to TN systems

 

[Msitekkie]

It's (theoretically) possible that by applying the calculations for minimum supplementary bonding csa that if it comes back in small enough (min 4mm) that your SWA armoured might be up to the job, in which case just extend the PME. Have a read of GN8.

There is no supplementary bonding requirement. This is mainly about mitigating the risks of a broken PEN conductor.

 

[marcuswareham]

It's (theoretically) possible that by applying the calculations for minimum supplementary bonding csa that if it comes back in small enough (min 4mm) that your SWA armoured might be up to the job, in which case just extend the PME. Have a read of GN8.

The minimum size for a bonding conductors with a PME supply is 10mm

see Reg 544.1.1 & table 54.8

I believe this is todo with high current under PEN fault conditions and the potential for diverted neutral currents

 

[Msitekkie]

It is up to the designer to assess the risks associated with a broken PEN conductor, and then put mitigation in place to reduce the risk, if the risks are unacceptable.

One way of mitigating against a broken PEN conductor risk is by making the outbuilding TT, although this is not as simple as a fix-all solution sometimes using TT can increase the risk and careful consideration should be taken into account when setting up TT systems very close to TN systems

Yes this is the scenario I am working with.

 

[marcuswareham]

There is no supplementary bonding requirement. This is mainly about mitigating the risks of a broken PEN conductor.

Although you say currently you have an earth rod connected to the PME earth at the workshop, this earth rod could be counted as an extraneous conductive part, if you were to leave it this way with that earth there would need to be a minimum of a 10mm conductor connecting that earth rod to the PME MET

This is also the case for any other extraneous conductive part in either of the buildings, like water pipes etc

 

[Msitekkie]

The minimum size for a bonding conductors with a PME supply is 10mm

see Reg 544.1.1 & table 54.8

I believe this is todo with high current under PEN fault conditions and the potential for diverted neutral currents

There is no bonding requirement. No water pipes or anything like that in the outbuildings.

 

[Rockingit]

The minimum size for a bonding conductors with a PME supply is 10mm

see Reg 544.1.1 & table 54.8

I believe this is todo with high current under PEN fault conditions and the potential for diverted neutral currents

But are we discussing main bonding or supplementary??

 

[Rockingit]

There is no bonding requirement. No water pipes or anything like that in the outbuildings.

But there ARE extraneous conductive parts???

 

[Msitekkie]

Although you say currently you have an earth rod connected to the PME earth at the workshop, this earth rod could be counted as an extraneous conductive part, if you were to leave it this way with that earth there would need to be a minimum of a 10mm conductor connecting that earth rod to the PME MET

This is also the case for any other extraneous conductive part in either of the buildings, like water pipes etc

It's (theoretically) possible that by applying the calculations for minimum supplementary bonding csa that if it comes back in small enough (min 4mm) that your SWA armoured might be up to the job, in which case just extend the PME. Have a read of GN8.

Sorry I don't understand how that would protect against a broken PEN conductor?

 

[marcuswareham]

There is no bonding requirement. No water pipes or anything like that in the outbuildings.

That make TT easier then, H

But are we discussing main bonding or supplementary??

Wouldn't it be main bonding

 

[Msitekkie]

But there ARE extraneous conductive parts???

Exposed conductive parts yes, extraneous conductive parts no, unless you include the earth rod itself in that category.

 

[marcuswareham]

Exposed conductive parts yes, extraneous conductive parts no, unless you include the earth rod itself in that category.

In its current configuration, the earth rod is classed as an extraneous conductive part as currently the PME earth is in use at the workshop.

However you want to isolate the PME earth and change to TT, in that case the earth rod will become the earth electrode for the workshop TT system.

 

[Msitekkie]

In its current configuration, the earth rod is classed as an extraneous conductive part as currently the PME earth is in use at the workshop.

However you want to isolate the PME earth and change to TT, in that case the earth rod will become the earth electrode for the workshop TT system.

So coming back to my original posting, does anyone have any comments about the specifics of achieving this seperation, or does my proposed approach seem correct? Interestingly I think I have found some suitable "Tufflon" compression glands from Pratley to achieve the isolation rather than having to use a plastic box to terminate the SWA to.

 

[timhoward]

I'm very tired and obviously missing something obvious here, and fully prepared to look a right idiot!

As a matter of semantics wouldn't you normally apply the term PME to the DNO side of the supply? So essentially the OP has TNCS?

I believe the OP is saying that the TNCS is exported to a non-metalic building, and the only 'extraneous conductive part' is a backup earth rod, presumably installed as a belt and braces backup in case of broken upstream PEN conductor. I'm assuming this is run in insulated earth cable until the rod itself, so the only exposed point is the very top of the rod?

Extraneous conductive parts obviously need bonding to a fixed potential, and the earth rod is connected to real mass of earth one end and the exported earth the other end.
Supposing the earth rod was actually a metal water pipe that ran out underground and up to a tap. You'd bond it.
I'm not grasping the difference, or the actual issue in having a backup earth rod in an outbuilding which has an exported TNCS earthing conductor. What am I missing here please?

 

[Msitekkie]

I'm very tired and obviously missing something obvious here, and fully prepared to look a right idiot!

As a matter of semantics wouldn't you normally apply the term PME to the DNO side of the supply? So essentially the OP has TNCS?

I believe the OP is saying that the TNCS is exported to a non-metalic building, and the only 'extraneous conductive part' is a backup earth rod, presumably installed as a belt and braces backup in case of broken upstream PEN conductor. I'm assuming this is run in insulated earth cable until the rod itself, so the only exposed point is the very top of the rod?

Extraneous conductive parts obviously need bonding to a fixed potential, and the earth rod is connected to real mass of earth one end and the exported earth the other end.
Supposing the earth rod was actually a metal water pipe that ran out underground and up to a tap. You'd bond it.
I'm not grasping the difference, or the actual issue in having a backup earth rod in an outbuilding which has an exported TNCS earthing conductor. What am I missing here please?

Hi Tim, maybe looking back at my original post would help - we got a bit off the point with discussing the earth rod itself.
When deciding to export the PME or not, the available guidance tends to talk about extraneous conductive parts such as water pipes & the associated bonding issues so people focus on that, hence I think some of the bonding discussion above.
The other issue that is often ignored (and which I am talking about here) is when you get a broken Neutral (PEN conductor) in the supply. In this scenario the earth becomes live along with any exposed conductive parts. Yikes! In the house this is less of an issue because of the reduced likelyhood of being in contact with earth (carpets wooden floors etc). A wooden outbuilding with plastic accessories is similar to the house.
To mitigate against the risk of electric shock where you have exposed conductive parts in an outbuilding (sometimes walls made of metal sheeting even) you use TT earthing in the outbuilding & isolate it from the PME. In this case a previous electrician has left the PME & TT interconnected and the scenario is further complicated by the SWA passing through and being T'd off in the 2nd outbuilding. It is also necessary to consider protection should someone say, put a pick-axe through the cable.
So the question is mostly about achieving this separation in the correct manner.

 

[marcuswareham]

So coming back to my original posting, does anyone have any comments about the specifics of achieving this seperation, or does my proposed approach seem correct? Interestingly I think I have found some suitable "Tufflon" compression glands from Pratley to achieve the isolation rather than having to use a plastic box to terminate the SWA to.

I believe others have used a standard plastic stuffing gland to isolate the amour from SWA, I guess as long as it grips & supports the cable and you cannot access exposed conductive parts from each earthing system at the same time (411.3.1.2) then all is ok

 

[Msitekkie]

I believe others have used a standard plastic stuffing gland to isolate the amour from SWA, I guess as long as it grips & supports the cable and you cannot access exposed conductive parts from each earthing system at the same time (411.3.1.2) then all is ok

Thanks Marcus, I have seen suggestions before of using a step-down adapter from 25mm to 20mm, but not that you could use an ordinary stuffing/compression gland to isolate the armouring when terminating to metal containment. Guess I should give that a try then, but would be interested to hear from someone who has actually used that approach.

 

[Soi disant]

I am aware TT & PME should not be joined,

Oh.

No more bonding of incoming metal service pipes then.

 

[timhoward]

Hi Tim, maybe looking back at my original post would help - we got a bit off the point with discussing the earth rod itself.
When deciding to export the PME or not, the available guidance tends to talk about extraneous conductive parts such as water pipes & the associated bonding issues so people focus on that, hence I think some of the bonding discussion above.
The other issue that is often ignored (and which I am talking about here) is when you get a broken Neutral (PEN conductor)

Thanks for responding. And sorry I misunderstood the question.
I understand the dangers of a broken PEN conductor but the current arrangement would seem to already make it safer, subject of course to the resistance of the rod.
I knew I'd seen this discussed in a video - I finally found the right one - 20 minutes 10 seconds in:

That example is at supply source, but the same would presumably apply in at outbuilding.

Converting the outbuilding to TT is of course still an option and you might prefer it.

 

[Msitekkie]

I believe others have used a standard plastic stuffing gland to isolate the amour from SWA, I guess as long as it grips & supports the cable and you cannot access exposed conductive parts from each earthing system at the same time (411.3.1.2) then all is ok

Thanks Marcus. looks like my earlier reply to your message didn't survive, so let me try again. I was saying I have heard of people using 25mm to 20mm step down adapters to isolate, but hadn't come across this approach before & should give it a try.
I also mentioned I would be interested to hear from anyone who has tried this approach.

 

[Msitekkie]

Thanks for responding. And sorry I misunderstood the question.
I understand the dangers of a broken PEN conductor but the current arrangement would seem to already make it safer, subject of course to the resistance of the rod.
I knew I'd seen this discussed in a video - I finally found the right one - 20 minutes 10 seconds in:

That example is at supply source, but the same would presumably apply in at outbuilding.

Converting the outbuilding to TT is of course still an option and you might prefer it.

In John's example there are no exposed conductive parts, so it's a different scenario & little risk involved - the foundation earthing mats he mentioned briefly, I understand may become a thing in the next version of the regs - necessary I gather for "Islanding" purposes if you want to take a solar installation off-grid when you have a power cut. I commented on that video asking John to demonstrate the SWA isolation in another viseo, as I haven't seen any of the usual YouTuber suspects do it, but JW hasn't responded (as yet anyway).

 

[timhoward]

I saw that you state in an earlier post that you are trying to avoid a plastic adaptable box, though to be honest that's exactly how I've done this in the past.
In one case it was three core SWA so easy to simply terminate the cpc and joint L and N to a new 2 core SWA.
In the other case I actually used a box within a bigger box and the incoming and outgoing 2-core SWA cables were a good distance apart, and the double enclosure meant that even if the outer shroud came off the inbound PME earthed cable it wasn't accessible.

 

[marcuswareham]

In John's example there are no exposed conductive parts, so it's a different scenario & little risk involved - the foundation earthing mats he mentioned briefly, I understand may become a thing in the next version of the regs - necessary I gather for "Islanding" purposes if you want to take a solar installation off-grid when you have a power cut. I commented on that video asking John to demonstrate the SWA isolation in another viseo, as I haven't seen any of the usual YouTuber suspects do it, but JW hasn't responded (as yet anyway).

David Savery has isolated SWA amour in 2 of his videos, when changing from a TNCS earth to TT, one in his EV charger video and another in his hot tub video, In both case a plastic box was used, As iv said iv used a plastic module enclosure before as it also gives me space for the fault protection RCD,

As mentioned by timhoward using a plastic box does help to provide more seperation between the 2 earthing systems and also makes it harder / less likely they will be connected in the future by mistake, etc

At a friend's house he has the SWA terminated in a plastic junction box, and then some plastic conduit between the junction box and a metal clad socket, the amour stops at the gland and is isolated but the inner cores continue through the conduit to the socket, looks like a nice neat way todo it and without any extra cable joints

 

[pc1966]

I would use a 25mm plastic gland (they IP68 things) as it seals/clamps the cable and allow it to remain isolated. Sure it is not going to deal with the same sort of pull-forces that a SWA gland would, but if the cable is adequately clipped/cleated to restrain it then I don't see any issue.

Allows a metal box if you want, as that would be on the TT earth then, so all that remains is to make sure there is little risk of a sub-main L to E (TT side) fault inside the box.

If you are really paranoid, have the SWA fed from a 100mA delay RCD, then if the rod Ra is low enough (and it ought to be 200 ohms or less anyway) the supply would trip on that fault, and the final circuits on 30mA double-pole RCBOs like the Wylex/Crabtree/Fusebox compact ones would be selective and offer the additional shock protection needed.

Finally if it is a workshop I would look at some sort of emergency lighting in case the power fails for any reason, as being plunged in to darkness with sharp tools all around is not nice!