2-Core or 3-Core

Darkness · 31 May 2011

I have a trench dug from my garage to another large building around 80m away. I am going to get some swa to lay in it before covering it up.

I am just wondering would I be best to buy 2-core, earth the steel wire armour at Point A (garage) but not use this earth at the new shed Point B. Here I will install a new earth rod.

Or buy 3-Core and use the third core to carry the earth to the new shed which will not require an earth rod?

Cheers for any input.

1john · 31 May 2011

What is the supply type?

What is the Zs at the garage?

What size cable feeds the garage?

What size cable are you planning to install?

What is the proposed load?

What size protective device are you using?

Darkness · 31 May 2011

Single Phase TT System

Supply to garage is 16mm² (40A MCB protected by a 30mA RCD) and I was planning to install 25mm² to be safe.

Load should be no more than 4000w at one time.

Lectrician · 31 May 2011

What distance is the garage from the supply?

peterduckett · 31 May 2011

Do you Really need to protect SWA at the house with an RCD?

Darkness · 31 May 2011

The supply to the garage is around 40m max due to the route and it is seperate from the house hence the RCD. Two feeds from the meter, one for house and one for garage.

Taylortwocities · 31 May 2011

I would have to RCD at the garage end too.

For 16mm² I would start off with a switch fuse (60 or 80amp would be OK for 16mm² XLPE). Your switch fuse would need to be insulated (ie not metal) as you have TT system.

16mm² is more than enough for 4KW.

JohnW2 · 31 May 2011

Taylortwocities said:
Your switch fuse would need to be insulated (ie not metal) as you have TT system.

I've seen this said before. Where does it come from, and does it apply to things other than switch fuses? - and does it still apply if there is an upstream RCD?

Kind Regards, John.

Spark123 · 31 May 2011

It doesn't matter about having a metal switch fuse if there is an RCD upstream, however I'd question having a bog standard 30mA protecting the whole lot. It would be better a with time delay RCD (Type S) with local 30mA normal RCDs.
As for 2c or 3c, I'd imagine if it is on RCD then 2c would be adequate using the armour as the earth but still better to have a local rod to the building.

flameport · 31 May 2011

JohnW2 said:
I've seen this said before. Where does it come from, and does it apply to things other than switch fuses?

If some fault internal to a metal consumer unit resulted in a live wire contacting the metal case, the casing would become live but the main fuse would not blow.

Therefore an upstream RCD is required, or use a plastic consumer unit.
It's the same situation as any other metal cased item connected to a TT supply.

JohnW2 · 1 Jun 2011

flameport said:
JohnW2 said:

I've seen this said before. Where does it come from, and does it apply to things other than switch fuses?

Click to expand...

If some fault internal to a metal consumer unit resulted in a live wire contacting the metal case, the casing would become live but the main fuse would not blow.
Therefore an upstream RCD is required, or use a plastic consumer unit.
It's the same situation as any other metal cased item connected to a TT supply.

Thanks, but I see that I expressed myself very badly

. I knew 'the reason' (in engineering/common sense terms). When I asked 'where does it come from' I was asking (or trying to ask!) whether the regs actually say this - and, if so, where.

Of course, the problem is not just that the case of a metal CU etc. could become live without any protective device operating, but also that all exposed or bonded metalwork in the premises would also become live. The same is presumably equally true even with an insulated CU/switch fuse/whatever if an L-E fault upstream of any RCD occurs within it.

Kind Regards, John.

Lectrician · 1 Jun 2011

It is TT - you HAVE to have an RCD at the start of the SWA to the garage, it does not need to be a 30mA, a 100mA Type S would be typical, with a 30mA (where required) in the garage.

Commenting about the insulated switch fuse is pointless - By saying this you are implying that the feed to the switchfuse will have no RCD protection on a TT, and therefore the armour will not either.....

40m of 16mm may well sit on a 60/80 fuse, but you are not taking into account volt drop.

16mm, 40m, 40amp will drop just over 4.18v at full load.

16mm, 80m, 20amp (4 ish kw) will drop 4.18v at full load.
25mm, 80m, 20amp (4 ish kw) will drop 2.78v at full load.

As lighting can only drop 3% (6.9v), both 25mm & 16mm are too small for the second submain. Only just over, but you need to allow capacity for the final circuits too. Very unlikely to load both the submains to their max at once too.

I would re-measure your cable runs with a meter wheel to make sure you are accurate.

mfarrow · 1 Jun 2011

JohnW2 said:
When I asked 'where does it come from' I was asking (or trying to ask!) whether the regs actually say this - and, if so, where.

It is 531.4.1:

"If an installation which is part of a TT system is protected by a single RCD, this shall be placed at the origin of the installation unless the part of the installation between the origin and the device complies with the requirements for the protection by the use of Class II equipment or equivalent insulation (Section 412). Where there is more than one origin this requirement applies to each origin."

If you refer to section 412, in particular 412.2.4 Wiring Systems, then this refers to non-metallic sheathing of cables as providing the reinforced insulation neccessary to meet the requirement.

I don't know whether you could count the inner sheath of an SWA cable as meeting this requirement. Also, you'd have to consider how you'd earth the armour in that situation: insulation of the earth pit, touch voltages, step voltages, safe testing etc.

JohnW2 · 1 Jun 2011

mfarrow said:
It is 531.4.1:
"If an installation which is part of a TT system is protected by a single RCD, this shall be placed at the origin of the installation unless the part of the installation between the origin and the device complies with the requirements for the protection by the use of Class II equipment or equivalent insulation (Section 412). Where there is more than one origin this requirement applies to each origin."
If you refer to section 412, in particular 412.2.4 Wiring Systems, then this refers to non-metallic sheathing of cables as providing the reinforced insulation neccessary to meet the requirement.
I don't know whether you could count the inner sheath of an SWA cable as meeting this requirement.

Thanks. Yes, that seems to cover it, provided the insulated CU, switch-fuse or whatever actually qualifies as Class II - although I can see some people getting confused by "If ...protected by a single RCD" into thinking that it doesn't apply to their installation.

In terms of wiring, 412.2.4 seems to rather 'spoil' it a bit, since (unless I'm misunderstanding) it does indeed appear to be saying (in conjunction with 531.4.1) that any length of any cable with a non-metallic sheath (i.e. virtually any multicore cable) has 'adequate mechanical protection' for the cable to be unprotected by a RCD in a TT system. That doesn't really seem to be fully 'within the spirit', so I wonder if I'm misinterpreting?

mfarrow said:
Also, you'd have to consider how you'd earth the armour in that situation: insulation of the earth pit, touch voltages, step voltages, safe testing etc.

Indeed. I find it difficult to be sure whether 412.2.2.4 theoretically allows a CPC to enter a pre-RCD Class II enclosure in order to connect to the armour (and CPC, if appropriate) an outgoing armoured cable - but, as others have said, I'm not at all sure that one should be doing this, whatever 412.2.4 may say. If the enclosure is the one which actually contains an RCD (so that the outgoing cable was RCD-protected, I suppose it could still be said that there was an issue with having a CPC within it, but that would perhaps be getting rather pedantic.

Kind Regards, John.

mfarrow · 1 Jun 2011

JohnW2 said:
any length of any cable with a non-metallic sheath (i.e. virtually any multicore cable) has 'adequate mechanical protection' for the cable to be unprotected by a RCD in a TT system. That doesn't really seem to be fully 'within the spirit', so I wonder if I'm misinterpreting?

No, that's exactly what it is saying, or how else would you get your electricity from the meter into the RCD enclosure?

JohnW2 said:
Indeed. I find it difficult to be sure whether 412.2.2.4 theoretically allows a CPC to enter a pre-RCD Class II enclosure in order to connect to the armour (and CPC, if appropriate) an outgoing armoured cable - but, as others have said, I'm not at all sure that one should be doing this, whatever 412.2.4 may say. If the enclosure is the one which actually contains an RCD (so that the outgoing cable was RCD-protected, I suppose it could still be said that there was an issue with having a CPC within it, but that would perhaps be getting rather pedantic.

Note much thinking required there - the reg is for CPCs of other circuits, not the one connected to the RCD.

diynot

If you need to find a tradesperson to get your job done, please try our local search below, or if you are doing it yourself you can find suppliers local to you.

Select the supplier or trade you require, enter your location to begin your search.

Are you a trade or supplier? You can create your listing free at DIYnot Local

2-Core or 3-Core

Similar threads