Earthing arrangement for shipping container on building site

[BrianHJR]

I’ve had a shipping container delivered to my building site for use as storage/utility room/worker toilet. One side is insulated as it was previously used for exactly this, and it already has some fairly rudimentary electrics installed.
On the previous site, their “groundworker’s electrical contractor” had connected it up to their boundary meter kiosk with a length of SWA with commando plug inline, going through a hole in the side, and straight into a metal consumer unit mounted on the lining.
It appears that the earth was exported along the SWA, there was no local provision for earthing, and there is no obvious linking of the protective conductor to the metal body of the container.

On my own site the container is sat on a thick layer of crushed limestone, about 20’ from the electricity meter box, where the incoming temporary supply is TT earthed via a rod underneath.

I’m trying to establish what the correct earthing arrangement should be in this situation?

 

[westward10]

If your source of supply is TT then it will be TT.

 

[BrianHJR]

If your source of supply is TT then it will be TT.

Thanks.
Can the TT earth be exported to the container though? Can the conductor supplying protective earthing from the TT spike a distance away be replied upon to provide suitable protective earthing to a large metal object? Or should there be a separate spike at the side of the container?

Am I correct in thinking that the metalwork of the container should have been bonded to the protective conductor regardless of the earthing arrangement?

 

[timhoward]

The distribution circuit for the container needs a CPC and this needs earthing, which would usually be done the supply end.
The 20 foot distance you speak of is not material; the cables impedance is a fraction an ohm compared to the impedance to real earth via the earth rod which will likely be many ohms.
I don't see an extra earth rod being required.
Normal bonding rules apply.

The advantage to an additional new earth rod is the chance to improve the earth loop impedance but in this situation (to my thinking) you want to avoid the distribution circuit's CPC becoming a preferable main earth path for the existing supply, so I'd insulate the CPC at the container end by glanding onto a plastic box and just connect the new rod to the earthing and bonding at the container.
Hope that makes sense.

 

[pc1966]

The container metalwork should be bonded to the internal CPC system in any case.

You don't need an additional earth rod as the supply CPC ought to meet the required Ra anyway. Another rod won't do any electrical harm and would marginally improve the reliability of it, however, you have to be careful where you drive one in just in case there are any buried cables or services!

 

[pc1966]

Just to add that in any TT system you have to be very sure that a live-earth fault before the RCD is very very unlikely as nothing will disconnect in that case!

If the supply has some RCD protection already, such as 100mA or 300mA delay, then an earth rod arrangement below 167 ohms would allow supply fault disconnection before the container's RCD. But you should not assume that!

So for the wiring in to the cabin and to the first (presumably CU incomer) RCD it should have additional protection, either double-sheathed cable or in plastic conduit, etc, so such a fault is extremely unlikely.

 

[timhoward]

The container metalwork should be bonded to the internal CPC system in any case.

Thanks for clarifying - That is what I intended to convey by saying "normal bonding rules apply", but I can see it wasn't very clear.

My assumption was that the container supply would be downstream from the RCD providing fault protection for the main supply.
But as said above this absolutely needs testing properly before proceeding.

I'm also suspecting that Section 717 applies, as that mentions containers. If so I believe it needs an RCD no greater than 30ma protecting the supply. There's other rules about using flexible cables too....

 

[pc1966]

I'm also suspecting that Section 717 applies, as that mentions containers. If so I believe it needs an RCD no greater than 30ma protecting the supply. There's other rules about using flexible cables too....

Yes, here with a plug/socket supply it ought to have a 30mA RCD already.

But most containers I have seen also have RCD inside as they could be hard-wired in which case the supply may not have 30mA RCD but possibly higher & delay if TT, or none at all if TN on SWA cable. Sure it is a pain that you could trip two RCDs on a test, etc, but much less of a pain than being on the receiving end of an electric shock with a stuck/failed/missing-supply RCD!

I did wonder about the flexible cables aspect, as in my own limited experience of these things for site offices they can be wired in T&E and not the flexible stuff mentioned in the likes of sections 708/717 (though the list in 717.52.2 includes 7-strand conduit wire was well as fins stranded stuff).

 

[BrianHJR]

Thanks everyone.
Theres going to be an electrician come to do all the testing but I need to have the installation more or less ready with cables pulled etc beforehand or it will cost me a fortune…
I will also run it past them in advance to make sure they’re happy with the proposals too.

 

[BrianHJR]

The container metalwork should be bonded to the internal CPC system in any case.

You don't need an additional earth rod as the supply CPC ought to meet the required Ra anyway. Another rod won't do any electrical harm and would marginally improve the reliability of it, however, you have to be careful where you drive one in just in case there are any buried cables or services!

Fortunately very little chance of hitting any unexpected buried services as the only ones there are ones that I’ve installed myself - 2 x water supply pipes and a temp electric run nicely tucked out the way down the side of my caravan!