Earthing - yes yet another electrical earthing question or two

Our house (build in 50's) is in a city suburb. It is fed by overhead cables looped house to house. None of these cables provide an earth.

We just have a phase and neutral running down the side of the house to a beefy fuse and then to the usual switchgear. The earth that there is, (another) beefy conductor from the metal copper cold water pipe to a brass block next to the consumer unit to which all the earth conductors from the ring mains join it. The (split load) CU has mcb's and an rcd (which works)

My questions are;

- is this adequate earthing?

- If it is not, what should I do about it?

- If changes need to be made, do I have to pay or is it the supply company?

We have never received any communication from the supply-company about this in 30 years of living here.

ot: Checked The old dear's house next door for earth - and found a thin conductor meanders from a block near the CU, outside the house, up the wall, to the front porch where it ends in mid air! I didn't want to worry her so said nothing. But she has an electric shower....! and no rcd fitted - in which case the lack of an earth may be a good thing for her.

ps Should a metal kitchen sink unit be equi-pot bonded to earth? Ours is - which means it's bonded to the metal cold water pipe. thanks.

Reply to
dave
Loading thread data ...

Sounds similar to my last place...

As usual a photo would help avoid confusion.... but anyway:

Using the cold water pipe as an earth was not uncommon in the past - my place used to use the gas pipe. Often installs like this would include an ELCB with the earth connection made through it. However, use of a service pipe as an earth is no longer permitted. It may still be safe, however the risk is that someone will replumb the water main in the street in plastic, and provide new feeds to the properties, and your earth will vanish.

Now a couple of things worry me in your description:

When you say "a brass block next to the consumer unit to which all the earth conductors from the ring mains join it", do you literally mean this? I would expect all the circuit earths to terminate on the earth bus bar in the CU. Then there to be a single earth connection (in at least 6mm^2 earth single) between the bus barin the CU and the main earth terminal (in fact in many installs the earth bus bar in the CU will *be* the main earth terminal).

Secondly, you mention a split load CU (fine), but an RCD - as in one. I would expect there to be at least two in your case.

No.

You will need to install a dedicated earth electrode, and connect this to the main earth terminal.

All the detail is here:

formatting link
need to ensure that *all* circuits have RCD protection to ensure adequate disconnection in the event of a fault. The traditional way of doing this was with cascaded RCDs - a 100mA trip with time delay first protecting all circuits, and a 30mA trip for the socket circuits. While not compliant with the 17th edition, it may be the simplest path to a workably safe system in your case.

You will need to instate main equipotential bonds to any other incoming service pipes etc (you existing earth connection to the water main, and function as your main bond to that)

formatting link
- If changes need to be made, do I have to pay or is it the supply > company? You pay - its all on your side of the installation. However it might be worth enquiring if you supplier will do you a free (or low cost) upgrade to a PME / TN-C-S installation.

If you do need to do work, then depending on what you actually have at the moment, it may be a simple as just installing an earth rod.

No surprises there then...

I don't sound much like a good thing from here! Are you sure there is really no earth at all? i.e. a connection to a pipe or something that may not be immediately obvious?

Perhaps there is a ELCB rather than a RCD?

For and example, see:

formatting link
ps Should a metal kitchen sink unit be equi-pot bonded to earth? Ours

There is no requirement for it to be, however they often are. No harm is done if it is. (chances are it would get fortuitous bonding via the plumbing and taps anyway).

Reply to
John Rumm

called TT

or go high impedance anyway. The RCD should do the job with a lousy earth, but its certainly not great practice, as if the RCD fails, you end up with a positively dangerous setup.

In what way is that a problem?

Hammer an earth rod into the ground, connect it to the existing earth connector block.

I'm assuming the OP has a whole house RCD, which should be safe enough. If thats right, the rcd protection would only need upgrading if it trips repeatedly some day in the future.

CH system.

your house is your responsibility.

Usually in cases like that there is an accidental earth, eg via the waterpipe etc. Not good, but not quite as grim as it first looks.

no

There are possible question marks over the exact nature of the CU & related bits, I'd suggest posting a pic.

NT

Reply to
meow2222

or with an intermittently dangerous one that works just long enough to convince you it is still ok!

Well I can't quite visualise how you would do it easily with modern T&E cable. Its the sort of thing that you used to see on old rubber or plain twin PVC installs where the CPC for a circuit would be a separate bare conductor, which if you were lucky made it back to the earth terminal at some point. Hence in itself - it may be providing a workable earth, but may be indicative other other lurking problems.

Possibly, but that does not sit well with the split load CU, which makes me wonder.

plus the usual unexpected darkness problem...

Not something you would necessarily want to rely on though - it only takes a bit of plastic pipe or too much gunk on a compression fitting and your touch voltage in the presence of a fault will start to rise.

The main problem here would be that of the earth impedance. Even if there was a damn good earth from a proper earth spike sunk nice and deep into clay, the chances are the fault loop impedance is still going to be north of 5 ohms. That means any fault current is going to be limited to a maximum of about 45A and quite probably much less. Enough to open a lighting circuit MCB, but probably not enough for a socket circuit (and not quickly enough to offer any shock protection), and certainly not enough on a shower one.

Hence any fault just sits there with the bulk of your mains voltage sat on the metalwork. The current flowing to earth via whatever path you do have will also be potentially warming things up and drying out your earth connection - potentially raising its resistance still further.

Reply to
John Rumm

yup. And if it turns out to be an ELCB it wont trip on all leakage faults either.

Its possible... generally bonding will be there though, and bonding is not critical.

If the water supply is in metal, then there's a long and good earth rod right there. Water pipe earths were a very good choice before plastic supplies showed up. Any supplies still in metal still are.

I'm not condoning any of this, just saying the issues with the set up arent as terrible as is often imagined.

NT

Reply to
meow2222

So is there a separate connector block for the neutral? Do the lead-in cables go to a combined unit? Is there by any chance an earth wire from the neutral connector block to the earthing block adjacent?

Reply to
The Wanderer

==/

John Rumm wrote "The main problem here would be that of the earth impedance. Even if there was a damn good earth from a proper earth spike sunk nice and deep into clay, the chances are the fault loop impedance is still going to be north of 5 ohms. That means any fault current is going to be limited to a maximum of about 45A and quite probably much less"

I take it that the reason for 6mm cable to the earth rod is this

45A ? As the fault loop impedance is usually considerably higher than this 5 ohms, such a cable weight seems unnecessarily large - is this cable size specified in the regs?

Rob

Reply to
Rob G

Indeed - especially as many are not installed as they should be.

In the absence of adequate earthing, or earth fault protection (RCD etc) to clear a fault, it becomes critical - its the last thing keeping you alive.

I agree that the pipe would be a "good earth" - however one has to put that into the context of a good earth - for a TT system. Putting that in other terms - its still a crap earth if you want to rely on it to clear a fault using only a fuse or MCB.

e.g. the last place originally used its gas pipe as an earth. Nice long run of steel pipe in damp clay under concrete. So in TT terms it was a

*very good earth* - it reliably and year round archived about 11 ohms. However, that meant that in the absence or failure of a RCD, the maximum fault current was only about 21A. This would not be able to clear a fault on most circuits bar lighting. Its also well short of the < 1 ohm *total fault loop impedances* required to open a 40A type B MCB in the 0.1 to 5 second range of times specified for fixed equipment on TN style systems (let alone the more stringent requirements for TT ones)
Reply to
John Rumm

I have actually glossed over some detail there - well spotted!

The 6mm^2 is specified as a conductor size for the main service bonding. The main earth connection size is a bit more complex. If the wire is not buried, then the min size is 4mm^2 unless it is protected against corrosion and mechanical damage - in which case only 2.5mm^2 is required.

If the wire is buried, then you need 25mm^2 for an unprotected wire,

16mm^2 for one protected against corrosion, or 2.5mm^2 if protected against corrosion and mechanical damage.

And just to top it off, these all assume that the measured impedance of the earth is >1 ohm (which it normally is). If its not then you revert to the guidance for a main earthing conductor used on a TN system.

So in answer to your question - yes, you choose a conductor size such that is it adequate to deal with the potential fault current, but also with the other requirement of adequate physical robustness.

Bonding conductors are selected to be adequate to limit touch voltages, but also have to allow for the fact the bonded service may provide fortuitous earthing, and hence they also need to withstand a certain amount of fault current.

Reply to
John Rumm

This thread appears to have gotten into cart and horse territory.

The priority in this situation is to determine the specification of your supply. It may be TN-C-S TN-S or TT. You need to discover that before proceeding further.

Other posters have, it appears, assumed you have a TT supply. In fact it is possible you may have a TN-C-S supply (for why see below).

In a TN-C-S system (aka PME-Protective Multiple Earthing) the neutral of the supply acts as a shared earth and is earthed at multiple points in its path from the supply origin. In addition - at take-off points such as your house switch gear - it is anchored to local points to form local equipotential planes - such as at your house incoming copper water pipes, gas pipe, oil pipes etc.

Take a look at the incoming main fuse board area.

A TN-C-S supply has an incoming Phase (Line) goes to the main fuse (typically 60A, 80A or 100A) whilst the incoming Neutral goes into an unfused terminal block which is usually beside and part of the fuse block plastic.

Outgoing from the Neutral junction box should be 2 cables. One goes to the Meter the other goes to a PME terminal box - which maybe just a common or garden junction strip. Also incoming into that PME terminal box should be earth cables ("Main equipotential bonding conductors") connecting to each incoming house service pipe. The main installation (house) CU earth is taken from here to the CU (internal) earthing bar to form the main house earth. If there are multiple CUs then another earthing box might be fitted to feed earth to each CU's earthing bar.

So Q no 1: Is that what your incoming supply & CU boards look like?

If yes, read on.

A split-load CU & RCD speaks of a newish (within last 10-15 years?) upgrade to your supply.

Was your supply possibly changed to TN-C-S then?

Having established your type of supply check:

  1. The earthing geometry against the diagrams in the IEE On-Site Guides (which for TN-C-S is more or less what I have set out above)

  1. If it matches the diagrams check that bonding (earthing) cables are the right size.

If in any doubt check with the elec company as to what supply they think you have (TN-C-S; TN-S or TT) Also look around the incoming supply fuse for the magic letters 'PME'.

Then specify the changes needed (which might be only a need to tidy up the wiring and clip free hanging cables to the backing board etc) & get it done.

If the defects are your side of the main fuse, you pay (or DIY which is what we are all about).

Nor or you likely to unless some major fault or incident occurs. It is quite possible it has been changed without your knowledge or by whoever upgraded your CU/ etc. Do you know how it was when you moved in? Who installed the split mode CU & RCD? It might have been changed over then.

////snip///

That is a contentious issue. Probably most are, but there is a risk of shock in (different) fault conditions either way. The current fashion (17th Edition) is to supply the whole kitchen from RCD protected circuits.

If unclear please repost for clarification.

HTH

Reply to
jim

OP has an RCD though

Reply to
meow2222

Neighbour does not though by the sounds of it though...

Reply to
John Rumm

And carry out the necessary testing thereafter...

Another option of course, which no-one seems to have mentioned (unless I missed it) is to enquire of the supplier or the DNO whether PME conversion is available. (That's if it's not PME already, which as The Wanderer and jim have suggested, is a possibility.)

Quite possibly she's in great danger, if it's not been PME'd and should the insulation in the shower unit fail. Someone who knows what they're doing ought to take a look.

Reply to
Andy Wade

I did request a photo for that purpose (and suggest he see if its available from the DNO)...

The fact that he spoke of a split load CU and RCD in the singular, could be explained by a PMR install even if the description of the earthing does not seem to tie up.

Yup, I would second that.

Reply to
John Rumm

On Thu, 16 Apr 2009 08:42:41 -0700 (PDT), jim wrote: First of all very many thanks for all the detailed replies and my aoplogies for not responding - but, I've been away and couldn't respond. I'll read all the followups and try to get the details - I can see my OP needed a better description - asap. thank you all again.

Reply to
dave

again my apologies for delayed reply.

I thought the best thing was (as suggested) was to take some pics of the settup here and so...

Elect distribution area to house is:-

formatting link
are no signs saying it is a PME installation (it was wired during the 50's but clearly updated CU and Earths redone since then).

Is there an electrical way to test if I have a PME installation? (measure the voltage between earth and neutral maybe? That is, the earth anchored at the cold water supply pipe and the incoming neutral).

There are definitely just two cables entering the building (drops from roof looped from the "previous building". These cable can be seen coming through the board in the pics.

Thanks

Reply to
dave

Definitely not PME at the house, but....

No. Providing PME is entirely a security of connections issue, and installing additional earth electrodes on the distribution netwrok. It's not something you can measure, but....

The leccy industry quite a few years ago was making efforts to have pme available on all its networks. You may already have PME available. A quick telephone call to the local distribution company (the lot who own the wires) should be able to confirm for you. You will have to ensure the installation meets the requirements of the latest edition of the IEE wiring Regs as far as earthing goes.

Reply to
The Wanderer

From what is shown on the photos it does not look like PME either... you would normally expect to see a substantial yellow/green wire connected to the neutral terminal block.

Where do the other set of tails go? (the ones that run vertically to the right of the supply fuse)? The layout seems a little odd since the split is before the meter.

Reply to
John Rumm

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D\

=A0 =A0 =A0 =A0|

=A0 =A0 =A0 =A0|

=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D/

Looks like TT, earthed via the water mains to me. All the Henley blocks look sealed, so there is no easy way to nick the electricity without getting caught - unfortunately...

T
Reply to
tom.harrigan

A quick question.

I worked in a terraced house the other day. The supply was PME and came through the wall from the next door neighbours cellar. All terminal blocks present and correct and installed by YEDL.

The next door neighbour asked me for a quote for some work doing.

The incoming supply to the neighbours house was TN-S and came in underground to the cellar with the old lead sheathed earth cable supplying the earth. YEDL had tapped off this TN-S supply to power the next door neighbour with a PME (35mm single armoured cable) supply.

I was a little concerned but the Ze was 0.11 ohm.

Is that normal practice?

Adam

Reply to
ARWadsworth

HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.