RCD Protection

[ban-all-sheds]

Well, as you say, there's the cabling to consider, as you'd have to get two lots into the outgoing terminals of the incomer, or take them to a couple of rail-mounted distribution terminals, e.g.

And of course you'd have to get more singles to use inside the CU.

And you're going to lose space for MCBs.

Then you'd have to shorten one of the busbars, as it would be designed to come out of the incomer.

And (I think?) you'd need to split the normally non-RCD neutral bar in two.

Mind you - I found this on the Hager website, so maybe if you want one, you can buy one ready made from them:

 

[ban-all-sheds]

Another assumption i made was that if the electrician goes for installing an earth rod he would simply be able to put a 100amp 100ma time delayed RCD in a separate box before the CU.

Yes - he could do that.

But if the electrician goes to the trouble of getting the supply converted to pme and sorting out any defects in the existing installation ( required by REC before they will put an PME conversion in ) the 100ma time delayed RCD would not be necessary , or have i made to many assumptions here.

No that's quite right - if you get a PME supply you no longer need an overall RCD.

 

[Damocles]

BAS, why is it necessary to have sockets on 30ma rather than 100ma rcd to satisfy disconnection time? I do not see what one has to do with the other. The time difference between 100mA and 30mA trip will be bugger all in the event of an earth fault. (which is why you need a time delayed one!) And completely irrelevant in the case of a live-neutral fault. ?

 

[ban-all-sheds]

The tripping current is irrelevant, but a 100mA RCD is allowed to have up to a 300ms trip time.

Of course, it might go faster.

But it might not.

And the regs say that where an RCD is used to provide protection against indirect contact or supplementary protection against direct contact it must trip in <200ms at In.

They also say it must not trip at less than 0.5In, so your 100mA RCD will pass a 50mA fault current indefinitely. Not good if it's passing through you....

 

[Damocles]

Is it just me, or does no one else see the absurdity of this.? The guy has an unsafe installation with no earth and the rec refuses to convert it to pme thereby providing an earth and making it mostly safe until he has connected up various pipe bonding to the currently non-existant and hence useless earth? If they really do have the remotest duty of care to the safety of their customers then a pretty stern letter of complaint seems justified.

 

[Damocles]

Um, so? I thought what was required was a circuit which tripped in 300ms. That's fine then.

 

[plugwash]

PME has risks of its own

the bonding should definately be connected BEFORE connecting a PME

actualy im not convinced that water pipe earthing is as bad as people make out

if the undergound metal then its an earth and should provide disconnection if is plastic then you just end up with a veriation on earth free equipotential bonding

 

[ban-all-sheds]

Um, so? I thought what was required was a circuit which tripped in 300ms. That's fine then.

Ah -- I was busy editing when you did this. Sorry.

No - it's not fine.

 

[Damocles]

Good point. So if either his services which seem already to be bonded are metal, or are all plastic. Then he should be safe (more or less) in the event of an earth fault. If there is completely no earth, the pipework may all go live, but he should be ok. How many weeks do you reckon it will be before anyone notices the earth is live?

To get back to the original problem, you have checked the service ends of the bonding cables to check whether there might be an existing earth spike there?

The guys here are just about convincing me that you have nothing to worry about and might as well leave everything as it is (???)

 

[securespark]

if the undergound metal then its an earth and should provide disconnection

An electrical installation should not rely on a service pipe for an earth connection. This went out in 14th Edition, Plug, in 1966!!!!

 

[Damocles]

It does sound ok BAS, but I do have reservations about it.

There is this issue about how reliable a RCD is. Short-circuit through the earth has fewer moving parts to go wrong. RCD plus good earth gives double indemnity, so to speak.

What I am puzzled about is what you are supposed to do in place of table 7 conventional circuits in the on site guide for a TT installation.

 

[ban-all-sheds]

I'm struggling to understand how your reply relates to anything I wrote...

 

[Damocles]

Bas, you said...

Damocles wrote:
Um, so? I thought what was required was a circuit which tripped in 300ms. That's fine then.

Ah -- I was busy editing when you did this. Sorry.

No - it's not fine.

I was agreeing with your previous post disagreeing with me. perverse, aren't I?

I was expressing a reservation about the reliability of RCDs as the only line of protection. Using two RCDs in sequence must improve protection over one alone.

Table 7 is relevant because it lists cicuit designs considered safe in an installation. It ought to list which designs are considered ok for a TT installation, but does not list any. Should I conclude from this that the IEE considers the only sort of circuit considered safe to connect to a TT/RCD supply is one having no lights,sockets,cables or anything else?

 

[ban-all-sheds]

My guess is that since all the circuits in a TT installation will have RCD protection, length-based disconnection limits are not an issue?

If you read the notes you'll see that all unannotated numbers are where volt-drop limit kicks in, so, e.g. a 10.0/4.0 radial on a 32A Type B is limited to 69m, and that limit applies to TN-S and TN-C-S, so by implication, TT as well.

I guess for those circuits where Table 7.1 shows a limit based on disconnection time you'd need to do the volt-drop calc to see what the limit is with TT.

 

[plugwash]

if the undergound metal then its an earth and should provide disconnection

An electrical installation should not rely on a service pipe for an earth connection. This went out in 14th Edition, Plug, in 1966!!!!

not saying tis allowed by the IEE wiring regs

but compliant with the IEE regs and safe are two totally different concepts