Protection by RCD

[neil_ev]

Hi all,

Just a quick query on RCD protection. Looking at 3.6.1 of the OSG it states that RCD protection is required for;

ii)socket outlets in domestic installations. - SO ALL SOCKETS SHOULD HAVE RCD PROTECTION? (unless specifically labeled as fridge, freezer ect...)

iii)for circuits of locations containing bath or shower. - EVEN IF THE LIGHT IN THE SHOWER IS RATES TO APPROPRIATE IP?

Also it does not mention outside lights. I'm sure these should be RCD protected?

Thanks in advance

Neil

 

[Plonker 3]

But should you carry out a alteration you have to check Zs values so that should comply which will operate any OCPD downstream of the circuit surely?

 

[spark 68]

But should you carry out a alteration you have to check Zs values so that should comply which will operate any OCPD downstream of the circuit surely?

What is to say a fault cannot occur after the alteration ?

Iam not talking about what if's here Dillib, just what is good practice and about minimising the risks.
I was merely pointing out why it matters where the RCD is placed, and my reasoning behind it.

 

[Deleted member 26818]

Aslo sparks if you look at the definition of a "circuit" in the regs it says

An assembly of electrical equipment supplied from the same origin and protected against overcurrent by the same protective device(s)

It is the term "origin" that is key here I think.

I'm not sure how valid the argument about leakage back into the circuit down stream of the RCD is, I suppose in theory it could happen, but as CPC's are commoned in the CU there is a possibility that a non RCD protected fault on another circuit could leak fault current back into the bathroom via the CPC and any exposed conductive parts.

In the event of an earth fault anywhere in an installation, all CPCs and earth/bonding (and in the case of TN-C-S neutral) conductors will rise to 230V potential.
Therefore any earth fault will allow fault current to bypass any RCD irrespective of where the RCD is positioned.
However as long as all Main Protective Bonding is in place and the main earth conductor is not compromised, there should be little danger.
Where an RCD is positioned part way along a conductor, an earth fault between the origin and the RCD will allow fault current to bypass the RCD, not in itself a particular danger.
However if the conductor's CPC is damaged/disconnected either due to the fault or some other reason, the fault current will have no other route other than to bypass the RCD and enter the location. This is where there is a danger.
As for the origin of a circuit, guidance is provided in BS7671 in Appendix 15, where it states that the origin is at a DB.

 

[HandySparks]

In the event of an earth fault anywhere in an installation, all CPCs and earth/bonding (and in the case of TN-C-S neutral) conductors will rise to 230V potential.

I don't think so.
Surely, relative to "ground" (transformer star point), the voltage on the CPC at a line to earth fault will depend on the impedance of the fault itself (which may be close to zero) and the relative impedance of the line and earth parts of the fault circuit? Unless there is no external earth connection, it'll never rise to the supply voltage.

 

[telectrix]

as i understand it, a line/earth fault will cause the potential of the earthed component to rise to line value, at the point of the fault, for the duration of the fault untill the OCPD activates. obviously, the resistance of the fault path will affect the duration of the fault.

 

[HandySparks]

Hi Tel.
I quite agree that the duration of the fault is set by the resistance of the fault path and the characteristics of the OCPD. Hence why we measure Zs.

Now, "a line/earth fault will cause the potential of the earthed component to rise to line value, at the point of the fault".

Well, yes, except that a line/earth fault also drags down the line voltage at the point of the fault; so the local line voltage is reduced. So, say for example, the line and earth parts of the fault path are of equal resistance, then the voltage at the fault will be half the supply voltage. In effect, the resistance of the two parts of the fault path form a potential divider across the supply.

Or am I misunderstanding?

 

[Deleted member 26818]

In a healthy circuit, the line and neutral have similar if not equal resistance.
Would this combined resistance halve the supply voltage?
In many instances, the CPC has a smaller CSA than the live conductors, as such some volt drop would be expected, but not by a significant amount.
In many cases where an earth fault occurs, the neutral is still connected.
In such cases, the CPC would become a parallel conductor with the neutral, as such reducing resistance.