Earthing a secondary CCU

[RF Lighting]

The CPC is big enough assuming thet the new submains in not supplying main protective bonding.

It is a poor choice for a submain though, and will normally require RCD protection if it is flush mounted.

 

[Spark123]

So, just to confirm, my 4mm earth conductor in the T+E is suffient ?

Is it just for earthing the electrical circuits or is it to be used for main protective bonding to water pipes etc too?

Beaten by RF by milliseconds

 

[mfarrow]

OOI, where did you get the k rating of 100 from?

Table 54.3, 90 degrees thermoplastic (i.e. worst case).

 

[mfarrow]

... will normally require RCD protection if it is flush mounted.

Why's that then?

 

[Spark123]

I'd assume the cable to be 70deg cable unless stated otherwise, the cable and the terminations must be rated for 90deg for the cable to be used at 90deg.

 

[RF Lighting]

... will normally require RCD protection if it is flush mounted.

Why's that then?

Regulation 522.6.6

 

[mfarrow]

Regulation 522.6.6

But of course... re-read your message after I'd posted - thought you'd meant something else.

 

[GaryMo]

I'd assume the cable to be 70deg cable unless stated otherwise, the cable and the terminations must be rated for 90deg for the cable to be used at 90deg.

That's what I was thinking, hence the confusion with a k value of 100.

 

[mfarrow]

No confusion intended. It's a worst case calculation, based on not knowing or seeing the installation. He might not even have a 100A cut-out.

The calculation was merely to demonstrate it's use and application, and how Table 54.7 can be ignored in most domestic installations.

 

[ColJack]

OOI, where did you get the k rating of 100 from?

nevermind that, since when did 239/100 = 0.75?

 

[ColJack]

what time do we use in the adiabatic anyway? it's been a long time since I used it..
5s? 0.4s?

 

[GaryMo]

OOI, where did you get the k rating of 100 from?

nevermind that, since when did 239/100 = 0.75?

I missed that bit

He should have stated a minimum of 2.5mm2

 

[GaryMo]

t is the time taken to disconnect the fault current for the particular device in use (from the time/current characteristics in Appendix 3).

You may need to refer to manufacturers data for time less than 0.1s.

 

[mfarrow]

nevermind that, since when did 239/100 = 0.75?

It doesn't

OK so we have 2.4mm². I'll therefore retract my last statement about 54.7 being completely useless.

It raises an interesting point though, and that is if you strike a nail through a cable at or near the consumer unit, assuming we were protected by MCBs then the cable would be very toasty . You may not see a fault level high enough not to trip the MCB before enough energy is let through, but it could happen.

 

[GaryMo]

Have you got an example?