Non-Protective Non-Equipotential Non-Bonding

[securespark]

Saw a job today where a well known DIY store started by two gents called Richard & David had installed a kitchen. On arrival I did an earth loop & checked the CU & PEB's.

A quick glance at both the gas meter and stopcock showed a 10mm² conductor fastened to both. So far, so good.

But hang on, that conductor just links one pipe to the other...it isn't connected to the MET...

A resistance test showed the gas PEB @ 18 ohms and the water PEB @ 146 ohms with reference to the MET!

 

[ColJack]

and the maximum permissable if an RCD is fitted is????

 

[Spark123]

0.05 ohms

 

[ColJack]

and where exactly is this stated in the regs..?

this has come up in several posts in the past days, so I'd like to know for future reference..

the only reference i've found so far is for suplimentary bonding.. but that's not main bonding is it...

 

[Spark123]

iirc is derived from the current required to disconnect a 100A service fuse within 5s and the 50v rule.
The only place I can recall it mentioned is in guidance note 3 under PIRs.

 

[ColJack]

so you just glossed over the RCD part of my question then?

100 A service fuse

R <= 50V/550A = 0.09 ohms.. ( for 5 seconds )

R <= 50V/980A = 0.05ohms.. ( for 0.4 seconds )

RCD

R <= 50V/0.1 = 500ohms ( for 100mA RCD )

R <= 50V/0.03 = 1667ohms ( for 30mA RCD )

so if a modern "17th edition board" or a TT supply is installed, what do we take as the required minimum?

 

[Spark123]

As far as I am aware, we still take 0.05 ohms as the maximum for main PEBs.

 

[securespark]

I have just taken an internal examination (you know what I mean!!) and that was one of the questions: the maximum resistance of PEB's must not exceed 0.05 Ohms.

As for the RCD, it relates to the maximum touch voltage of 50.

So, for a 30mA RCD to reliably & safely disconnect the supply in the event of an earth fault, the maximum EFLI cannot exceed 1666.67 Ohms.

As for the figure of 1667 in the good book, I disagree. It should be 1666, as 1667 computes a touch voltage > 50.

As for your question about the required minimum for TT supplies, the disconnection time is listed in the regs is 0.2s.

 

[ban-all-sheds]

I have just taken an internal examination

On yourself?

Are you double-jointed?

Back to the plot - I'm beginning to worry how, with the seeming encouragement of the 17th, people are increasingly regarding fitting an RCD as the way to fix any bonding and fault-loop etc problems...

 

[securespark]

I think as a temporary measure, it is an acceptable fix (like putting an RCD on scouseneil's TT supply), but long-term, faults should be rectified.

I remember a board change I was on in 1991. My boss seemed quite keen that an RCD would happily protect the householders from an excessive Zs. But, I argued, the Zs should be within spec anyway, RCD or no RCD.

I have since argued that point many times and often regarded as a bit anal because of my insistance that Zs figures should meet requirements.

 

[ricicle]

I remember a board change I was on in 1991. My boss seemed quite keen that an RCD would happily protect the householders from an excessive Zs. But, I argued, the Zs should be within spec anyway, RCD or no RCD.

I have since argued that point many times and often regarded as a bit anal because of my insistance that Zs figures should meet requirements.

Unless you are on a TT and the Zs can't meet requirements for the disconnection times with fuses/mcbs.
But I do concur - what would you rather rely on, an electromechanical device or some correctly installed bonding ?

 

[securespark]

Mais oui, mon ami!

I was taking it as read that I was referring to TN systems. TT systems of course need RCD protection regardless.