insulation resistance testing bombshell

[the doctor]

Hello All

Something happened this week, that blew my head off:Chair fall

I cannot go into detail, but have joined forces with a Part P provider. They like the idea of a real electrician writing and teaching and have asked me to take a sort of ambassadorial role:worship

Part of this process involves them bankrolling the new book. However they have said that some of my views do not hold the party line so need 'fettling'. Fair enough I suppose...The man who is proof reading the book is an industry 'heavyweight' I cannot name him, but he is named on page 10 of the new BS7671:Blushing

So, you can imagine, that even me, the part p doctor is well punching above my weight here. However, I know what i know and reckoned he could not 'nail me' so to speak.

That all changed last week when he told me that I and almost all the sparkies in the UK do insulation resistance wrong. At first I did not see it but when it dawned on me, he is right and I, well humbled to say the least.

Does anyone want to know what they are doing wrong then? Guinness

 

[NozSpark]

Taking the CPC out of the earth bar by any chance????

 

[Cirrus]

Do tell!

 

[davetheglitz]

Oh Yes!!! Very interested!!!! On an installation N and L are isolated - E is not. Something to do with that?

 

[the doctor]

how did you know that nozspark? - is it an open secret then?

 

[NozSpark]

Got pulled on an assesment for it :Blushing

 

[kerching]

What Noz said, perchance?...... :coat

 

[Sidewinder]

Ahh, but is has to be "IN" the earth bar first to remove it!

So for IV then perhaps not, but for PIR perhaps yes...

This depends on "how" you do your IV.

 

[the doctor]

Yep

Noz is right...

How long has the rule been in that the cpc has to be connected in? At the old electric board school they taught a guy to test and then connect it in

---------- Post Auto-Merged at 22:16 ---------- Previous post was made at 22:10 ----------

In recent months a person was killed by the practice... let me explain...

1 spark puts in cable down metal jumbo stud wall.

2 cable hit by screw or burr, just down to line conductor

3 spark does IR test- cable passes

4 spark energizes- live wall now

5 boiler springs leak and wets floor under wall

6 customer electrocuted

7 if cpc had been connected in to bar with bonds, bad reading may have showed up through contact between concrete floor and metal stud

i thought initially, so what. When BS7671 explained it was a revelation. He said he was not surprized. He reckoned 99 out of 100 sparks do it wrong, apart from Noz of course

---------- Post Auto-Merged at 22:19 ---------- Previous post was made at 22:16 ----------

I want to know, when did reg 612.3.1 change to say that cpc has to be connected in and bonding present? Who has old regs books lying around?

Me? :red card I need to lie down a while...

 

[Sidewinder]

doc,

I have had the letters etc. on the scenario you describe.

However, on initial install one should verify cables prior to connection to the CU first, thus you are "NOT" removing the cpc.

Once connected, then IMHO they must be verified as still correct.

This would give you the opportunity to rectify any possible install damage prior to final connection/2nd fix etc.

 

[davetheglitz]

Could you detail why this is wrong? Can only think that an interlinked circuit would be a problem - but the readings taken will always be worse case. Perhaps we should be checking for interlinked circuit as an additional test.

 

[Sidewinder]

Oh, it is one of the ways of proving electrical separation, thus must be undertaken, else how can you prove this?

 

[tom1]

same here got told on an assesment pm me the assessor name noz i wonder if it was same bloke

 

[the doctor]

My BS7671 man is correct.

If the circuit is not connected to board you could have a line wire nicked touching the metal stud and always get a perfect reading. If the cpc was connected to the earthbar and bonding, the extraneous metal touching the building fabric could give a path back through the floor to the nicked line on the L-E test....

it has proper messed with my head. :yellow card

---------- Post Auto-Merged at 22:29 ---------- Previous post was made at 22:27 ----------

sorry tom, not yet....

they say i am under a 'confidentially agreement'

 

[Zeespark]

You arent the only one with a confused look on your chevy chase Doc.

AndyGuinness

 

[Sidewinder]

doc,

i do know a few of the 7671 committee along with some of the GN3 committee.

I do agree, however, I still stand by my description.

IV, test cable fully isolated to check for damage.

Connect up then re-test, as described, this proves electrical separation also.

 

[sparkytim]

Always done it that way, often picks up conductors crushed in back boxes etc leaking to earthy walls. If CPC is isolated from earth then it can be easily missed.

 

[kerching]

I, for one, concur with the musings of Reptilian Reginald in #16

as Punchamilo said 'that's the way to do it'....well, that is the way I have done it for as long as I can remember; it just seems to be good practise. belt 'n' braces and all that malarky.............. :coat

 

[steptoe]

:C

Im confused as to how IR is checked here,

do we have to disconnect the earth, L&N and test, then reconnect and test again?

Ive only ever done it the same since I was taught,

check continuity(ring finals obviously) then connect, when everything is connected then global IR, if something shows up then disconnect til you find it,

maybe that was the lazy way so only one IR to do, but maybe in the old days more people took more pride so the chances of a fault were fewer.

so, how am I supposed to test it now? cos Im still lost.

 

[davetheglitz]

I've been looking at this as an EICR - in which case the scenario that is described would be detected.

My interpretation of you are saying is that on installation the insulation measurements should be made with the cpc connected to the earth bar. Fair do's.

However - without the water present how would you know whether the cable was nicked? I'm sure there's a liable case waiting to happen here - and you will always find another electrician ready to label another as a muppet for an honest mistake on a very unlikely scenario.

Should we be flooding our customers houses as a test for electrical insulation?