Do GN3 and OSG contain errors?

Talk Electrician Forum

Help Support Talk Electrician Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Mizunoid

Member
Joined
Aug 15, 2022
Messages
21
Reaction score
6
Location
UK
On page 69 of Guidance Note 3 it shows a picture of a test of a whole consumer unit for Insulation Resistance.

The info for the test says it shows 'an L-N test' but in the picture it shows one probe in the L of the main switch load-side and one probe on the earth bar.

Is the guide wrong or am i missing something?

On a separate but related note, i keep seeing tests where people remove the CPC from the earth bar to IR test it, but GN3 repeatedly says 'it is essential that the incoming earth connection is connected to the MET of the installation (and that this is connected to the means of earthing)'. Again - does this mean you have to leave the CPC in?

I mean, the wording is really cloudy - if the incoming earth was connected to the MET, the phrase 'and that this is connected to the means of earthing' makes no sense since the incoming earth IS the means of earthing.

What am i missing?
 
I think you have slightly misread it. What it is saying is a line and neutral insulation resistance test of line and neutral, connected together to earth. Not a test between line and neutral bùt line and neutral connected then tested to earth.
Never disconnect a cpc to test, you are testing to the entire earth of the installation not just the cpc of the circuit under test.
 
If you read bs7671 the requirement is for the whole system to meet minimum insulation level of 2 megohm ( might be 1, I can't remember off the top of my head) . Hence 'global' insulation test at the CU).
 
I think you have slightly misread it. What it is saying is a line and neutral insulation resistance test of line and neutral, connected together to earth. Not a test between line and neutral bùt line and neutral connected then tested to earth.
Never disconnect a cpc to test, you are testing to the entire earth of the installation not just the cpc of the circuit under test.
The picture doesn't show them connected together. For the line to neutral test it shows live to earth bar. For an example check OSG page 118 or GN3 page 69. Same picture.

All i can think is that the neutral and the line are both in an RCBO and somehow that makes them all connected?
 
s 'it is essential that the incoming earth connection is connected to the MET of the installation (and that this is connected to the means of earthing)'. Again - does this mean you have to leave the CPC in?

yes. if its disconnected then it could pass IR testing but still have a potentially dangerous fault that would have been detected if it was tested correctly
 
The picture doesn't show them connected together. For the line to neutral test it shows live to earth bar. For an example check OSG page 118 or GN3 page 69. Same picture.

All i can think is that the neutral and the line are both in an RCBO and somehow that makes them all connected?
I agree the picture isn't clear and doesn't show an apparent L to N link.
 
It kinda annoys me that we have to pay so much money for these publications and yet there are loads of errors in them.
 
It kinda annoys me that we have to pay so much money for these publications and yet there are loads of errors in them.

The number of errors gets bigger with each edition …….

I would like to see these documents reviewed by the Plain English Society before it’s updated again
 
It kinda annoys me that we have to pay so much money for these publications and yet there are loads of errors in them.

18th had plenty mistakes... then the **** up correction also contained mistakes...

rushed out to make as much money as possible with little regard to making sure its correct
 
What really amuses me is that the 18th edition in paper version differed from the online version

Who are the gormless idiots who are responsible for its publication?
 
Can use RCD sockets and spurs
Can’t use them
Can use them

So I suppose that f you were doing an EICR and there was an RCD spur/socket that it could be proved was installed during the non-compliant times it would be wrong, yet if done a day later when Amd came out it would be correct

But that only applies when there is an 'R' in the month..
But not including weeks when there are NO repeat broadcasts of Midsomer Murders, Vera, or A Touch Of Frost... And Tesco's have no shelf labels saying "Aldi Price Match"...

Other than that is all quite straight forward...
Wots the issue?? :unsure::rolleyes::p🍻:p:cool:o_O;):rolleyes::rolleyes:
 
It wouldn't be an IET document if it didn't have errors in it, I can never understand how the IET have never heard of or adopted proof reading before publication
 
Top