Blimey, wasn't this a diy job?

[Electrifying]

Yes, it is an 'expected' test and should normally be done. I'll give you a clue........it's the 4th test on most guidance lists and goes by the name of 'polarity'. Perhaps you would be so kind as to tell us how you would confirm 'polarity' on a 'radial' socket circuit, by testing, during 'initial verification' - without using the method I described or a wander lead??

I think a problem here is that guidance, and 'guidance lists', seem to be thinking of 'polarity' in a somewhat restricted sense. In relation to pre-energisation 'polarity' testing, 2.7.11 of GN3 says:

"For radial circuits, the (R1+R2) measurements, made as in test method 1 (Section 2.7.5), should be made at each point"

If one followed that guidance, one would, indeed, pick up L-N reversals and even N-E reversals, but a circuit with an L-E reversal, such as the fatal one we've been discussing in this thread would, presumably pass such a test with flying colours - after all (R2+R1) is the same as (R1+R2).

Kind Regards, John.

No, no, no...... 'polarity' is the confirmation that single-pole accesories/devices are wired in the 'line' conductor only (put simply).

So, answer my question - How do you confirm 'polarity' on a 'radial' socket circuit, by testing, during 'initial verification' - without using the method I described or a wander lead???

And when you can't answer, will you then please concede - for others who are reading - that the method I described, or the wander lead method, are 'required' tests.

You see, the whole point of this......if we're going to speculate about the OP.......is that had 'polarity', of the socket outlet in question, been confirmed in the correct manner, the tragic accident may not have happened...........but some people (including qualified sparks), are under the impression that 'flicking' the socket switch during the R1 + R2 test confirms 'polarity'....IT DOESN'T........that only applies to lighting radials!

 

[JohnW2]

No, no, no...... 'polarity' is the confirmation that single-pole accesories/devices are wired in the 'line' conductor only (put simply). So, answer my question - How do you confirm 'polarity' on a 'radial' socket circuit, by testing, during 'initial verification' - without using the method I described or a wander lead??? And when you can't answer, will you then please concede - for others who are reading - that the method I described, or the wander lead method, are 'required' tests.

Hey, I'm just the messenger. I've reminded you what GN3 says about "How do you confirm 'polarity' on a 'radial' socket circuit, by testing, during 'initial verification'". If you believe that more than that is 'required' (which clearly should be the case, if one wishes to be able to detect an L-E reverse), I'll have to look to you to indicate who/what (other than common sense) 'requires' it.

Kind Regards, John.

 

[Electrifying]

John, are you not reading my posts???

You said:

Yes, it can/could be done on a radial circuit - either by your adaptation of the ring final method or with a long lead - but are they 'expected' tests which would normally be done?

RF Lighting said:

No it is not part of the standard sequence of tests.

I said:

Yes, it is an 'expected' test and should normally be done.

I'll give you a clue........it's the 4th test on most guidance lists and goes by the name of 'polarity'.

Perhaps you would be so kind as to tell us how you would confirm 'polarity' on a 'radial' socket circuit, by testing, during 'initial verification' - without using the method I described or a wander lead??

Which you didn't answer, instead you said:

I think a problem here is that guidance, and 'guidance lists', seem to be thinking of 'polarity' in a somewhat restricted sense. In relation to pre-energisation 'polarity' testing, 2.7.11 of GN3 says:

"For radial circuits, the (R1+R2) measurements, made as in test method 1 (Section 2.7.5), should be made at each point"

If one followed that guidance, one would, indeed, pick up L-N reversals and even N-E reversals, but a circuit with an L-E reversal, such as the fatal one we've been discussing in this thread would, presumably pass such a test with flying colours - after all (R2+R1) is the same as (R1+R2).

Which was you totally not understanding the concept of polarity - by the way, that method, as I already stated, only works on lighting radials!!......

so I said:

No, no, no...... 'polarity' is the confirmation that single-pole accesories/devices are wired in the 'line' conductor only (put simply).

So, answer my question - How do you confirm 'polarity' on a 'radial' socket circuit, by testing, during 'initial verification' - without using the method I described or a wander lead???

And when you can't answer, will you then please concede - for others who are reading - that the method I described, or the wander lead method, are 'required' tests.

So then you reply:

Hey, I'm just the messenger. I've reminded you what GN3 says about "How do you confirm 'polarity' on a 'radial' socket circuit,

NO, You haven't.

You've told us what GN3 says is the method to use......but from your answers I really don't think you know 'what' polarity is, nor how to test/confirm it.

If you believe that more than that is 'required' (which clearly should be the case, if one wishes to be able to detect an L-E reverse), I'll have to look to you to indicate who/what (other than common sense) 'requires' it.

The regulations require it!!
You are required to tick the box to say that you have confirmed polarity on that circuit.

You are confirming that single -pole devices are connected in the 'line' conductor only.

So, I ask you again,

How do you confirm 'polarity' on a 'radial' socket circuit, by testing, during 'initial verification' - without using the method I described or a wander lead???

 

[JohnW2]

You seem determined to keep me out of hibernation - have you got a wager going with someone?

So, I ask you again, How do you confirm 'polarity' on a 'radial' socket circuit, by testing, during 'initial verification' - without using the method I described or a wander lead???

I cannot think of any other foolproof way; if I wanted to do it, I personally would use a long lead. I frankly find it hard to believe that (m)any people routinely use 'the method you described'.

However, this discussion slipped onto the question of 'what is required' and you seem to be dismissing what I said about the guidance in GN3 - which many would regard as a statement of what is 'required'/acceptable. I quoted what 2.7.11 ('Polarity') of 2.7 ('Initial Verification') said about the method for testing 'polarity' in a radial circuit - and, whether we like it or not, many people will (not unreasonably) take that to indicate that such a test is what is 'required' - or, at least, is acceptable; it says nothing about having to do anything beyond that.

As you and I agree, that method will not detect incorrect polarity in some situations (specifically an L-E reverse) - but it is surely the IEE, not me, that you should be telling that they don't understand the concept of polarity?

Kind Regards, John.

 

[Electrifying]

Yes, but GN3 is just that - guidance on how to achieve the correct test results required by the regulations.......you can use other methods.

For instance - testing Zs. GN3 gives you a method of testing Zs, but that's not the only method........you can test Zs at the consumer unit - and it doesn't trip the RCD.

GN3 telling you to use the R1 + R2 method to confirm polarity on a radial circuit is a bit misleading - it only works for lighting......you throw the switch and you get a reading- polarity confirmed.

Do the same on a socket outlet and the reading will switch in and out, regardless of which conductor is connected to the 'line' terminal.......so it doesn't confirm polarity......as you rightly state, a 'Line' / 'CPC' reversal.

As I stated, the only way to then confirm 'polarity' of that socket is, after testing R1 + R2, connect the 'Line' and 'Neutral' conductors at the CU and test R1 + Rn at the socket as well.

If you get a reading for both tests, then polarity is confirmed.

This is stuff that comes from understanding what you are testing for, rather than reading from a book - which isn't a dig at you, by the way, as you aren't a practising spark.

You could do the other suggestion and drag a wander lead around with you........but if it's a three storey house with fifteen socket outlets it would be a bit of a drag! (and you have to do it twice if you want to confirm no neutral / cpc swap, remember )

To be honest, this discussion - and what happened to the plumber in the OP - has me wondering how many people do correctly test polarity at all socket outlets before ticking that box.

 

[JohnW2]

Yes, but GN3 is just that - guidance on how to achieve the correct test results required by the regulations.......you can use other methods.

Yes, but I thought we've agreed that following that guidance would not necessarily 'achieve the correct test results required by the regulations'.

GN3 telling you to use the R1 + R2 method to confirm polarity on a radial circuit is a bit misleading - it only works for lighting......you throw the switch and you get a reading- polarity confirmed.

It's surely more than 'a bit misleading'. It says nothing about being restricted to lighting circuits and nothing about throwing any switches. The (unstated) implication is that a reasonable (R1+R2) constitutes an adequate demonstration of correct polarity.

As I stated, the only way to then confirm 'polarity' of that socket is, after testing R1 + R2, connect the 'Line' and 'Neutral' conductors at the CU and test R1 + Rn at the socket as well.

I've already agreed with that (assuming that one is using that method) - but I've also questioned whether (m)any people actually do it. As I've said, if I wanted to undertake a foolproof dead test of polarity, I'd use a long lead.

This is stuff that comes from understanding what you are testing for, rather than reading from a book - which isn't a dig at you, by the way, as you aren't a practising spark.

I couldn't agree more. Under most circumstances, I don't really care one iota what regulations and guidelines may say about methods of testing - or, indeed, what is tested. It's usually the life, limb and home of myself and those around me that is at stake, and I will undertake tests that I feel, on the basis of my knowledge understanding, are necessary in order to protect those things.
However, whilst I'm well-known for being critical of those who 'work by books, regulations and guidleines' rather than on the basis of knowledge and understanding (or even without having adequate knowledge and understanding), I don't think one can (or should be able to) really blame anyone who dutifully follows IEE guidelines to the word for believing that they have done 'what is required'.

For instance - testing Zs. GN3 gives you a method of testing Zs, but that's not the only method........you can test Zs at the consumer unit - and it doesn't trip the RCD.

GN3 actually allows that Zs of final circuits can be calculated by adding (R1+R2) to 'Ze' - which can either be 'Zs' as measured at the consumer unit (which many would call ZDB), or can even be determined by enquiry (for TN systems).

Kind Regards, John

 

[Electrifying]

For instance - testing Zs. GN3 gives you a method of testing Zs, but that's not the only method........you can test Zs at the consumer unit - and it doesn't trip the RCD.

GN3 actually allows that Zs of final circuits can be calculated by adding (R1+R2) to 'Ze' - which can either be 'Zs' as measured at the consumer unit (which many would call ZDB), or can even be determined by enquiry (for TN systems).

I was referring to an alternative method of testing Zs of a final circuit - at the CU as opposed to testing at the furtherest point - saves balancing on top of a ladder and trying to fix three leads to your ceiling rose to test non-trip............but you won't find the method described in GN3

 

[JohnW2]

You slipped this bit in whilst I was replying .....

You could do the other suggestion and drag a wander lead around with you........but if it's a three storey house with fifteen socket outlets it would be a bit of a drag! (and you have to do it twice if you want to confirm no neutral / cpc swap, remember )

True, but 'your method' would not involve any less wandering around the house/sockets, would it? - as I see it, the only difference would be that you wouldn't be trailing a test lead behind you.

To be honest, this discussion - and what happened to the plumber in the OP - has me wondering how many people do correctly test polarity at all socket outlets before ticking that box.

I think we both probably know the common answer to that. I don't really want to wander in the territory of things that we don't discuss here, but provided one is going to undertake live tests (carefully) immediately after energisation of the circuit ...............

The tragedy which started this thread would probably not have happened if I, or many others here, had done the wiring, even if we had managed to reverse the L & CPC, since we would have had a nice bit of G/Y sheathed copper between socket (connected to the other G/Y-sleeved conductors, even if they were in the L terminal) and backbox - so that, assuming the faceplate screws were in, there should have been a tell-tale bang and a click when we first attempted to energise the circuit!

Kind Regards, John.

 

[JohnW2]

I was referring to an alternative method of testing Zs of a final circuit - at the CU as opposed to testing at the furtherest point - saves balancing on top of a ladder and trying to fix three leads to your ceiling rose to test non-trip............but you won't find the method described in GN3

Yes, I presumed that's what you meant (it's easy enough so long as you join conductors appropriately at the ceiling rose and then measure at the CU such that the fault current goes through the (R1+R2) of the final circuit) - but I was pointing out that GN3 allows you not to bother directly measuring Zs of a final circuit at all (or necessarily measuring any EFLI if you have a TN system!)

Kind Regards, John.

 

[Electrifying]

I was referring to an alternative method of testing Zs of a final circuit - at the CU as opposed to testing at the furtherest point - saves balancing on top of a ladder and trying to fix three leads to your ceiling rose to test non-trip............but you won't find the method described in GN3

Yes, I presumed that's what you meant (it's easy enough so long as you join conductors appropriately at the ceiling rose and then measure at the CU such that the fault current goes through the (R1+R2) of the final circuit) - but I was pointing out that GN3 allows you not to bother directly measuring Zs of a final circuit at all (or necessarily measuring any EFLI if you have a TN system!)

Kind Regards, John.

Quite, but I, as an electrician, am oposed to some scaremongering H&S do-gooder, who has probably never tested anything in his life, telling us we don't have to test it, we can make the figures up. I would rather test it.

You see, the one big flaw in adding R1 + R2 to a Ze figure gained by enquiry, is that you haven't confirmed that you do actually have an earth - let alone an earth with the loop impedance that they quote you over the phone.

As you can't get around live testing for 'supply polarity' or RCD/RCBO functional tests, I don't see what the issue is with testing Zs.

 

[JohnW2]

[
Quite, but I, as an electrician, am oposed to some scaremongering H&S do-gooder, who has probably never tested anything in his life, telling us we don't have to test it, we can make the figures up. I would rather test it.
You see, the one big flaw in adding R1 + R2 to a Ze figure gained by enquiry, is that you haven't confirmed that you do actually have an earth - let alone an earth with the loop impedance that they quote you over the phone.

I couldn't agree more - although I would have thought that a 'scaremongering H&S do-gooder' would be the last person to be responsible for saying that you didn't have to do it! After alll, the regs and guidelines do 'allow' you to test it.

Having lived most of my life with TT systems, I've come to regard Zs measurement as a bit of a joke for me - since what I'm actually measuring is primarily the impedance of parallel paths which theoretically may not still be there when I wake up in the morning (if the man with a lorry, a large drum of plastic pipe and the ability to work very quietly appears sometime during the night ) So long as he doesn't turn up, I could actually achieve all the required disconnection times without any RCDs.

Kind Regards, John.