Neutral earth fault

[derek preece]

Just to make clear, incase anyone reading this didn't know, a Martindale socket tester will not pick up a neutral/earth crossed connection which is why the correct dead tests are so important.

 

[Guest 004]

Just to make clear, incase anyone reading this didn't know, a Martindale socket tester will not pick up a neutral/earth crossed connection which is why the correct dead tests are so important.

No but doing a loop test will

 

[derek preece]

I always carry out the sequence of tests in the order I was taught at college, I do a Ze test before starting the design for a new installation, all the rest of my live testing is done after all the dead testing has been done. I only do new installations though so that is why I don't rely on Zs testing, I just use calculation.

 

[Guest 004]

I always carry out the sequence of tests in the order I was taught at college, I do a Ze test before starting the design for a new installation, all the rest of my live testing is done after all the dead testing has been done. I only do new installations though so that is why I don't rely on Zs testing, I just use calculation.

I was never 'taught' to do R1Rn
Indeed I am struggling to find any reference for this

Certainly not in NICEIC Inspection, Testing and Certification, not in BS7671
If it is BS7671, I would like to know where

 

[BigTom83]

You don't record r1 + rn, but you do do it. Look in on site guide, how to test a ring. Tells you step by step.

 

[Guest 004]

You don't record r1 + rn, but you do do it. Look in on site guide, how to test a ring. Tells you step by step.

cannot see R1Rn test

Please use correct R (not r) as it can and will confuse people

OK have seen it in OSG

 

[BigTom83]

If you have got red on site guide, bottom of page 81. Continuity of ring final circuits, step 2.

 

[Guest 004]

If you have got red on site guide, bottom of page 81. Continuity of ring final circuits, step 2.

Indeed Thank You, have edited previous post

 

[derek preece]

There is also, on pg 38 of GN3, an explanation of (r1+rn)/4

 

[Guitarist]

There is also, on pg 38 of GN3, an explanation of (r1+rn)/4

Absolutely. All that said, a thorough R1R2 will pick up cpc in the wrong terminal anyway.

 

[brman]

Absolutely. All that said, a thorough R1R2 will pick up cpc in the wrong terminal anyway.

Exactly! Isn't this the point? Even if an R1+Rn wasn't done (which it should be, according to the guidelines) a R1+R2 would still pick up this fault.
And two modular sockets are two separate sockets so both need to be tested. An easy to mistake to make but an error in the testing all the same.......

 

[Guitarist]

Exactly! Isn't this the point? Even if an R1+Rn wasn't done (which it should be, according to the guidelines) a R1+R2 would still pick up this fault.
And two modular sockets are two separate sockets so both need to be tested. An easy to mistake to make but an error in the testing all the same.......

Yep. Let alone the fact that it was connected wrongly (bad enough on its own), the guy who tested the installation didn't do it properly. Simple.

 

[Deleted member 26818]

Well as I've never done the 2391, I can only go by my own training and experience.
The r1+rn/R1+Rn tests are listed in GN3 as being step 2 in the testing of an RFC.
However this is often not conducted (especially with new builds) in it's entirety with only the r1+rn part being measured to prove continuity.
With any initial verification, 100% of the installation should be tested, and to my mind this is imperetive when the testing is of a third parties' construction especially as it is likey in this case that there are a number of third parties conducting the construction.
It's all very well taking short cuts when testing your own work, but not when testing other's workand especially when there are likely to be many 'other's'.
With normal double sockets, it is usual to only measure R1+R2 at one of the sockets, so it could be the case that the person conducting the dead tests was not aware that these particular sockets were wired individually, or may be that they knew, but did not consider it necessary to test each individual socket.
This suggests that either the person was inexperienced. Perhaps newly qualified or even not qualified at all.
It may be that the testing is being conducted on a fixed price and the managers/supervisors were pushing for the testing to be conducted quickly, perhaps the person was even instructed not to test each socket?

To my mind, each of the tests listed in GN3 should have been performed in order, and the test and inspection should cover 100% of the installation.
However if the tester has been given an instruction not to conduct certain tests, or not to test 100% of the installation, what can they do?
At the end of the day, they are an employee, and must as such consider that they will be sacked if they don't follow their employers' instructions.

 

[Guest 004]

Nice one Spin

So are we agreed, that an R1R2, would have picked up the fault, and despite what some books say, R1Rn was not needed? (going away a little from the OP).

 

[IzzyS]

r1+r2 would have picked it up in this case, but without the r1+rn you could miss a line-cpc reversal. Which of course would be detected at the loop test stage, but that's not really the way. Especially with fancy metal sockets... so r1+rn is essential in my opinion.

 

[Deleted member 26818]

Nice one Spin

So are we agreed, that an R1R2, would have picked up the fault, and despite what some books say, R1Rn was not needed? (going away a little from the OP).

Yes, I'd agree with that.

 

[bigbaddave]

Would a earth / nutural cross over work on a rewirable fuseboard?

 

[Nivag316]

Visual inspection before testing would help as per inspection schedule. Makes yo wonder how many other suspect installs are out there.

- - - Updated - - -

Visual inspection before testing would help as per inspection schedule. Makes yo wonder how many other suspect installs are out there.