Lighting fault

[tommy.dale]

H guys I wonder if anyone could shed some light on a fault that I have just attended, I fitted some fire rated led dim able lights in a pub, about 60 of them.
there was about 6 led downlights previously fitted that was working (fitted before I took job on) and was working since I done my bits upto 3 weeks ago, anyway i have gone back to do some more work and these 6 lights aren't working.....

my fluke t51000 isn't showing any voltage, I seperate the t&e fromthe lights and find the incoming feed from board which I get 230v on, perfect right?
i add the links in to daisy chain the lights and the voltage disappears completely yet I separate the t and e and voltage is there again,

when I attach the feed by itself to one of the fittings it comes on but not when any other Load is added.
there was also 2 fittings with the n-CPC reversed yet I saw these working??

anyone experienced this and could help me out before I rip everything out tomorrow?

tom

 

[tommy.dale]

I would guess a IR test was missed due to the CPC's & neutrals cross connected at some of the fittings !

i didn't install these fittings that is why I locked it off ready for tomorrow and will rectify and carry out the appropriate tests then, thanks for the feedback

 

[tony mc]

I realised that Tommy from post 9 as the 6 faulted lights had nothing to do with the 60 you had fitted! You only have to get the other 4 out as you have already removed 2 finding the CPC Neutral faults. It is obvious that no testing was carried out with regards these lights and I would document my findings on the new certificate backed up with photos of the faults that you trace when you submit your invoice! Let us know what you find !

 

[i=p/u]

I know it's a Tn-c-s by just reading this.

 

[Leckie]

R1, R2 and Rn tests first, then IR tests

 

[Leckie]

Oh, and disconnect l,n and cpc for continuity tests. Reconnect cpc only for IR tests

 

[GLENNSPARK]

Oh, and disconnect l,n and cpc for continuity tests. Reconnect cpc only for IR tests

what you on about now here eh?
use the wander lead, connect at the origin of supply and start going from point to point watching the values as you go....when its shot up you know the high resistance joint was at the last point tested...

 

[Leckie]

what you on about now here eh?
use the wander lead, connect at the origin of supply and start going from point to point watching the values as you go....when its shot up you know the high resistance joint was at the last point tested...

That might work, or might be misleading. It might be a bad connection to L or N. it might be a low IR between L or N or E. so verification of R1 R2 and Rn first, with any parallel paths disconnected as far as possible, then IR tests, but with the cpc connected to include any parallel paths.

 

[telectrix]

think what glenn was suggesting was to wander lead each conductor in turn, L,N, (& E) to find the point of high resistance.

 

[Leckie]

I agree with the method, I'm just saying that if its an IR problem that method might not indicate the fault and also that if it does iR test anyway after continuity test as it doesn't sound like this install has been tested in the first place.