Testing of lighting circuits

Question

I have been doing testing for low voltage installations and for the longest time I cringe whenever I start on the lighting circuits majorly for hotels where there is the lighting control systems because of the following challenges 
 1. CAT5/CAT6 cabling for control majorly for GRMS systems. 
 2. Circuitry where a single breaker serves 2/3 circuits looped from the main terminal of a switch to the next switch while only one cable seems to come from the breaker and makes the tracing of the circuits very difficult. 
 3. Borrowed neutrals and earths which gives very crazy results and sometimes the testing equipment is not able to give any analyzable readings. 
 How have you been able to do tests that conforms with the guidance of BS7671 and mostly is is required in the model form. and what are the possible tests to carry out in some of this circuits.

mapj1 · Accepted Answer

There really should not be borrowed neutrals between circuits on different breakers - that is a clear failure , though interlinked earthing is common, especially in buildings with metal parts in the construction, as any fittings that attach to the building metal will be connected. 
 Common places neutrals meet when they should not are on multi-way switches for lights on stairs that span lighting zones, certain types of emergency light fitting, and in domestic cases, wall lights that have taken the live from a light switch, where there is no neutral, and then a neutral from the back of the nearest socket... RCDs are slowly eliminating that last one. 
 The most important test in terms of shock protection is probably R2 to the light switches, and any other parts that folk will touch in normal operation - a failure to verify the earth to some inaccessible metal part is a lot less serious than a wall light or metal switch that s supposed to be earthed but is not. 
 I am unclear about your comment about one breaker serving multiple circuits - by definition one breaker serves one circuit - it is just that circuit has a convoluted layout, and may be fed from the middle or a branch point of the radial rather than one end - a 'Christmas tree' layout is not wrong as far as BS 7671, but I would agree it may be a pain to test as it has multiple ends. In such a case I would quickly test all possible 'ends' without dismantling, with an R2 wander lead and record the detail of the point with the highest or higher reading. 
 Insulation can be tested with L-N strapped relative to earth, and not bother with the L-N tests, as in an operational system, there will be loads wired L-N. There should not be much wired L-E or N-E. 
 See what others say as well. I do not do this sort of thing as my day job. 
 Mike

Testing of lighting circuits

Join us to get the best from IET EngX.

Community Rules & Guidelines