Fatal Incident - Safe Isolation failure?

 

I suspect that the DIY-minded general public get away with quite a lot, more by luck than judgement.

More by luck and circumstances sometimes. I carried out an I&T today on a holiday chalet which is changing ownership. There is evidence of D.I.Y. meddling. A three bar infrared wall heater showed no earthing. It is a class 1 type. It is fed from a 13 Amp. socket hard wired to a switched  fused connection unit with a neon indicator. All good so far. But its supply circuit's C.P.C. was left disconnected behind the switched fused connection unit.

If the case became live, a safety aspect is that the chalet floor is suspended wood. The floor coverings are plastic and nylon and there are no extraneous conductive parts in the room. 

Z.

Zoomup: 
 

Mike's comments remind me of an incident many years ago when wound ballasted fluorescents were popular.

A batten fluorescent fitting in a domestic garage was misbehaving. I was up a metal  step ladder having an initial look and fiddle. I received a mild but unpleasant shock.

The terminal block inside had three terminals. L. E. and N.  E. was connected via a small metal tab to the metal batten body and was designed to be the earth terminal.

A clown had connected the E. terminal to the permanently live loop conductors. Earth (C.P.C.) was not connected at all.

I was saved by the fact that the steps had plastic feet on them.

 

Z.

 

 

I have found the same issue with fluorescent lights several times. 

You could take a lead from SSE jointers. They were blue neoprene examination gloves when working, as they will protect from non-piercing direct contact, such as brushing a conductor or slipping when using insulating tools. I have some extra tough orange neoprene ones which might be even better, as seen on certain car maintenance videos, being made from synthetic rubber. Any additional safety feature is a good idea and these gloves still give excellent dexterity, unlike the gum-rubber insulating gloves for fuse changing live,  etc. I agree with John too, the volt-stick pens are much better than a two-pole tester, they work on the electric field between true Earth and live parts, check before and after use on a known live cable. Although many people don't seem to like them, the neon screwdriver works nearly as well as a volt stick, and is still VDE certified as safe to use. However, I can offer no guarantee of 100% reliability of any to these things, but that is not really what you are wanting is it?  

The two pole tester beats the non-contact volt stick for proving dead.

When I was taught safe isolation twenty years ago we were taught to prove live before proving dead, so:

• Prove the circuit LIVE using the two pole tester, checking L-N, L-E and N-E to ensure correct polarity, also the two pole tester works correctly
• Isolate the circuit and secure it.
• Prove the circuit DEAD using the two pole tester, checking L-N, L-E and N-E.
• Prove the tester still works using a known live supply, such as the incoming terminals of the isolation switch or a proving unit.

That way there’s less risk of testing between two conductors or terminals both with the same voltage on them or between a live and a disconnected conductor and not identifying an issue. 

The “up to date” method omits proving live and goes straight to isolation then proving dead, to reduce the risk of  electric shock, but it means the circuit hasn’t been proved healthy to start off with, increasing the risk of not identifying problems.

When I was taught safe isolation twenty years ago we were taught to prove live before proving dead, so:

• Prove the circuit LIVE using the two pole tester, checking L-N, L-E and N-E to ensure correct polarity, also the two pole tester works correctly
• Isolate the circuit and secure it.
• Prove the circuit DEAD using the two pole tester, checking L-N, L-E and N-E.
• Prove the tester still works using a known live supply, such as the incoming terminals of the isolation switch or a proving unit.

That way there’s less risk of testing between two conductors or terminals both with the same voltage on them or between a live and a disconnected conductor and not identifying an issue. 

Consider the theoretical situation where the isolating device somehow disconnected N and PE but left L connected…

(there are a few devices that deliberately switch PE, from some in-line RCDs to PME_style EV controllers to islandable on-site generation, so perhaps not an entirely implausible situation, let alone the complete idiot-wired setups where they get three wires into three terminals but not necessarily in the right order….)

   - Andy.

“Consider the theoretical situation where the isolating device somehow disconnected N and PE but left L connected…”

I had a job that I posted a topic about a few years ago, some electricians had installed storage batteries to a TT earthed installation with solar panels, a hot tub and a ham radio station, nothing like keeping it simple, after they had installed the batteries RCDs kept tripping, which was what I was trying to sort out.

Completely unrelate to the actual problem I discovered they had installed a four pole isolation switch which was supposed to turn off the mains and on-site supplies simultaneously, but they had actually miswired  it  so that the live conductor to the main house was disconnected, but the neutral became live Including the bare metal neutral bar in the consumer unit. So if you used this upfront isolation switch the consumer unit rather than the main switch within the consumer unit the CU became a very dangerous  place to be working. This could be detected using a simple plug in socket tester, voltage testing device or an installation tester, but you do have to be checking.

Treating every electrical installation as if it was installed by an idiot is the starting point for safe isolation, the second being treat everything as if it is live and you need to wave the volt stick about a lot, but don’t rely solely on the volt stick.

After watching the safe isolation on YouTube I mentioned above, but could not find I bought a Martindale VT7 single pole contact tester, bearing in mind my Fluke T150 does a single pole contact test as well so now I have two ways of doing this test, the Martindale VT7 should indicate if a metal enclosure or similar is live, but I don’t have enough experience of using it to make a recommendation.