Shock protection on luminaries

I know that other standards sometimes have different ideas about shock protection to BS 7671 (e.g. BC and ES lampholders having bare live parts). Does anyone here know what the requirements for shock protection (when basic insulation fails) for domestic wall lights - specifically the type supplied with a flex, in-line switch and plug (i.e. expected to be in-reach). I'm not sure which product standard would apply. My natural reaction was to expect them to be either earthed or marked as double insulated - but are there any other options allowed by the product standard?

As you might guess, I've been asked to fit one and it doesn't look at all right to me - steel fixing bracket seems to be exposed to wiring with just basic insulation at the back of the fitting, and is then in direct metallic contact with the dome nuts holding it all together at the front.

I didn't much like the idea that it was sold as retail but supplies as a kit of parts - flex had plug and switch fitted, but supplied with bare ends to connect to the fitting - nasty if a little one got hold of it and tried to plug it in... but that's a different issue.

          - Andy.

Parents
  • Unfortunately, sleeving the wires isn't going to be particularly practical in this case - loose terminal blocks to deal with and the internal wiring goes through exposed metalwork that doesn't have enough free space to accommodate the extra thickness of the sleeving.

    So to make the best of a bad job, I've changed the flex to 3-core and managed to earth all the exposed metal parts. I don't like relying on RCDs alone for shock protection from single faults if it can be avoided. 

    Can anyone confirm if it's BS EN 60598-1 I should be looking at?

        - Andy.

Reply
  • Unfortunately, sleeving the wires isn't going to be particularly practical in this case - loose terminal blocks to deal with and the internal wiring goes through exposed metalwork that doesn't have enough free space to accommodate the extra thickness of the sleeving.

    So to make the best of a bad job, I've changed the flex to 3-core and managed to earth all the exposed metal parts. I don't like relying on RCDs alone for shock protection from single faults if it can be avoided. 

    Can anyone confirm if it's BS EN 60598-1 I should be looking at?

        - Andy.

Children
  • It sounds as though you are in "I don't like the look of that" territory Andy. I assume that the wiring inside the fitting is essentially stationary, so would require basic insulation failure and movement of the wires to touch any metalwork? Do you think this is even a reasonable possibility, and if so why? Do we get failure of PVC basic insulation when stationary, and is it extremely rare to never? How can the wires move, presumably there is a cord grip to keep them in place? Risk assessment of anything requires analysis of the highest level, including answers to questions like these. It also requires proper numerical values of possibilities, such as one in 50 million instances. The fully dumbed down ridiculous ones that are now used for everything are a meaningless abomination of the proper process. Reducing accidents to zero is similarly not possible. If one doesn't know the numbers, you are not suitable to carry out the assessment! These come from previous statistics of accidents, the only possible source of data. This is often so badly collected it too is meaningless. I would refer you to the one place where proper data is collected, this is postmortem examination. Pretty much every possibility is examined, and the results collated. Still there are some cases where no cause can be found, but we even have a statistic for the possibilities.

    The question you need to answer is simple, how many people are killed each year by this product (or very similar ones)? As few people overall are killed in the home by products and from electric shock, the answer is probably the one you are looking for, it is not significantly dangerous. Would the changes discussed make any difference, say from one every 100 years to one every 200 years? Note this will never be none every 200 years, the rate is unmeasurably low.

  • Andy said:

    "One of the other worries is where the wires to the lampholder run through a hinged elbow (obviously intended to be moved repeatedly in use) - they're just two coarsely stranded fairly thin but stiff single insulated wire, (probably 7-strand and about 0.5mm² at a guess) with a bit of braided sleeving over them."

    Z.