on ESC Website

[JohnW2]

The problem with keeping things floating that it`s difficult to confirm achievement 100% of the time, is some conductor is by design or accident referenced to earth and we are unaware of it then the safety we enjoy in a bathroom and a proper isolated SELV safety socket is easily defeated.
PS SELF in a bathroom would not be considered SELV at 50V either. ... Yup one side gets strapped to earth and your float idea disappears so back to square one and earthing/bonding.

I've been thinking ablout this (dangerous )

If one had a 2-wire, no-earth referenced (i.e.'floating') supply (hence 'non-polarised'), then it is hard to see that any sane person would deliberately ("by design") connect one of the live conductors to earth.

As for "by accident", that is obviously possible, just as it is that Line may accidentally be connected to earth withj our current (earth-referenced) system.

However, if that happened 'by accident', it would presumably be equally likely to happen to either of the two live conductors. If it happened more than once (within an installation or within a network), then it is presumably quite likely that both live conductors would (separately) have been connected to earth - with potetially 'interesting' consequences!

Kind Regards, John

 

[ericmark]

I seem to remember maybe 14th Edition, when we had to earth metal window frames, someone thought one could pass an extension lead out of the window, and a fault could make the frame live.

We had the same with other metal items, it was considered a dog could knock over a lamp, which smashed its bulb on hitting a radiator, and the pipes could make a radiator in another room live, and one could be unaware of the accident. So we bonded the pipe work.

However, this did not last long, and it was realised in some situations all being bonded to add to the risk, as well as removing risk, if I touch a live wire and an isolated radiator I will not complete a circuit, may get a shock due to inductive and capacitive linking, but not enough current will flow to likely do me any harm.

So we have to assess the risk, will connecting an outside lamp to the home's earth, when that earth may be TN-C-S reduce or increase the chance of getting a shock? Yes, loss of PEN is rare, but we can't ignore the risk, we must assess it.

So with SELV we have the same situation, the risk has to be assessed, and since we have not built the equipment, we are not in a position to assess the risk, someone has done that for us, if we see this sign then someone has type tested the item, and decided it is SELV, I picked up my laptops power supply, and it does not have that sign, ups. With a clover leaf plug, the output may be earthed, but am I safe to use the laptop plugged in on my BBQ table outside? And when PAT testing should I test if the 20 volt plug is connected to earth?

 

[ebee]

I seem to remember maybe 14th Edition, when we had to earth metal window frames, someone thought one could pass an extension lead out of the window, and a fault could make the frame live.

We had the same with other metal items, it was considered a dog could knock over a lamp, which smashed its bulb on hitting a radiator, and the pipes could make a radiator in another room live, and one could be unaware of the accident. So we bonded the pipe work.

My understanding that this was a myth brought about by some misunderstandings, I`ve seen a few myths buzzing about at the wholesalers over the years myself./

 

[ebee]

John,
it is surprising to some but in a two wire floating system it is possible for one conductor to be strapped to Earth and go unnoticed for a long time, when found it is either by coincidence or a big bang.
From your bolted fault to resistive leakage then capacitive and inductive coupling can all play a part, enough to consider deliberate earthing by some method or another , hence TN systems and TT being the favourite methods of protection chosen. TNC and IT are under pretty much more strict control measures and some of us (including me) are not really fans of the TNC part of TNC- S either .
If I were in charge of the UK I would mandate TNS as the norm and a few TT if you really do not have a realistic alternative , TNC-S not allowed at all and TNC & IT only under very specialist installations under very specialist control and therefore very rare.

Even then there is a very real risk that TNS could become very like to TNC-S and go unnoticed.

Even in my limited experience compared to the UK total housing stock I have come across two (OK not exactly related but some comparisons might be drawn though)
1/ Loss of N by the 4 core cable in the road, supplying quite a few properties on the three phase main, then one property caused a bolted fault making approx a third of those properties single phase supplies Zero volts and the other two thirds having 415 volts, Quite a few plugged in appliances did not survive this stress apparently.
2/ A meter being connected with L & N transposed, apparently it had been like that for some time (unlike item 1/ above) and so some changing a lamp etc with the room switch off the L supply was still connected to the lampholder pins as only the N was in fact switched, it all seemed to work correctly as far as the users could know and, as far as I could tell, nobody got a belt off it.
When I discussed it with the DNO they were reluctant to get it corrected within 24 hours until I pointed out the potential dangers then they corrected it within a few hours.

So yes, these things do happen and probably not just to me so may be quite a bit more than we would like to think they do.
Although there are not dead bodies littering the towns and cities there must be more close calls then we might like to think.
Sometimes you might get some trades mucking up the polarity of a circuit or two but when say a meter becomes wired reversed it can be pretty bad - A mistake is merely evidence that someone has tried to do something, has anyone never made a mistake?

But hey, I do not expect to be crowned King any time soon.

EDITED - The "Bolted Fault" I mentioned in 1/ above was not the usual E to L or even N to E but rather N to L, either a near dead short or an exceptionally high load of almost no impedance (resistance), thereby rendering that one phase N polarity as it came in and the other two phases 415V each away from it (and the usual 415V to each other) . Consequence , things with transformers in had overheated and some not survived it but some had survived.
Customers had been out shopping for a few hours and there was a distinct smell of overheating, the had some brand new TV stuff plugged in (Near Xmas so they`d been tuning in etc ready for the big day). I got called because the intruder alarm I installed years earlier stopped working.
Days/weeks previous a meter fitted had noticed some strange readings on a property and got a team to investigate, when the team went all seemed ok but when this overheat damage occurred in some properties they came back and traced the fault, fractured N conductor in the road so they dug up road to repair and discovered the bolted fault - turns out that someone in one property had just about shorted out one phase after the fracture by doing something "Abnormal". Once traced they cleared it and made the joint and filled the road back in.
No doubt some of those properties inhabitants had reported no electric so it was only a matter of hours not days but enough time to cause some damage in some properties, as far as I`m away only to appliances, no housefires etc.

They deduced that the fracture had been playing now you see me now you don`t as traffic caused ground strains so different meter readings by different teams until the fracture disconnected for good sort of thing.

The report give to the householders made interesting reading - it confused them and I must admit I read it thru then read it again to grasp the problem properly

All good fun and made more fun by someone doing something "abnormal". I`m guessing that without the abnormal bit the "N point" might have stayed closer to its normal potential as loads in the neighbourhood balanced out to some extent so the voltage swings not being as severe or of such long duration