'high security' ring main

Lack of medication?

But don't worry, he will soon say that it was a spelling mistake and that you are picking on him.

I take it you don't understand then.

More a case of it making it harder to fault find later, plus a slight increase in the chances that you will break a wire by work hardening it.

Which would be preferable to the alternate arrangement where it degenerates into a section of a radial circuit being unearthed altogether and the user not knowing.

One would test just as one would for any other new circuit.

It addresses a particular risk of circuits with high earth leakage - it does not eliminate the risk completely - something that can never be done. It does however reduce it for a marginal extra cost in materials and effort. As with all engineering its a case of compromise.

its marginal at best..

suppose its a simple circuit.. one socket two wires between the socket and the consumer unit.

that is twice as many joints terminated as a single wire circuit. so there is twice the rate of failure. So that means the dual circuit is not twice as reliable as the single circuit and the first fault will in the dual circuit will (on average)occur in half the time the first fault will occur on the single circuit. The second will occur somewhere near the time the first would occur on the single circuit (being as it has half the failure rate due to having half the components). This is true for any replicated system where the faults are undetected and not fixed.

There are a load of equations that tell you how reliable a system is, but they all assume you can detect the fault and hence repair it, which you can't easily do in this case.

You really have no idea do you?

Ok, if that's the official view. Mine - having come across lots and lots of poor connections on installations done by pros due to poor tightening etc, is that I'll continue to twist conductors together here.

No its not.

Not much point in doing it for a one socket circuit, you are not going to amass enough leakage on a single socket to warrant high integrity earthing anyway.

There are not "twice as many joints", there would be four - live, neutral, earth, earth loop - so one extra. The one extra joint now means you need two joint failures to leave the socket unearthed rather than one.

And in English this time?

I think you are missing the point here. The object of the exercise is to reduce the chances of leaving a section of circuit CPC floating with the potential to source a significant current at mains voltage due to leakage. Ensuring that each socket has two parallel earth connections in no way hinders this. Using two terminals also halves the chance that a single termination failure would result in this.

But you have twice as many joints in the earth so twice the failure rate, no way to know its failed and the second fault will leave it in exactly the same state as the first failure in a normal circuit.

Where is the high integrity in a system where the faults occur at the same rate and have the same effect?

You are missing the point. read up about high reliability systems and failure rates versus time to repair and it will all become clear. I am not going to attempt to explain it any further than I already have as its pointless trying. You can either read about it or ignore me, take your pick.

I think here is one of the false assumptions upon which your argument flounders...

Firstly you are assuming that a "normal" circuit would be wired with a single earth connection at each socket. Whereas, most electricians I have observed, will use both terminals (myself included) when two terminals are provided. The reason knocks on to the second part of your assumption; that two connections are of the same inherent reliability as one, and hence the benefit is doubled along with the probability of failure and hence you have a zero sum game. This is not the case IME. When dealing with loose sleeved wires such as the earth, it is inherently easier to make high quality terminations of one wire per terminal than two. There is also less possibility of a shift in the wire dressing will alter the clamping pressure in the terminal.

You are also ignoring the benefit provided by better earth level clamping provided by having at least twice (and possibly) more CSA of copper clamping the leakage potential on the CPC.

None, however that is not what we are dealing with here.

Also note that in a commercial environment where use of this circuit topology is more likely, the fact that you have even identified the circuit as a high risk one, suggests a more rigorous routine testing would be appropriate than for a general purpose circuit. This would reduce the likelihood of a fault remaining undetected.

No need to "read up" on it Dennis - I seem to recall "testing and maintenance" and "reliability engineering" where not particularly difficult courses when I studied them.

I don't believe a word of it Dennis - you like nothing better than an argument descending into ever decreasing circles.

Oh tempting.... could I use a grubbing mattock rather than a pick?

Your experience is irrelevant. Try proving that two connections are as reliable as one. You can't as it just isn't true.

So you are now saying they are going to disassemble the circuit and test it and then reassemble it frequently so that dormant faults are found, without actually making it less reliable by all the messing about? You appear to gloss over the fact that its difficult to test the circuit once its complete in the first place. just buzzing a few wires isn't going to detect all possible faults and taking it apart and reassembling it may well induce the same faults when you do it.

However you didn't understand them it appears, assuming you did such a course.

After the intitial inspection the circuit can easily be tested at any time without any disassembly in a matter of minutes.

I can satisfy myself that after many hundreds of inspected joints that this is the case. You can of course suit yourself.

With wiring in general, some failure modes are more common than others. Loose sleeved CPCs tend to be more problematical when there are several in a screw terminal.

Remember also that the IET armed with the expertise of many competent engineers, and access to far more empirical and experimental data that either of us, also promote this system. It is not something I invented just to wind you up.

In your opinion, which you are most welcome to.

No. I am saying in a commercial environment it is appropriate to carry out periodic testing. This does not necessarily need to be invasive, but can be conducted with an earth loop tester and some inspection.

You do write some crap... Its hardly any harder to test than a bog standard radial, and certainly no harder than a ring circuit - and testing that is routine.

Have a read of the on site guide - it explains in detail how to test basic circuits. It really is not rocket science.

Got a pass with distinction in both old bean...

You can if you want but i work with real numbers not a few hundred. If you really think you can find the failure rate of such simple circuits from inspecting such low numbers then you really don't have a clue.

The IET have strange views as many of their members will tell you.

I would like to see you test for the faults without taking stuff apart with the kit electricians use. that is the average electrician wouldn't have a clue or the equipment needed to test the circuit with dismantling it and prove it was actually working as intended after assembly. That would mean as the user sees the equipment BTW.

Really? That means they aren't worth the paper they are written on.

Says the man who can count up to twelve using his fingers.

So says the bloke who is unable to use an induction hob without causing a fire.

It would take 5 seconds to prove that both earth circiuts are intact.,

So says the bloke that can't pass water without needing a change of underwear.

I'd like to know how you do that without disconnecting anything, or using test gear I'd only expect to find in a lab.

We live and learn.

A simple earth loop reading will do the trick.

Which, according to one set of testing instructions I've seen involves making and breaking a number of connections.

Are you are thinking about an earth loop reading for the origin of the supply where you would need to disconnect main equipotential cables?