Danger Earth Fault

[ebee]

Disregarding the presence of an RCD , I might not be present or it might not be working.
We go to great lengths to ensure disconnection of Earth Faults.
Not just in normal dry places but particularly in special locations such as bathrooms.
Indeed, not long ago, it was permitted to have a lighting circuit in a bathroom and the target time was within 5.0 to disconnection.
Then that changed to any lighting circuit serving a bathroom etc must be 0.4 seconds.
Nowadays all normal final circuits must be with 0.4 seconds or sometimes less.
Anyway, before disconnection occurs, a dangerous voltage potential could exist.
That might be solely dependant on conductor sizes ratios of bog standard T & E cable, unless fortuitous connection of another earthed conductor is present to effectively reduce the earth path. Such as another circuit containing a cpc sharing the earth path or bonding (main or supplementary) .
So, if we consider just a bog standard T & E cable supplying something in any location then what voltage might be present under earth fault prior to disconnection?
Seems like 16.0 T & E and 4.0 T & E might be the highest and 1.0 T & E might be the lowest.
Sobering thought?

Of course we are only considering TN systems and not TT

 

[JohnW2]

.... So, if we consider just a bog standard T & E cable supplying something in any location then what voltage might be present under earth fault prior to disconnection? Seems like 16.0 T & E and 4.0 T & E might be the highest and 1.0 T & E might be the lowest. Sobering thought? .... Of course we are only considering TN systems and not TT

As you illustrate,you aree perfectly capable of doing the (very simple) sums, but one problem is that the real-world answer depends upon factors external to the installation, which will usually often be unknown.

Presumably reflecting what you have done, IF the substation transformer (hence the neutral connection to true earth) were right next to the CU, and IF the supply voltage remained at 230V during the duration of the fault, then the voltage, relative to true earth, at the point of the fault would be roughly as per this table:

... which confirms, as you say, that the highest voltages at the fault, relative to true earth, arise with 4 mm² and 16 mm² T+E.

Assuming that the building is properly constituted as an equipotential zone, then what actually matters are potentials relative to the installation's earth (not 'true earth'). That means that, in the real-world situation (substation being remote from the installation) then IF the supply voltage at the CU remained at 230V during that fault (which it won't, since Ze will not be zero), then the figures in the above table would still apply to the voltage relative to the installation's earth - and that remains true whether it is a TN-S, TN-C-S or TT installation.

However, in the real world the catch is that (since Ze will not be zero) the supply voltage will not remain unchanged during the fault. Particularly if the installation is fairly distant from the transformer, I imagine that the supply voltage at the origin of the installation could drop markedly during the fault, thereby markedly reducing the voltages at the fault relative to the installation's earth.

However, none of this is 'news', so I wonder what your point is. I thought we all understood that 'touch voltages' remain 'dangerously high'; during a fault, that being the reason why we are required to take steps to limit the faults to very short durations (before disconnection)?

Kind Regards, John

 

[ebee]

Yes I know all that John, what was striking was the theoretical difference between the best and worst size relationships at the point of the fault . Besides I do prefer to think about 240 being the "real voltage" as it always has been - ish rather than the now notational 230 volts . Europe did not change their voltages and neither did we, we simply all rewrote it to fit in with each other.
Anyway, I always feels a bit safer with 1.0 T & E or singles of the same size conductors than some of the other sizes, even if somewhat deluded

I don`t think we will ever re-declare at 1000V +/- 100% as being nominal voltage for most things but it would save some politicians some considerable thinking

 

[ericmark]

With a fuse we see a graph disconnection time and overload current. So the true short circuit disconnection time will be variable, as will be the voltages, with an MCB the magnetic part will disconnect in such a short time likely ½ cycle, same with RCD, so in real terms only looking at fuses.

I find the fuse confusing, we had semi-conductor fuses on our heaters as the MCB was not fast enough to protect the semi-conductor contactor, so fuse was actually faster than an MCB.

But in the main we want disconnection before anyone and touch a live part, so looking more about damage to cables, semi-conductors etc. As until insulation melts we should not be able to touch a live part.

 

[JohnW2]

Yes I know all that John,

As I said, I realised that you were as capable of doing thee (very simple) sums as anyone elsee.

.... what was striking was the theoretical difference between the best and worst size relationships at the point of the fault .

Indeed. I have no idea how 'they' arrived at the decision as to what CSA of CPC to have in T+EE cables. I think that, in some countries, COCs have to be at least the same CSA as the live conductors - which makes a certain amount of sense.

However, I'm not sure it really has much practical relevance, since the CSA of CPCs would have to be at least double that of the live conductors before the touch voltages became remotely safe - and that'snot going to happen (nor, to my knowledge, is it the situation anywhere ion the world). ... and, of course, RCDs don't make any difference to (the magnitude of) the touch voltages, either.

Besides I do prefer to think about 240 being the "real voltage" as it always has been - ish rather than the now notational 230 volts .

So do I, but we are expected to think of 230V these days and, in any event, the issues we're talking about are essentially the same whether the supply voltage is 200, 230, 240, 250 or 300 volts - even if 'the numbers' vary a bit!

Anyway, I always feels a bit safer with 1.0 T & E or singles of the same size conductors than some of the other sizes, even if somewhat deluded

It's obviously 'the best' T+E, but I don't think you should feel significantly 'safer' because the touch voltage is (briefly) then 'only' 115V (or 120V,or whatever)

Kind Regards, John

 

[JohnW2]

With a fuse we see a graph disconnection time and overload current. So the true short circuit disconnection time will be variable, as will be the voltages, with an MCB the magnetic part will disconnect in such a short time likely ½ cycle, same with RCD, so in real terms only looking at fuses.

That's sort-of true,but less relevant in the context of wanting (being requires to have) disconnection times of 0.4s.

In fact, I imagine that at very high currents, a fuse may well operate more rapidly than an MCB. There is no theoretical limit to how quickly a fuse can 'vaporise' but, in the case of an MCB (which involves a 'mechanical switch') factors such as inertia and friction will impose a limit to how rapidly it can disconnect.

Kind Regards, John