Relative risks of TN-C-S and TT

[John D v2.0]

When i say touch voltage i mean the voltage of two simultaneously touchable items. Which would be around 2m apart.
Step voltage would be the voltage between two parts you would step on.
Interesting discussion here though.

 

[JohnW2]

When i say touch voltage i mean the voltage of two simultaneously touchable items. Which would be around 2m apart.

Fair enough - that's also what I have been talking about. However, it does seem that different people have different understandings of "touch voltage".

As I said, in terms of the PD between two touchable things (if that's what you are calling "touch voltage"), if all proper CPCs and bonding are in place then, contrary to what you (I think) suggested, that PD (which you are calling "touch voltage") will be lower with TT than TN, since the fault current through (and hence 'voltage drop' in) the CPC carrying the fault current will be lower with TT.

If you 'complicate' TT by introducing the possibility of bonded low-impedance extraneous-c-ps, then you would be more-or-less talking about the equivalent of TN-S.

Kind Regards, John

 

[John D v2.0]

See what you mean.

I'm not sure if it's a good question for this stage of the debate, but why don't we use isolated supplies? Is the whole system of earthed supplies to protect people directly or just to warn that equipment is in a poor state and needs replacing or repairing, before we end up with two different pieces of equipment perhaps in two different dwellings becoming faulty on different live conductors and causing a shock hazard.
Could we get the same protection by using fused RCD plugs in the same way we use normal fused plugs at present, and don't bother with earthing at all?

 

[JohnW2]

See what you mean.

I'm please to hear that you agree.

I'm not sure if it's a good question for this stage of the debate, but why don't we use isolated supplies? Is the whole system of earthed supplies to protect people directly or just to warn that equipment is in a poor state and needs replacing or repairing, before we end up with two different pieces of equipment perhaps in two different dwellings becoming faulty on different live conductors and causing a shock hazard. Could we get the same protection by using fused RCD plugs in the same way we use normal fused plugs at present, and don't bother with earthing at all?

In very circumscribed and carefully controlled environments isolated (not earth-referenced) supplies are obviously possible.

In the big wide world, I think at least one problem is that it would be impossible to prevent accidental connections arising between one or other side of the supply and true earth (e.g. a fault from L or N to what we would currently call an extraneous-c-p). The moment that has happened, the local supply network becomes 'earth-referenced', with the attendant risks to people with supplies on that network who have anything connected to earth in their premises. I think there are plenty of other reasons as well.

Kind Regards, John

 

[plugwash]

Floating supplies work great on a small scale but they don't scale up well.

Capacitance is everywhere, both deliberate filter capacitors and stray capacitance in cables. As the network grows you will reach a point where it is effectively connected to earth from an electric shock point of view even if there are no faults.

 

[John D v2.0]

The moment that has happened, the local supply network becomes 'earth-referenced',

I realise that but on its own it wouldn't cause a direct issue and with rcd protection to both items surely they would both disconnect as soon as there was an actual problem?
I guess what I'm really wondering is would universal RCD protection and no cpcs at all be safe

 

[ban-all-sheds]

IT installations become dangerous if you end up with two appliances with an exposed-conductive-part fault.

As for the rest - apologies, but I've been absent for a day, and right now don't have the will to catch up on all the back-and-forth, so apologies offered and ritual castigation (ooh, matron) accepted if this is superfluous or not what's being discussed:

Re touch voltage it is, AFAIAC, the voltage wrt earth to which an exposed-conductive-part rises in a fault condition. It could be wrt either true earth, or installation earth, probably the former, since the scenario is Mr Person clapping hold of the metal case of a toaster which an element has just contacted. (Other scenarios are available). If some dozy plumber has unnecessarily bonded a kitchen sink it could be installation earth.

We hope that there is some protective device which operates before the PD between the case and earth reaches a value which makes it a Bad Idea for him to have clapped hold of it. That's why we care about fault loop impedances. And why we care about a 1667Ω impedance wrt 30mA RCDs. Without an RCD, if we have, say, a 100Ω impedance, what voltage would need to appear on an earthed exposed-conductive-part for a 13A BS 1362 to operate? Or a 30A BS 3036?

 

[JohnW2]

I realise that but on its own it wouldn't cause a direct issue and with rcd protection to both items surely they would both disconnect as soon as there was an actual problem? I guess what I'm really wondering is would universal RCD protection and no cpcs at all be safe

Well, as BAS is probably implying, one problem situation (albeit very rare) would be if one item developed a fault from L to its exposed-c-ps and another (simultaneously touchable) one developed a fault from N to its exposed-c-p. If someone touched both, they would get an L-N shock, which would not trip an RCD (and certainly wouldn't cause an OPD to operate - at least, not until they had literally turned to a cinder, even if then!).

With earthing, as we know it, in the picture, an RCD and/or OPD would operate when the first of those faults arose (certainly if it were an L one), and the OPD would certainly respond to the second fault (since L & N would then be joined via CPC).

I think there are other issues as well (perhaps relating to less improbable situations), so I'm thinking

Kind Regards, John

 

[EFLImpudence]

However ....Because it is connected to true earth (at the transformer) through a path of (roughly) half the EFLI (at least with TN-C-S). All things being equal, with TN-C-S that would mean that the potential of the exposed-c-p would rise to about half of the supply voltage above true earth (i.e. nominally about 115V), not 230V.Maybe I should not have written "a lot less" but, as above, 115V is appreciably less than 230V.

Fair enough - but more than the 50V being mentioned and referred to as 'safe'.

My point was the contrast with TT, when the potential of the exposed-c-p will rise to something approaching '230V'.

Consequently, then, does it follow that you think it is a bad idea to TT outbuildings - and houses?

Notwithstanding -

What 'ground' - that in the vicinity of the earth rod? If so, it is 'bonded' by the Erathing Conductor and rod.

Ok. but only in the (very) close vicinity of the rod.
I presume the rod should be installed just outside the entrance of a shed - but is that how it is usually done?

 

[JohnW2]

Fair enough - but more than the 50V being mentioned and referred to as 'safe'.

Indeed - but, as I keep saying that 'unsafe' potential relative to truth earth is only actual 'unsafe' if either (a) one's equipotential zone is not as equipotential as it ought to be, or (b) if there is an adequate equipotential zone, but a person has part of their body inside (and able to touch an exposed-c-p), and part of their body outside, it.

Consequently, then, does it follow that you think it is a bad idea to TT outbuildings - and houses?

Not really - since, as you have just pointed out, even with TN the PD to true earth is more than high enough to be very dangerous IF either (a) or (b) above apply - so the fact that that PD is even greater with TT is really of no consequence (it was people other than me who raised that issue).

If (a) above applies, then there is not much to choose between TT and TN as far as safety/danger is involved - and, if one is concerned, one should make the zone more equipotential.

A judgement is necessary as to whether or not (b) applies and represents a degree of hazard that needs to be addressed. If one decides that it does need to be addressed, the better (lesser of two evils, but far from foolproof) approach is probably to have TT with the rod very close to the door (as below). For what it's worth, in my house there are no exposed-c-ps touchable by someone standing outside the kitchen or cellar doors - so I'd be happy to have that situation with TN (if I had TN!). However, there are such parts touchable from outside my greenhouse door - but that is TTd to a rod very close to the door.

Ok. but only in the (very) close vicinity of the rod. I presume the rod should be installed just outside the entrance of a shed - but is that how it is usually done?

As above, if (b) applies, then it is probably theoretically 'better' to have TT with the rod as close as possible to the door/window/whatever that could give access to an exposed-c-p within the house/building (as with my greenhouse).

Kind Regards, John

 

[John D v2.0]

If someone touched both, they would get an L-N shock, which would not trip an RCD

Surely both RCD would trip at that point? Unless you're talking about one appliance with two independent exposed conductive parts, in which case i agree. I would expect there to be a requirement to ensure all ecps to be connected together in a single appliance.

 

[JohnW2]

Surely both RCD would trip at that point? Unless you're talking about one appliance with two independent exposed conductive parts, in which case i agree.

I'm talking about two different appliances (or whatever), one with its exposed-c-p(s) connected (by a fault) to L and with the other having its exposed-c-p(s) connected (by a fault) to N. If both were protected by the same RCD, then someone touching both would get an L-N shock, with the L and N currents both flowing through that one RCD, hence no trip.

If each and every appliance/whatever had its own dedicated RCD (you did at one point talk about everything having 'RCD plugs') then two of them would indeed trip in the scenario I'm talking about.

You were talking about a hypothetical system with no CPCs (hence no earthing). I suppose a half-way house would be to have CPCs, bonding together all exposed-c-ps, but not connected to earth, but that would be a mixed blessing. Although it would mean that the second of the faults above would result in immediate operation of an OPD (since L and N would be connected via CPCs), whilst there was only the first fault, every exposed-c-p in the installation would be connected to either L or N (probably unknown go anyone) - which, as I mentioned at the start of this could present a hazard if/when one side of the supply became (accidentally or incidentally) earth-referenced (if there were some earthed (and not bonded) things existed in the building).

Kind Regards, John

 

[JohnW2]

If each and every appliance/whatever had its own dedicated RCD (you did at one point talk about everything having 'RCD plugs') then two of them would indeed trip in the scenario I'm talking about

.... I should perhaps have added that a downside of this is that no RCD would trip until someone got an electric shock (which, despite the RCD trip, could possibly be fatal). With earthing present, a fault from L to an exposed-c-p should trip an RCD as soon as the fault appears, usally before anyone has had a chance to get a shock.

Kind Regards, John

 

[John D v2.0]

I think that's the point, the rcd trip time is finite and more likely (you say definitely but i think it's still possible when there's no one touching the equipment) to be when someone has a shock. Although the existing tripping time for faults in hand held equipment is 0.4s, the tripping time would start when the fault occurred not when the user made contact with the equipment. Thereby making it less risky with the existing system.

 

[JohnW2]

I think that's the point, the rcd trip time is finite and more likely (you say definitely but i think it's still possible when there's no one touching the equipment) to be when someone has a shock.

Maybe I'm missing something but, in the scenario I described, I can see absolutely no reason why an RCD would/could trip (ever) unless/until someone (or something) touched exposed-c-ps of both faulty appliances. The fact that one exposed-c-p is at L potential and one is at N potential means nothing to an RCD until some current flows, and it can only flow if someone or something provides a conductive path between the two exposed-c-ps.
[we are, of course, talking about the scenario you hypothesised in which there were no CPCs]

Kind Regards, John