Earthing neutral

I don't see the need to isolate the PE between the transformers and gear- I would earth them at both sources and the gear without hesitation.

That wire (between the sources) is logically the system Neutral (even though it's not distributed to the loads in this case) - which for a TN-S system can only be earthed at one point (otherwise it becomes a TN-C-S system and you have a whole pile of other potential problems to deal with).


It probably helps to read it along side the preceding diagram (9A in appendix 9 of BS 7671) where the N is distributed to the loads - to which the same notes apply.


Also it's not the case that the regulations are saying that this conductor must be treated as a PEN. The text is simply describing it as being similar to a PEN (and as it's an earthed neutral conductor, it is does have some similarities). Also note that this text is only a note to a diagram in an informative appendix to the regulations - it's important to understand the it is does not form part of the regulations proper and should not be taken as such - appendices that are part of the regulations are marked "Normative" instead of "Informative".


   - Andy.

Ok- but what is the theory behind keeping a PE insulated? I know I keep asking because to be frank those behind BS7671 know WAAYYYY more than  I do.

Aah - why do we rush round sleeving bare earth wires, after marking them as inspection failures and have the earth as an insulated core in flex, when it could just as easily be the conduit or metal trunking that is performing the same function, and we do not have to insulate that.

Nothing to do with neutrals this time, but the rise of voltage during fault is still relevant, as is the double fault to danger principle. In a mains flex 

Long bare cpc wires inside accessory and junction boxes have a habit of getting pushed against a line conductor or terminal when the box is pushed closed.

ProMbrooke:

Ok- but what is the theory behind keeping a PE insulated? I know I keep asking because to be frank those behind BS7671 know WAAYYYY more than  I do.

1 - To limit the impression of the HV side fault appearing on the LV Neutral - ie the Rise of Earth Potential transfer voltage


2 - To ensure it is only earthed at one point - if it were bare (and buried) then it  would effectively be multiple earthed


Regards


OMS

Sorry IT woes mean this is now out of order.


Aah - why do we rush round sleeving bare earth wires, after marking them as inspection failures and have the earth as an insulated core in flex, when it could just as easily be the conduit or metal trunking that is performing the same function, and we do not have to insulate that.

Nothing to do with neutrals this time, but the rise of voltage during fault is still relevant, as is the double fault to danger principle.

A mains flex or the final circuit wiring  is likely to be the bit with most of the voltage drop during an appliance fault - if L-E voltage is 230V before the fault then at the point of junction it may be 115 volts - with a voltage gradient sloping down the live from 230 to 115, and at the same time an identical voltage gradient slopes up the earth conductor. so to stop other things being 'tingly' for the short time between fault on and ADS kicking it off, we prefer to cover them. This assumes the PE and L are comparable resistance.

But in conduit and trunking, the impedance of the earth path is vastly less than that of the live core, due to the greater metal cross section, so the voltage division and exposed touch voltage issue is less serious.

In places with a TT supply, the earth conductors rise to closer to 230V during fault, so firstly the ADS has to be faster and secondly those of us who think about it try to specify plastic conduit, or if we have to have lots of bare metal earthed stuff, then try to make sure that parts that are within reach of each other are either bonded to each other or are arranged so folk cannot get to it.

(and so the same earth, so it all goes live together, and you do not care - aster all the only time you really would like your kitchen sink and taps to be live not earthed,  is when trying to fill a live handled kettle ... though insulation is better still.)


At the substation end, it is the same sort of thinking but in reverse - can a substation transformer fault cause the voltage on exposed stuff to rise until the ADS gets it, be that HV breakers or expulsion fuses, as mentioned above.

To ensure it is only earthed at one point - if it were bare (and buried) then it  would effectively be multiple earthed

Agreed - even if not buried a bare conductor might be subject to accidental or incidental earthing.


Remember that this conductor is the N of the system (as well as providing the reference to Earth) - and multiply earthing neutrals can be a bad idea. Generally we attempt to keep things safe - at least under single fault conditions (either insulation failure or conductor continuity failure) - and a multiply earthed neutral can be a problem in that respect. Think of a N conductor earthed to separate electrodes at two different points, and then a break in the N conductor somewhere between the two - any current still flowing through the N will cause a significant voltage gradient around each electrode - so not only is the system itself no longer entirely referenced to earth, but you have a likely lethal voltage between the two electrodes and whatever is connected to each of them - not good for a single simple fault. That's one reason why we don't normally have PEN conductors in consumer's installations and in DNO systems where they are used they go to extraordinary lengths (well beyond normal BS 7671 requirements) to keep PEN conductors intact. With N fully insulated and Earthed at just one point any single fault shouldn't be dangerous - bits of the system might well stop working and (insulated) bit of the N conductor may well be a long way from 0V - but it should still be safe.


The problem is more obvious where the N is distributed and so expected to carry large currents from the loads - but even in this case where it isn't the N could still be carrying some circulating currents between the sources (imagine one source trying to produce 229.99V L-N and the other 230.01 volts) - so you can't rule out the possibility of N carrying a few amps between the sources - so the same approach applies.


That's not to say you couldn't use a multiply-earthed (TN-C-S) system if you really wanted - but then it wouldn't be a TN-S system and that particular diagram wouldn't apply.


   - Andy.

Again, I see no harm in multiple earthing of a PE.

Because it's not just a PE in that diagram - it's also the system N - it ties the neutral points of both sources togther - and multiple earthing of a N can be a problem.


   - Andy.