My drawing is for worst case N-E voltage.

TN-S or TT is the worst case,  TNC-S will have the Earthed reference closer to the consumer so will give lower N-E voltage.

be very clear with neutral -earth voltage you mean neutral to CPC voltage - clearly in PME land, that is zero at the company fuse,

or if you mean the voltage between neutral and terra-firma earth, as  present on accidental electrodes like outside metal fence posts and  so on - to which you may add an outside light or camera. That voltage rises steadily the further you are from the best earth electrode on the network, usually, but not always, lowest at the substation end.

With PME the installation earth terminal is not at zero even if there is a DNO earth rod connected directly into it as that rod has a resistance.

In reality there can be several installations connected and a fair length of DNO cable after the last DNO earth rod. 

The idea that the DNO TNCS-PME earth terminal supplied for the consumer installation is at zero volts just is not true. 

Yes the MET could be up to 18.5v relative to the star point earth back at the substation.

but I am referring to the potential you can measure at the origin of the installation i.e. from the Neutral to the MET. 

In a PME TNC-S the voltage drop on the neutral supply cable will appear as small potential differences between the various multiple earths.

so the N - E potential at the installation origin will be close to zero (only the voltage drop through the meter N and the earthing conductor).  

This does presume zero current in the earth but since that should be minimal (except in fault conditions) the voltage drop in the earth path will be negligible.

It also assumes equal conductor size for Line and Neutral, so for PME the parallel paths via the Neutral conductor and through the earth will reduce the voltage drop on the neutral side a bit.

parallel paths via metallic services bonded to neutral at each property will certainly reduce the voltage gradient along the road - even a mere 15mm dia water pipe is equivalent copper area to more than a 25mm meter tail. However, current through the earth itself, in comparison, will be small and almost negligible. When you hammer in an earth rod, even 8 feet of one into soft Essex clay, you still see tens of ohms of Ze. A 4ft rod into sandy soil, and you will get more than 100, maybe 200 ohms. This will support currents that hurt but will not blow any fuses worth mentioning

Compared to the hundreds of milliohms of the copper and steel paths, the earth path in parallel is almost insignificant.

Also, most of the volts are dropped within one rod length of the electrode, as that is where the current density is highest- hence the step voltage risk to animals, and advice to either insulate the top foot or so, or to bury the rod, or to fence it off  for a few feet all round .

In reality around a traditional rod, on PME a small carrot shaped region of earth around the electrode is pulled up to the local neutral voltage, and the region of influence dies away very rapidly, and that is more or less it. The neutral to earth voltage (and I mean Terra firma earth, not some bit of G/Y string connected to the neutral) can be tens, but is usually single figures, as 3 phases conspire to at least partly cancel.

It could be assumed that there is a 36.8 volt drop just on the phase between the suppliers transformer and their fuse at the customers installation.

It also assumes equal conductor size for Line and Neutral

Indeed it does - which might not be a safe assumption since some older 3-phase systems employed a half-sized N. So if you happen to be on the end of an old 3-phase cable where only one phase is significantly used and no parallel paths with N you could see two-thirds of the v.d. along N instead of half - so maybe 24.53V might be closer in some cases.


   - Andy.

I am not aware of any DNO installing superconducting Neutrals ? so there will always be some potential difference between the various earths along a PME PEN (except when zero load).


In the case of a reduced neutral the Line conductors should be sized for the full load on all three phases with an absolute maximum of 37 volt drop. 

If only one phase is being fully loaded then the DNO has an undersized neutral and their design assumption is false in providing a reduced N.

If the single phase only load is within the capacity of the half sized neutral then their will be a correspondingly lower drop on the one loaded Line.

If the voltage drop on the neutral + the drop on the line exceeds 37v then the DNO is in breach of ESQCR.  Either because of undervoltage at full load or overvoltage at no load.


In either case I don't see the N - E voltage at the origin of the installation exceeding this worst case figure of 18.5v.

In all other situations the N - E voltage will be considerably less.