What voltage would you expect on an IT system with an undistributed neutral between live and earth?

Between where and where are you seeing line voltage, and at what impedance - the whole point is that it can 'float' however a dead short to earth anywhere is bad.
if this is a 3 phase system I can well believe that the centre of the triangle will be near zero, so the line to earth voltage will seem close to line, due to all 3 line to ground capacitances being equal and having more or less equal voltages being the equilibrium condition/

Mike.

Sorry Mike, can you correct my simple thinking..

High resistance to earth meaning voltage dropped and a lower voltage to earth than line

How many phases have you got, and what voltages do you see on each of them ?

In the simple view, just measuring voltage draws no current, so no voltage drop even if loop impedance is infinite. A real volt meter is tens of megohms, but the real earth loop impedance is a bit uncertain but could easily be a few hundred k ohms, so ~ 2 orders of magnitude lower than your volt meter.

But add an impedance between 1 line and ground, that is a lot lower than that loop impedance combined with the reactance of the capacitance of the wiring (you can estimate that -  how much wire have you got on each phase ?  - allow 50-100pF per metre for small diameter stuff.) and if the system is IT, the loaded phase voltage to ground will drop away, and the others will rise to compensate and the loads wont notice as the line to line voltage is unaffected.

Mike.

Just to add a couple more questions - which version of an IT system do you have - isolated from Earth or impedance earthed? If the latter, what sort of value of impedance?

Are there any appliances or anything else with filters, that could reference the system to Earth? (e.g. via a capacitor or three.)

   - Andy.

or even an insulation monitor...

  - Andy.

Thinking out loud: When you say 'undistributed neutral', do you mean that you have L1, L2, L3 wires, without any neutral or earth wires being distributed to you (i.e 3 phase). Or, that you have an L and a Neutral (single phase), but the neutral isn't connected into follow-on circuits (not distributed any further)?

  and then where/how is the local earth (the T in IT) connected? I'm guessing [academically] this is a 3phase system and a local earth rod, with no loads on the three line conductors.

There can be sufficient upstream line capacitances to the mass of earth such that the 3-ph system is 'balanced' , or there could be sneak path(s) that bring one line close to earth potential such that the other lines show as line to line voltage. Using a load may tame the confusions.

If a line is 'floating' then it will have capacitance to whatever it is floating 'from' .

then where/how is the local earth (the T in IT) connected?

If it's the isolated version of IT, T is just connected to the exposed-conductive-parts of the loads - the supply end is separated from Earth. (In the impedance earthed version, the star point of the supply is connected to Earth via an impedance - but that star point may or may not be brought out as a N conductor (typically not, as that opens the possibility of N-PE faults which can be hard to detect).

  - Andy.

Thanks, That's useful clarification.

I was reading recently of the supply upgrade issues in Cornwall area where they had a neutral-Earth impedance (ASC?) reactor that allowed them the time to fault find line to earth faults without having to take the lines off supply, and how they were upgrading because of 'issues'. Unfortunately I can't find it in my browsing history .

I do hope that was HV !! Where we distribute neither neutral nor earth in the UK at least on overheads, so rural fault currents are actually quite low, considering the voltages.
In some countries is quite common to have neutral earthing resistors of single ohm sort of value at the source end on HV systems that have metallic earth, to limit the destructive power and impressed additional earth voltage of any phase to earth fault.

Cornwall, and a few other places in the UK, suffer from very variable ground conditions, all the way from wet mud to solid rock, and the latter can be awkward as step voltages don't  decay over distance as fast as they do over a more uniform soil.
Mike.

Edit According to P27 of this Western power https://www.nationalgrid.co.uk/downloads/3238

they use Peterson coil (a parallel resonant coil) earthing to limit the HV fault currents to a very low value to allow fault tracing, especially  in Cornwall and on the Gower Peninsula in Wales.

Yes (your Edit) the report I mentioned was a WPD (Western Power Distribution). It included discussions about the replacement of the resonant coil system to meet various requirements, and how that was proceeding.