I have been reading about the different types of grounding options there are for a new power system. I read that if a ground fault occurred in an ungrounded (capacitive grounded) system and is intermittent or allowed to continue, the system could be subjected to possible severe over-voltages to ground, which can be as high as six or eight times phase voltage. This can puncture insulation and result in additional ground faults. I wonder why they say "possible" severe over-voltages could occur, aren't they sure? I wonder if it's a sure thing such that it could be calculated exactly 8 times as much say when variables are at certain values? I wonder what precisely the condition are, what the driving force is, that guarantee severe over-voltages will occur?

Transient overvoltages causing insulation stress, can occur whether a system is "grounded" or not.

On the other hand, high frequency spikes can cause pin hole punctures on insulation, leading to failure of equipment; I experienced this at first hand, during the early days of vacuum circuit breakers; there was no means of dissipating the V.H.F. voltage arising from the sudden chopping of current, normally present in oil circuit breakers.

Due to failures of a newly commissioned 3.3kV motors, I sent one that had only been switched on and off for commissioning, to the rewinders for investigation; it failed on insulation testing and pinholes were found in the insulation on the leads from the terminals to the windings. The ERA got involved since it was a widespread problem, the problem was resolved by the introduction of snubber circuitry across the switching contacts.

Regards

Thank you for the information. The snubber circuit is an interesting devise. I read that the basic function of a snubber is to absorb
energy from the reactances in the power circuit. I understand the basic theory; the driving force behind the transient overvoltage is the energy stored in the field of an inductor or capacitor. I just wonder how this energy accumulates and how If possible the building of voltage could be predicted. I wonder how frequency is a factor in transient voltage. I wonder if a snubber circuit works on frequency like a filter. Perhaps a snubber circuit is the solution to all these problems, but I don't think one would design a power system with the expense of including a snubber without first predicting the risk of transient overvoltages. In your example, it was after the system design that excessive transients were discovered.

A couple of different ideas come to mind - which may or may not be relevant (so apologies in advance if I'm going off at a tangent here)...

Firstly, a secondary of a transformer can be capacitively coupled to the primary - so with say a 11,000V / 400V transformer, with the 11kV side referenced to earth but the 400V side was unearthed, the LV side would have tendency to be pulled towards the HV potential, which can then overstrain the insulation between the LV side and earth, even though there'd still only be 400V between LV live conductors. Exact figures would depend on the radio of the capacitances across the transformer and between the LV system and earth and the configuration of the supply, but as an example, 11kV in the UK is usually between phases - so is actually around 6.35kV to earth, if capacitive coupling tied the mid point of that to the LV system it would be around 3175V - which is just shy of 8x 400V. So that might be one theory.

Another one is ferroresonance - (i.e. the inductors (and capacitances) inherent in a system can sometimes allow it to oscillate) - which can then produce over voltages (a bit like the old electric fence controllers) - don't ask me to explain - but you could try the "Cahier technique" paper number 190 on the subject (Google will find you a copy).

- Andy.

I googled Cahier technique. I actually got this IET discussion at the top of the list, which is interesting. I think your first example my be that conceptual explanation I was hoping existed. It seems to make sense. I'll need to work on it. I will be taking Power Quality soon.
Thank you for the information.
Alan