Multiple Ground Faults on Isolated Ground systems of mixed voltages

[Bloozntooz7868]

I am having a bit of trouble getting my head round a problem regarding to systems of varying voltages running in Isolated Earth configuration. It is a floating vessel which comprises of a Topsides earthed system at 13.8kV, and feeders from the Switchboard go to the Marine side (Hull) Switchboards through Delta/Delta 13.8kV/0.69kV Transformers, so providing an ungrounded system (This is a normal setup for vessels in order to enable them to run essential services with a single ground fault).
From the 690V Board, there are many sub-systems which feed other Transformers, predominantly 690/440V and 690/230V, for lighting etc etc. There are also 3 x 440V Marine side Essential Power Diesel Generators that run through 440/690V step up Transformers to feed the main 690V Board in case loss of Topsides power.
The Ground Fault detection is done through 'Bender' relays - open delta type sensing, which provides alarms in case of a GF.

So essentially, there are a whole bunch of ungrounded systems at multiple voltage levels.

What I would like to understand is what happens if we have a 690V ground fault combined with, say, a 230V ground fault. What happens vectorially with the voltages? What happens to the different system voltages on the various Boards - are they affected? If yes, then how? If not, then why? Or am I just getting all confused over nothing?

A vector diagram would help immensely in addition to providing a narrative.

Thanks a lot in advance.

 

[Fuse4]

Dear Bloozntooz7868

In this Excellent book "Industrial Power System Grounding Design Handbook" by J.R. Dunki-Jacobs • F.J. Shields • Conrad St. Pierre, you can find about your question. Look the cover of the book in

http://www.groundingdesignbook.com/

anf Fig. 12.2

http://www.groundingdesignbook.com/sketches.html,

here there are the idea how work separate systems in ground fault contidions.

Other resources are "Application considerations for high resistance ground retrofits in pulp and paper mills" by Robert Beltz, Ian Peacock, and William Vilcheck, them say "... A grounding system is isolated from other grounding systems by delta windings in three-phase systems. It only takes one delta winding to accomplish isolation; not both primary and secondary windings ...". Also, the autors have a paper "High resistance ground Retrofits in pulp and Paper Mills" in the IEEE Industry Applications Magazine March/April 2001.

The behaviour of delta winding would be similar to the isolated neutral winding in a transformer and A single ground fault in the transformer delta side system do not pass to the other system, but if a grounding system is with fault, it pass to the transformer delta side in a phase to phase shortcircuit/oveload.

 

[Bloozntooz7868]

Fuse4,

Thanks for the reply.

I'm not sure that I will be going out to buy the book specially for this, but it does look to be the kind where I could find out the answer.
I take it that because you are the only responder (and I appreciate that), the question cannot be readily answered by other Eng-Tips members.

Thanks again - I appreciate the lead.

 

[Bloozntooz7868]

I have attached a hand drawing of what I believe are the vectors for an Ungrounded Star, and an ungrounded Delta system, and shown it with a ground fault, on different phases, on the two different voltage systems.

Link

Please comment on them - good or bad.

The first one (Ungrounded STAR) shows the shift of the faulted phase to ground, and the other Phase Voltages increasing by SQRT(3) to ground (this means that cables rated VL-VL opposed to Vp-VL will need to be used).

The second one (Ungrounded DELTA) shows the resultant vectors, with the two systems 'piggybacking' on each other.

If any of this is incorrect then please let me know.

I understand that there will be no Zero Sequence current, because there is no route back for the current (apart from the Capacitive currents).

What I don't fully get is, how can the ground (Hull) charge up to 690V for a 690V system ground, but then for it also to be said that its charged to 440V for a 440V system ground at the same time? If the Hull charges up then won't it be charged to the highest of those voltages (in this case 690V)? and in this case, what effect will this have on the lower voltage system (440V) due to them now 'sharing' a common line?

Thoughts anyone?

 

[waross]

The hull doesn't charge up.
The hull pulls the faulted phase down to zero and the un-faulted phases assume line to line voltage in respect to the hull.
When an on-shore delta system is corner grounded, the whole world does not assume a charge as a result.
If you measure voltage from one phase of a 440V system to one phase of a 690 Volt system, you will see interesting voltages as you move your measuring point from phase to phase. The measuring instrument may affect the readings.
If each system has a ground on it the Voltage measurements will still be interesting but may be more stable and not affacted by the measuring instrument.
If you are grounded and contact a phase of a healthy, ungrounded, 440 Volt system you will receive a shock of about 254 Volts.
If you are grounded and contact a healthy phase of a normally ungrounded, 440 Volt system with a ground on one phase, you will receive a shock of 440 Volts.
You have separate systems and the systems should not interact. However given some of the strange wiring that I have corrected on vessels, there may be some strange, unknown connection such that the systems may interact during simultaneous ground faults. It shouldn't happen but don't bet your life on it.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[Bloozntooz7868]

Thanks waross,

Sometimes it doesn't take much for the penny to drop. By reading your opening few lines I suddenly realised where I was going wrong - all phases simply take their positions relative to the grounded point, and not the hull assuming a potential of the phase(s).

I take it that my sketches are electrically correct (just for future reference)?

This was more of a theoretical question - but we have had problems with various 300mA RCD's (GFCI's) tripping on a a single phase-to-ground fault, but I guess this is more to do with capacitive current rather than any real fault current (but that's another story).

Thanks for clearing things up - appreciated.
David.

 

[crshears]

Hi blues 'n' 2's, I would have weighed in on this one but I was away, and since then Bill has said it better than I ever could. I'm glad you posted this one, though; it will be good fodder for for a project I'm pushing where I volunteer [ visit www.sskeewatin.com ].

CR

"As iron sharpens iron, so one person sharpens another." [Proverbs 27:17, NIV]