Isolated Ground

See:
and

Assuredly you are markedly better off {and usually money ahead} designing in electroamgnetc compatibility that does not require special grounding.

Suggestion: It is better to be more precise when it comes to grounding. Essentially, there are a system ground and an equipment ground. The equipment ground can be "isolated equipment ground." An isolated equipment grounding conductor runs, in the same conduit or raceway as the supply conductors. This conductor is insulated from the metallic raceway and all ground points throughout its length. It originates at an isolated-ground-type-receptacle or equipment input terminal block and terminates at the point where neutral and ground are bonded at the power source.

I would extend Yme48 questions, the case is in 13.8 kV distribution line. There is an idea by my collegua to reduce phase to ground/temporary fault by operating the system to be isolated neutral when the fault occured, so there will no longer zero sequence flowing in the system. This idea can be implemented by install additional breaker at transformer neutral ground. This scheme is also equipped by a reclosing relay to initiate the ground braker to reclose, and if the fault still exist, the respective line/feeder will be tripped/opened by rspective breaker's feeder as usual.

Do you have any idea/comments/inputs regarding this idea??
Please

vandax,

The full fault current will flow until the neutral breaker trips. What advantage is there over the normal practice of tripping the main feeder breaker and reclosing? I can foresee a lot of problems on unfaulted feeders that lose their ground reference.

Assuming you are following NEC (please verify):

The term "isolated ground" is misleading, and does not appear in the NEC. The applicable paragraph is NEC 250.142(D), Isolated Receptacles.

As mentioned above, this paragraph permits a receptacle to have the ground terminal insulated from the yoke, a separate insulated conductor is run from the receptacle back to a ground bus in the source panel or upstream panel. It is not truely "isolated", the resistance between the grounding terminal and yoke is very low once the required insulated grounding conductor has been installed. Such receptacles are identifed by their orange color or by a small orange triangle on their face (I don't believe the orange color is an NEC requirement, I think it's UL).

NEC, in general, requires all grounding systems to be bonded together in at least one point. NEC requires that the neutral and ground be bonded in one point (or sometimes two).

Be aware that overcurrent protection in the neutral is typically prohibited by 240.22 unless it simultaneously opens all phase conductors. Also, as mentioned above, an overcurrent device in the neutral would typically serve no purpose, with two specific exceptions:

1. It can facilitate maintenance if there are high neutral voltages present.

2. It is required if you are using the OC device to switch between two separately-derived 4-wire systems. This is rare, and can be more technically challenging than it first appears, and I'd generally recommend dual 3-wire (vs. 4-wire) systems with downstream transformers if you do have a legitimate reason to require transfer between separately derived systems.

Hope this helps.

In response Jghrist's comment, I need to clarify that we still use the ground relay(51G), which should be set under the 51N at feeders. So, whenener single phase fault occurred at the line, the 51G will sense and initiate to open the "new additional breaker beetwen neutal Tx - ground", and the fault current stop to flow once the system change to be floating neutral (isolated neutal).

15-30 second after, the new breaker reclose, and when the system still sense the ground fault (permanet fault), the feeder's breaker will be opened by 51N.

This method is expected to significantly reduce number of outage due to phase to ground temporary fault.

Comments are wellcome

vandax,

Perhaps you should start a separate thread because this is not really related to the main topic.

In your first post, you said you wanted to reduce the phase-to-ground fault. The full fault current will still flow until the xfmr neutral breaker trips.

If you leave the neutral isolated for 15-30 sec, the system will have no ground source. One phase may likely be connected to ground still, maybe through the impedance of a squirrel. Phase to ground voltages on the unfaulted phases (all feeders) will approach the system phase-to-phase voltage. Any phase to ground connected equipment will be subjected to overvoltage. Surge arresters may fail. Phase-to-ground voltage on the faulted phase (all feeders) will approach zero. Large voltage unbalances may occur on three-phase motors if they are on the secondary of wye-wye connected transformers, resulting in many motors tripping off line.

I still see no advantage over the normal practice of tripping the faulted feeder and reclosing to clear temporary faults.

Sorry, I still use this thread, since its relevance..

jghrist,
1). all our LA in distribution line has consider Vpp, so there would be no problem with the arrester or other eq.

2). most of our TX connection are delta wye, so it would not make voltage unbalance in load side/motor.

3). Up to now, we just worried if the new recloser is lock out, the system will be permanently ungrounded.

Please imagine, you have phase to ground temporary fault, but your system still on by floating neutral for 15".

Comment please

vandax,

I would worry not only about voltages during the time of the floating neutral, but transients that would occur during the opening and closing of the neutral breaker. Analysis of transients should be done with EMTP or similar software.

I still don't see any advantage to such an arrangement.

vandax, if you have no qualms about the issues raised by jghrist, why don't you just permanently remove the neutral? Then you would have no outages at all when a phase faults to ground!
It appears to me that your argument against his (very pertinent) reasons indicate that you are not concerned about the basic reason that a 13.8kV system has a grounded neutral, that is, to keep a consistent phase-to-ground potential.

jghrist,
Thanks for your susggestion. That's rigth, we don't concern to analysis of transient yet. The tool we have right now is ETAP only.

Dandel,
Actualy, this idea is come out by my collegua, and almost all of us agree to implement this immediately as current point of view/perspective. Why we don't implement ungrounded system permanenty? We all agree it is about safety/step voltage concern.
Itjust make the system for 15" ungrounded.