Neutral grounding (earthing) of parallel LV generators

[BrianWaddell]

The numerous publications I have found all talk about the standard methods of neutral earthing of parallel LV generators ie. all generators solidly connected to earth/one generator only in a group connected to earth/neutral earthing transformer. I understand that if (as recommended) only one generator in the group is connected to earth then the zero sequence current will return through that generator alone. There must be a limit to how many sets can be connected in parrallel before the magnitude of the zero sequence currents (ignoring fault level) is too great! I need to get hold of some references/guidance on firstly the calculations and secondly typical limiting values of the zero sequence current can anybody help?

 

[dpc]

If you have multiple generators on a grounded low voltage system, especially a 4-wire (480/277V) system, the best approach is to not directly ground the generator neutrals at the generators. Bring all the insulated neutrals back to the main switchgear and let the main bonding jumper ground all the generator neutrals at that point. This will eliminate all of the ground relaying problems you will have with multiple grounding connections, or worrying about which generator is grounded, etc.

 

[RRaghunath]

Bonding all the neutrals (and Connecting to ground) as suggested can cause harmonics circulate in the generator windings.

It is better to have isolation in individual neutral connections and ensure only one generator neutral is earthed at any given time.

The above applies to solidly grounded system and no such restriction is relevant in case of high or medium impedance (estimated earth fault currents not exceeding 100A)grounded systems.

Trust the above is helpful.

 

[dpc]

Harmonics can be an issue in some cases, but if all generators have the same winding pitch, the generated third harmonic voltages will be equal and little current will flow. However, third harmonic current can be fairly high and still not cause major concerns. The effective current is the square root of the sum of the squares of the harmonic currents. 480V generators are commonly operated in parallel with all neutrals grounded.

Actually, with a low-resistance grounded system, the harmonic current issue is more significant since the grounding resistors often do not have a continuous current rating.

Switching of neutrals to maintain single grounding point is a complicated and messy arrangement that should generally be avoided if possible, in my opinion. Just one more place for mistakes to be made.

 

[ruggedscot]

We had a refit recently of our gensets and they are wired in a different manner, usual three phase y connected sets but the main difference is out of all the sets connected to the bus we only operate at any given time with one y point connected. Fancy control involved - first set up and on to the bar normally closes the neutral. All other sets after this connect on phases but not the neutral.

Went through this with design engineers and the reasons being for harmonics and circulating neutral currents. The earth point being in the main panel means that there would be only one earth point.

Anyways isnt this the normal way to earth down a multiple set system so that the protection works on the alternator sets ? Using restricted earth fault relays on the multiple feeds allows the bad set to be dropped off of the bar without losing the other sets and maintaining the continuity of supply ?

Rugged

 

[RRaghunath]

dpc,

Thanks for pointing out. It is true 480V systems are solidly grounded always. What I said applies only to MV systems.

With regard to paralleling of generators, I am not sure about the circulating third harmonic currents being safe for the generators.

In India, I have always come across the practice of earthing only one generator in the group. The earthing isolators come with auxiliary contacts and the monitoring is also part of the neutral earthing scheme, to prevent human error causing harm (true even for MV generators) to the generators / system.

With regard to grounding resistors (in case of MV generators), the resistors reduce the magnitude of circulating currents substantially and hence do not pose any problem of heating to the resistors. It is also worth noting that, though the resistors are short time rated for the estimated magnitude of earth fault currents they also possess certain continous current carrying capability.

 

[alehman]

In the U.S., it is not true that 480V systems are always solidly grounded. Both low and high Impedance grounding (resistance or reactance) is sometimes used on 480 systems, not that I would recommend it in most cases.