MV generators

[Elecme]

Do we have to ground the MV (11kV)deisel generators through impedance ?.... and what is the difference between high, low impedance and solidly earthing rather than the limitation of the fault current and the interruption at 1st or 2nd fault......etc.

Elecme

 

[jbartos]

Suggestion:
1. If the grounding calculations results allow for the high-resistance system grounding, it is wise to implemented.
2. If the grounding calculations results do not allow the high resistance grounding, then next in the system grounding hierarchy is the medium resistance system grounding.
3. Solidly grounded neutral system grounding is not used on Medium Voltage (MV) level since it produces an unacceptable energy release due to short circuits in most instances.

 

[bigamp]

Elecme,

As jbartos suggests, there are many considerations and many different approaches.

High resistance (or impedance) earthing limits the earth fault current to between 5A and 10A. This level of current is generally accepted as being low enough not to cause generator damage even if the fault is in the generator. This is best way to go if the generator is connected via a unit transformer. If the generator is directly connected to a bus to which other load is connected (motors, transformers) and the generator can be the only source of supply then you may have a problem being able to implement a time graded earth fault protection system because the available earth fault is very low. If you want earth fault protection for the individual feeders connected to the bus, you will need to have low ratio core balance CT's fitted to the feeders to be able to detect the low level of earth fault current available. You would also need to ensure that these CT's did not start to produce output current (and so cause a false trip) during high loading or through faults. With high resistance earthing, if the generator is the sole source of supply, there will not be sufficient earth fault current available to operate protection schemes such as generator diffenrntial or restricted earth fault.

For a directly connected generator, I believe that a better approach is to size the earthing resistance to limit the earth fault current to something like 400A (typical North American approach) or to generator full load current. With these systems, a conventional time graded earth fault protection system will work and you will be able to detect earth faults in individual feeder circuits connected to the bus. In addition, there will be sufficient earth fault available to be detected by unit protection (differential or restricted earth fault) so that internal generator faults result in an instantaneous trip.

I have seen other systems where the generator is run as a 3-wire machine whilst running parallel with the grid but upon loss of grid, a switch automatically closes in to directly connect the star point to earth. It could just as well have connected the star point to earth via a resistance.

If you are operating with two or more generators in parallel then you also need to consider the dreaded circulating third harmonic current and an effective way of preventing this is to ensure that only one generator star point is ever connected (via resistor) to earth. You can achieve this by connecting each generator star point to a "star point bus" via an automatically operating switching device and put in controls so that only one switching device is ever closed in. The star point earthing bus is connected to earth via a single earthing resistor.