Grounding of 10kV parallel generators

[Thenewbie]

Hi,

I have just found this great forum and had to register immediately.

I have a theoretical problem at hand concerning parallel medium voltage generators prime feeding the grid.
I have drawn up a fictional network of five gen sets connected to a 10kV switchboard. In order to try to understand things involved. The switchboard is connected to a 16MVA delta-wye transformer, stepping the voltage up to 110kV.

What confuses me is the relation between transformer and generator sets grounding. If the transformer would be connected to a stable 110kV grid the star point could be grounded or not (if not, then maybe simply just having a over voltage protection to ground from the star point)

I have understood that if having the star points ungrounded at the generators, and if a phase to ground fault at e.g the generator switchboard then there would be serious difference between phase voltages from the gen sets. Lowering the voltage at the faulty phase and higher voltages for the remainig two phases.

And if assumed a very small resistance for the ground fault, close to zero, then the voltage would totally collapse on the faulty phase and cause the main voltage 10kV as phase voltage on the healthy phases.

If solidly grounding, or low resistance grounding, the generator star points then a phase to ground fault would cause seriously high ground currents. Maybe damaging the generator windings, although having protection relays for the gen sets. I have understood that there could also be problems with third harmonic currents during normal operation.

If using high resistance grounding for each gen set then the phase to ground current would be lower and maybe allowing some delay to be set in the protection unit, without compromising the generators. But what would happen concerning the selectivity between generator- and main transformer circuit breakers?

Seems that I have stumbled over this grounding issue, and I would appreciate if someone could kick my brain regarding correct choice of grounding. And would You have any information regarding how the resistance of the neutral grounding resistor is chosen?

 

[rbulsara]

Start out with studying IEEE Standard 142 on grounding, the Green Book and also IEEE Red book.

Rafiq Bulsara

http://www.srengineersct.com

 

[waross]

Consider a direct connected transformer for each generator. Sync on the high side. Differential protection and reverse power protection will back up ground fault protection.

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

 

[dpc]

Solid grounding of medium-voltage generator neutrals is a bad idea in general, unless there are single-phase line-to-neutral loads that must be served at the generator voltage. Solid grounding results in high fault currents and more severe damage to the stator core (steel) for internal faults.

Resistance-grounding is recommended. The exact method depends on a lot of factors. High-resistance provides the best machine protection but it is difficult (generally not possible) to selectively coordinate the ground relays in the event of a line-to-ground fault. Low-resistance grounding (50A to maybe 1000A) allows for better selectivity and coordination at the expense of greater machine damage in the event of a fault. There is also a decision to be made regarding the location of the resistor and the number of resistors in service at any one time, when there are multiple generators. Third-harmonic currents can be an issue with multiple ground sources. The magnitude of third-harmonic currents is dependent on the pitch of the generator windings and other factors. The expected steady-state 3rd harmonic current must be considered when sizing and rating the neutral resistors.

There are a lot of good references on the subject of generator grounding. You might want to take a look at this:

faq238-1287



David Castor

http://www.cvoes.com

 

[Thenewbie]

I found the green and red books. Thank You very much for the this great hint!

Did already read a little and will certainly study it more carefully.

First i thought about using individual block transformers for each generator. The cost of 3,0MVA 0,4/10kV transformers + required cabling on low voltage side make it "not so realistic"

I thought about "using" 3,0MVA 10kVA 4-pole generators with CT:s on star point side (for current measurement and winding protection) Voltage measurement transformers on primary and possibly droop CT on R phase, for A.V.R.
Feeding out through vacuum circuit breakers, to air insulated switchboard.
The gen set controllers would control the diesel engines and generators, also including following protections:

- Over- short circuit current protection
- Under- over voltage & Under- over frequency protection
- Overload protection
- Reverse power protection
- Earth fault protection
- Vector shift protection

These protections would be backed up by some kind of programmable protection relays and arch protection.
Then 1 synchronized vacuum breaker out from switchboard to the distribution transformer 10 / 110kV 15MVA.

The main issue would be to get familiar with parallel generators and medium voltage systems. Best way would be through a theoretical project.
But the grounding of medium voltage applications is something I´m not familiar with. Including grounding of multiple generators in parallel run.

As I quickly read through the green book, maybe a low resistance grounding resistor for each generator could be a possible combination. If the generators have 2/3 pitch windings, maybe they would minimize the third harmonics.
Then CT measurement to both generator controller and chosen auxilary controller.

Seems there is really a lot to consider, a really good calculation program would probably easy things a lot.

 

[rbulsara]

That is nice! Yes, there is a lot to consider and learn.
I hope that this is not your homework assignment. They are not encouraged here.


Rafiq Bulsara

http://www.srengineersct.com

 

[Thenewbie]

I have already graduated so no homework assignment, no commercial interest. An effort to expand control system knowledge, and trying to learn new stuff beside.

 

[living2learn]

Use a neutral grounding reactor in lieu of a resistor to eliminate harmonics, but still limit fault current?

 

[Thenewbie]

I have spent some hours going through the links provided by dpc. Great stuff! I have contacted some generator manufacturers, but seems really hard to get info like all the reactance- reactance- and time constant values. This of course having a great impact on short circuit currents and ground fault current.

Found some articles about different groundings for MV systems. One article recommended the separate low resistance grounding for each generator, if the generators where to be connected at a common bus. In this case I had planned to "connect" five or more sets to a common bus. But in my opinion here would be some serious differences in ground fault currents, if only some sets are running compared to all of them running. If all running and a ground fault at a single generator, then I have understood that the contributed short circuit current might overshoot the protection provided by low resistance grounding.

Concerning the use of reactors: I think there might have to be a reactor involved somehow on the low voltage side of the 15MVA 110/10kV main transformer, in order to limit short circuit currents on 20kV side. Either at the star point of the transformer or as a series reactor on the 20kV. In order to limit the short circuit current distributed by the grid. But have not reached to that point, still have to try to find out many things before that.

Concerning the grounding issue: One interesting alternative could then be to use individual high resistance grounding for each gen set and a zig-zag grounding transformer at the MV switchgear. Thus allowing higher ground fault current at switchgear and still keeping ground fault currents at a reasonable level at the gen sets. This in order to protect the generator windings.

As this "project" is only a study, and I don´t have to adapt to any existing numbers regarding dimensioning. So gives me a little ease that way. The biggest issue for Me is to try to learn something of this (what actually looks like a big mess at the moment) I will still have to study characteristics of each component involved, and of course the theory behind using different MV/HV components.

I have only the most basic information regarding MV and HV applications. The MV/HV theory was actually very superficial where I studied. Luckily standards put limits to e.g touch voltages etc etc So having some "certain" values which the "system" would have to meet. Still hard, and very time consuming, to try to move forward piece by piece. And of course to find information regarding standards, as these are commercial info and being very expensive.

The first intention was actually to study the control of diesel engines through J1939 and the control setup for parallel running. All this has expanded from this all the way to the 110kV grid and thus those components involved.