Islanded Generator Testing 1

[TkMork]

Good Morning/Afternoon/Evening to All,

I am inquiring as to some guidance on testing precautions and feasibility. The Generator under testing conditions is a G.E. LM2500 rated at 25 MVA, 13.8 kV, and 0.85 P.F. It has a Basler DECS-200 AVR system and we are looking to test its capability as a stand-alone unit by using a load bank configuration with different load profiles. The peak of the profile would be to attempt to simulate what would be a large inductive motor DOL starting such as a boiler feed pump rated at 1750 HP, 4000V, and 225 FLA with a continuous 3 MVA load already applied that experiences a peak of 10 MVA due to in rush from the boiler feed pump starting.

My question involves around the feasibility of this testing. The generator is being tested in its AVR condition which I believe to be its isochronous mode. It normally is operated in droop mode. I don't doubt the generators capability of supplying the intended load in a continuous fashion but I am questionable of the generator in a stand-alone setting experiencing large inductive motor starts. Unfortunately, the unit is very old and there is lack of performance data or even reactance values on hand.

What kind of information should be acquired before administering such a test?

What kind of precautions should be taken when applying loads to the generator in steps? To simulate a motor start, I would imagine using something along the lines of a step function where VA is high with low power factor at start and then incrementally raise power factor as VA decreases as the motor speeds up to operating condition.

How would you go about determining if the excitation system within the generator is capable of handling motor starting scenarios?

Thanks.

 

[wayne440]

...What kind of information should be acquired before administering such a test?...

The first information that should be determined is "who gets blamed/fired/billed for damages if it breaks during the testing".

 

[edison123]

No, it will be a constant high start KVA and a very low pf of less than 0.1 up until about 90% of rated speed. The start KVA depends on the motor design and and the start duration on if any soft start like a fluid drive is used.

Muthu

http://www.edison.co.in

 

[waross]

Adding KVA will tend to pull the voltage down. The AVR will increase the excitation to hold the voltage up.
Adding a 10 MVA surge on a 25 MVA set, should be within the capability of the AVR.
A second concern will be the MW loading. That will be about 250% to 300% of the motor full load MW.
The concern now is the ability of the set to pick up block loading and the response time of the governor.
I would suggest block loading the set with about 5 MW to check the overall response to block loading.
I anticipate that the MW load will be of greater concern than MVA loading.
Assuming 5% droop, and a 5 MW block load on a 21.25 MW set, the frequency may stabilize at 59.3 Hz.
It will probably drop more than that before the governor can ramp up the fuel.
But it will not stabilize, as the motor is accelerating. when the motor is up to speed, lets assume 80% efficiency including power factor.
Let's use 1.95 MW for the running, loaded motor. The frequency will stabilize at 59.72 Hz for a short time. Then the isochronous feature will trim the frequency back to 60 Hz.
My suggestion: Block load the set with 5 MW. on top of the base load. Observe the frequency dip. This should be a good indication of the ability of the set to pick up the motor starting load. Watch the voltage dip during the test. The voltage dip may be twice as much when actually starting the motor.
The base load will help buffer the motor starting.
I don't anticipate a problem starting that motor on that set. You have lots of over capacity.

--------------------
Ohm's law
Not just a good idea;
It's the LAW!