trouble parallelling generators 3

[snorkles]

Generator paralleling problem

Our plant has Two-2 M watt, 480 volt alternators. Scenario: Generators are paralleled via APD onto a common bus. Voltages, kwatts, kamps and frequency meters all concur.

Problem 1: The power factor meters are 180 out. Generator 1 meter is pegged to max. IND. Generator 2 at max.CAP.

Problem 2: When Load is applied it is not equally shared. kW meters are equal however kA meters are not. Gen 1 is producing more power than Gen 2

Question. What adjustments were needed? Should operating Gen. 1 be loaded, prior to bringing on incoming Gen. 2 ? And finally, what's happening with the power factor meters?

 

[electrageek]

snorkles

What is APD? What happens when only one generator is running on the Grid? Does each unit have the same problem when running by itself?

The problem may be your droop CT connections, to the Voltage Regulator (VR), if you are in the Automatic Voltage Regulating mode (AVR). The droop CT is suppose to be connected in such a way as to cause the AVR setpoint to be lowered as you supply more and more reactive power to the grid (i.e. Power Factor more CAP). The CT along with the Generator PT's helps "droop" the Generator Voltage so as to provide a stable system. However if the CT is connected backwards then the more reactive power you provide the higher the AVR setpoint will be and the more reactive power you will supply and the system will become unstable and max out on reactive power. This would cause the situation you described.

The second problem is related to the first problem. The current in the Generator is a sum of the real power (KW) and the reactive power (KVars). The KW power is provided by the prime mover and is function of either fuel or steam and provides real work. The KVars power is the energy controlled by the excitation system and provide the energy needed to create magnetic fields. We control KW by controlling the fuel and KVars by controlling the excitation. If you are maxed out on Power Factor then you are maxed out on excitation and the current can be very high.

To check the droop CT I normally started by checking the connections, polarity, and verifiing that it is installed per the manufacturers instructions. I usually start with a very small percentage setting for the droop on the automatic voltage regulator. Second I start the unit and match the frequency and voltage and synch the Generator to the Bus. I then put a small amount of reactive power on the Generator if it trys to immediately take too many KVars I know the CT's is probably backwards. I it doesn't then I continue to increase the KVars to about a 0.9 PF. Now take a wire jumper and temporily short out the CT at the shorting block. The voltage on the Generator should rise slightly. Wthen the jumper is removed the voltage should drop back down. This would prove that the CT is installed correctly.

 

[snorkles]

Thanks for the responce geek!

APD is auto parallel device...Individualy run,Genset power factor meters are within expectation.

Up until the last routine run, these generators have been sharing the load evenly.When on Generator power we are an island.Another test run will be given the 29th.I still wonder if operator error played a factor.Our operators are not well trained.

 

[electrageek]

snorkles

Is there a Kw and KVar load share system in place from someone like Woodward. Small Gen sets sometimes use a Woodward 2301 govenor. Maybe the Kvar load share, which works through the excitation system, has a failed component.

 

[snorkles]

Gensets controls are woodward and Schneider.I will investigate further...

 

[busbar]

Small comment: On the out-of-whack PF-meter readings, there can be more than one variation of acceptable connections. It’s possible that a quadrature connection is needed…i.e., Bø current and Aø-Cø potential. {Or, Aø current and Aø-Cø potential, or Aø current and Aø-N potential.} It's a lot more reassuring not to have ‘guess,’ ‘mentally compensate’ or ‘overlook’ odd readings in power gear.

 

[electrageek]

snorkles

Because this use to work, I would think we are looking for a failed components somewhere. If the units run islanded then there would not be an issue for PF control because the Kvar load is fixed by the system. However there still is a need to either share Kvars or to have both units droop percentages set about the same. Are the units set up for cross current compensation. This is where the droop ct's of each generator are interconnected to prevent one unit from hogging all the Kvars.