power factor not stable

[wuqn]

the system has three paralelled generators. But the power factor of the two is not stable. one's PF increase then the other decrease. I don't know why. what's the problem?
how can i fix it? thanks.

 

[waross]

Check your parallel switches.
When generators are connected in parallel there is a circuit consisting of a CT, a resistor and a bypass switch. The purpose of this circuit is to balance the power factor between the generators. If the contacts are going bad in one of the switches it may cause the power factor to shift between generators. Check the circuit for bad connections. The resistor may be built into the voltage regulator.
BE VERY CAREFUL WORKING ON CT CIRCUITS WITH THE POWER ON!
If a voltage regulator is unstable it will cause the power factor to shift between generators.
respectfully

 

[oldfieldguy]

wuqn-

waross is right. With paralleled generators, you have to use a circuit to balance reactive power between the units. If you have means to monitor the excitation current from each exciter, you will probably see the excitation current fluctuating also.

In addition to waross's suggestion, you might find that there are controls to "tune" the circuit to prevent it from "hunting", resulting in your symptoms.

In the past I've found problems with the switches that enable these paralleling circuits as well as problems with the control components on the exciters.

old field guy

 

[waross]

The circuit in question is often called a quadrature circuit.
The voltage regulator (AVR) senses the voltage across line "A" and line "B".
A CT is installed on line "C" and the resulting current is dropped across a resistor.
The current in line "C" is 90 deg. out of phase with the voltage from line "A" to line "B".
However reactive current is 90 deg. out of phase with the real current and so the reactive component of the current in line "C" is in phase with the voltage from "A" to "B".
The resistor is placed in series with the sense terminals of the AVR.
Normal in-phase current and voltage from the CT is at right angles to the sensed voltage and has very little effect on the voltage at the AVR.
Quadrature current resulting from a lagging power factor develops a voltage that is in phase with the sensed voltage.
This acts to lower the excitation so as to reject some VARs to the other generators, which in turn are increasing excitation to supply the added VARs.
Many AVRs have the resistor built in and the circuit is more complex for a three phase sensing AVR.
If you are the first to work on the problem, remember the first rule of trouble shooting.
"Gremlins don't change the wiring. Look for a component failure and don't change any connections."
If you are the second person to work on the problem remember the second rule of trouble shooting.
"The first person may not have known the first rule of trouble shooting"
Open secondary circuits on CTs are deadly, and polarity matters.
respectfully

 

[Alex68]

The generators are exchanging reactive power each other.
You can choose two ways to avoid these pf oscillations:

1- If one generator is in voltage regulation, then the other two generators should be in pf tracking. You should check that your AVRs have this kind of control.

2- The best way to avoid oscillations is to activate the AVR compound with a negative gain. This introduces a droop in the (Q,V) regulation.
The compound signal of the AVR allows to compound the reactive losses in the transformers:
v' = v + a i sinphi
Remebering that q = v i sinphi, then
v' = v + a q /v
So the (a q/v) is the compound signal.
If a<0, we obtain that a is like the droop of the frequency regulation. Typical values of a are:
-0.08 < a < -0.06
In this way the paralleled generators will correctly share the reactive production without oscillations.

http:\\m2ec.manunza.com

 

[ScottyUK]

If one generator is in voltage regulation, then the other two generators should be in pf tracking.

This can be a risky strategy: as load power increases the generators in PF controll will attempt to maintain their programmed PF by increasing or decreasing excitation. This can lead to the generator in voltage control either reaching maximum excitation limit or lowering excitation down to the stability limit.

Suggestion 2 is the correct way to deal with the problem.

Are the generators of similar rating?


----------------------------------

Sometimes I only open my mouth to swap feet...

 

[rbulsara]

wuqn:

If your first priority is to fix the problem, call the service technicians of the generator and/or control gear manufacturer and have them fix it.

If you want to learn some basics of generator paralleling, here is one of the links. As you would appreciate its rather complex issue, but modern controls makes it easy. The key is to know how your system is set up. Read up reative dropp compensation and cross current compensation methods for reactive power control. IEEE uses slightly different terminology which is mentioned in this pdf.

http://www.basler.com/downloads/VR_parallel.pdf

 

[wuqn]

Thanks everybody.
ScottyUK: the generatos are similar.
rbulsara: the service is not good.
Alex68: Can you show how to activate the AVR compound with a negative gain ? do I have to reduce the AVR's gain?

 

[Alex68]

wugn,
please do not modify the AVR parameters by yourself if you are not extremely sure of what you are doing!

You should check the AVR documentation and look for the transfer function diagram or something like that.
One of the input to the reference of the control chain is the compound, beside the PSS and over- / under- excitation limits.
You should set only the gain of the compound signal, but the correct value is the result of a calculation or a simulation.

Alex68
http:\\m2ec.manunza.com\

 

[waross]

Did this installation run well at one time or is this a new installation?
respectfully

 

[powerwagon75]

Waross, you just asked what I was going to. As you stated earlier, "polarity matters.

Wuqn, if this is a recent installation, or if some electrical maintenance/replacements/upgrades have been done, I would doublecheck polarity of parallleling CT's. Have seen them out of polarity with each other numerous times, and don't count on the wire numbers--physically check for an accidental reversal.

 

[wuqn]

i found the problem. droop variable resistor's wire is broken

 

[waross]

Congratulations wugn and thanks for the feed back.
Respectfully