normally alternators are not run in leading power factor. one argument is that armature current will increase in leading kvar area that it may damage end rings. But as seen from V-curves armature current will increase in either direction of unity power factor, so why armature current increases more in leading kvar side? Another argument is slipping of pole, but again that is true for both leading and lagging power factor. My question is then why alternator is not recommended to run in leading kvar or power factor side?

 

Hopefully Phil Corso will weigh in on this for a more factual explanation!

There is reaction between the two magnetic fields which exist in any electrical machine. In an alternator, the reaction is between the field produced by the flow of DC current in the rotor and the magnetic field produced by the flow of AC current in the windings of the armature (stator).

As excitation (DC field current) is reduced ("collapses") during a leading VAR/power factor condition, the armature field "expands". This causes uneven flux distributions. The increased leading reactive current in the armature increases the heat in the end-turns (end-iron) of the armture windings.

The reactive capability curve of any alternator has three distinct "curves": each of them represent limits. The curve that is parallel to the Watt axis on the positive VAR side represents the limit of the ability to cool the generator rotor when "over-excited" to produce VAR (increased DC currrent in the rotor windings=increased heat).

The curve that is parallel to the VAR axis on the positive Watt side represents the limit of the ability to cool the armature windings (increased current in the armature windings=increased heat).

The curve that is parallel to the Watt axis but on the negative VAR side represents the limit of the ability to cool the end-turns/end iron, when reactive current is flowing and there is no reduced magnetic field from the rotor.

The inability to cool the end-turns/end iron is the major reason for not operating most alternators with a "large" leading VAR flow/power factor.

Is it important to run your alternaor(s) in a leading power factor configuration? Sometimes the manufacturer of the generator can provide some guidance or support for such operation; have you contacted the manufacturer for information or assistance?

There are large AC machines built to be operated in leading power factor conditions: synchronous condensers.

One of the better textbooks on electrical machinery is Charles I. Hubert's "Electrical Machines..." It provides some very good descriptions of machine design and operationg considerations.

markvguy

 

Responding to petrosumit's Apr 11, 9:56pm post... the "V" curve graph of armature-current vs field-current certainly is a measure of a generator's capability, but at best it is a "coarse" approximation.

Your argument for operational symmetry in both the lagging and leading modes has some validity, in theory. But, actual capability is generator-specific. A more meaningful relationship can be better illustrated with an Operating Capability Chart. In such a chart, active power, kW, is the abscissa (y-axis), reactive power kVAr, is the ordinate (x-axis), and power-factor the radii emanating from the kW=0, kVAr=0 or x-y axis intersection. Furthermore, because the machine is an alternator, plotting is limited to quadrants III and IV, where III represents laggng kVAr, while IV represents leading kVAr.

The resultant curve approaches that of a semi-circle for the theoretical, lossless, machine. However, for the "practical" machine, neglecting differences in construction (cylindrical or salient-pole), as well as operation (isolated or parallel with others) the curve is distorted due to the following:

a) The active power-line is fixed by the prime-mover, and the rated pf.

b) The lagging-kVAr portion of the curve is altered by limits imposed by field-heating.

c) The leading kVAr portion of the curve is altered by limits imposed by stator-heating, and by transient stability restrictions.

In conclusion, the resulting asymmetry in the Operating Capability Chart is more debilitating for leading-kVAr operation.

Your statement about "end-rings" is unclear. Do you mean the retaining cover associated with cylindrical-type rotor? Or the end-windings associated with the stator?

Regards, Phil Corso, PE {Boca Raton, FL, USA} [[email protected]] ([email protected])

 

Responding to markvguy's Apr 14, 12:35am request for a more factual explanation:

By now you will have noticed that my earlier response to petrosumit's post, covered the development of an Operating Capacity Chart (sometimes called a Performance Curve). This post addresses your request for a more factual explanation.

First, addressing the perceived attention given to armature end-connections. Stator current increases. However, the end-connection temperature-rise is much higher than the slot component, thereby dangerously increasing its length with dire consequences. It's not the absolute temperature that's the problem it's the increase in length.

Regarding a factual explanation. I realize that the Performance Chart may have been too much. Hopefully, the following will be simpler to the eye-of-the-reader:

Case 1: Parallel, Equal Division. Consider identical machines, A & B, equally sharing a common inductive load. Their terminal voltages, armature currents, and field currents are the same. Now increase A's excitation while simultaneously decreasing B's. Terminal voltage remains the same, and power output is still shared equally. Because A is over-excited its armature-current lags, opposing its field-flux. Simultaneously, B is under-excited so its armature-current, leads, aiding its field-flux. Each machine produces a new reactive current which increases A's armature-current, but reduces B's.

The result is that A's armature-current is greater than machine B's. And, because delivered power is the same the internal generated voltage of each machine changes. But, because the load is inductive its lagging-current is delivered only by machine A which is considerably larger than the leading-current of B.

In conclusion, such operation is not deleterious to the leading-current machine, but could be to the lagging one.

Case 2) Parallel, Uneven Division. Same initial conditions as Case 1, except to better illustrate the leading-current phenomena, assume the load is non-inductive, i.e., pf=unity. Now reduce A's power input without changing excitation. The internal generated voltage changes, but because of its synchronous impedance drop its armature-current reduces. And, because the drop is inductive, lagging-current is produced.

Now, because A's conditions have changed, B's conditions must also change to offset A's. Hence, B's armature-current becomes much greater in magnitude, and leading in nature.

In conclusion, such operation is quite deleterious to the leading-current machine.

Regards, Phil Corso, PE {Boca Raton, FL, USA} [[email protected]] ([email protected])

 

The reference to end-iron and end-turns was in regards to stator (armature) end iron and statur winding end turns, not generator winding end-turns or generator rotor retaining rings.

Can a pole be slipped very easily in an over-excited condition? It would seem that reducing excitation--field strength--would increase the chances of slipping a pole much more than increasing excitation (field strength).

markvguy

 

Understood your point. In my case if alternator is running (0.8 pf lag) in parallel with grid and we reduce excitation of generator then it starts going towards unity power factor and if further reduce excitation it starts operating in leading power factor (load of the generator is held constant) then according to performance curve we cannot go in leading side beyond 0.9 pf lead, taking into account practible stability limit. this point you have mentioned in your reasoning also. So my question is which factor defines this stability limit and how armature current limiting region reaches, even though we are operating (at 0.9pf lead) much below rated stator current.

 

Responding to Markvguy's Apr 19, 11:41 pm comment... to whom was your post addressed? If for me, then:

1) The end-turn (end-ring) diatribe in my Apr 16, 8:14 pm post was intended to dispel the perception that end-turn damage was the only concern. (See thread # 1026220640 "ENGR: reverse reactive power") The question posed in my Apr 15, 10:06 pm was to Petrosumit.

2) A pole can not slip very easily in the overexcited case. I am unaware of any ANSI protective device to cover the situation.

Regards,
Phil Corso, PE {Boca Raton, FL, USA}
[[email protected]] ([email protected])

 

Responding to Petrosumit's Apr 19, 11:39 pm question:

The stability-limit line is determined by calculating the minimum load-increment that will cause the machine to lose synchronism, i.e, for each correspnding kW and kVAr values. The allowable kVAr range is "usually" larger for a salient-pole machine than a cylindrical-rotor machine.

Regards,
Phil Corso, PE {Boca Raton, FL, USA}
[[email protected]] ([email protected])

 

If your question is understood correctly, the defining factor--limit--of operation of an alternator in a leading power factor condition is the ability to cool the armature (stator) end turns, where the increases reactive power flowing during leading power factor operation concentrates. The manufacturer determines these limits, and expesses them using the reactive capability curve for each generator.

The three limits expressed by a reactive capability curve are all heat-related limits: the ability to cool the generator stator windings when predominantly pure watts are flowing out of the generator; the ability to cool the rotor when operating in an overexcited condition; the ability to cool the armature end turns when operating in an underexcited condition.

If you are looking for some "physical" factor which can be monitored, there are two: the power factor represents the ratio of "active power" (pure watts) to "apparent power" (the combination of active and reactive power): pf = kw / kva, and is one of the two factors which can be derived from looking at the reactive capability curve. The other limit is the magnitude and "polarity" of reactive power (vars, or kvar) for a given watt, or kw, load.

Perhaps the confusion of looking at the V-curves stems from the fact that the V-curves represent total current flowing in the armature--the combination of current related to pure power (watts, kw) and reactive power (vars, kvars), or, what is sometimes referred to as "apparent power" (va, or kva). VA, or KVA, is the combination of active and reactive power. So, kva increases when operating in either an overexcited or an underexcited condition due to the increase in the current flow of the reactive power component. V-curves don't just represent pure watts, or only the active power current; they represent the total current--both active and reactive, or the "apparent" current.

markvguy

 

Responding to Petrosumit's Apr 19, 11:40pm conditional question:

You stated stator current is reduced after the excitation decrease! Therefore, the kW or MW, must have also reduced.

Q1) Is the above correct? And,

Q2) What does "much below" mean in per-unit or percent of rated value?

Regards,
Phil Corso, PE {Boca Raton, FL, USA}
[[email protected]] ([email protected])

 

Try this link, the chapter on Synchronous machines. This is one of the best documents this author has found on the web--leave it to the Canadians!

http://canteach.candu.org/library/20030801.pdf

This document also information on v-curves, heating limits--even droop speed control--and on and on. And it's all in a very easy-to-understand format with lots of great graphics.

If you go to http://canteach.candu.org there is lots of additional information on power plant operation.

markvguy