SYNCHRONOUS CONDENSERS
~Synchronous Condensers.~--A synchronous motor when sufficiently
excited will produce a leading current, that is, when over excited
it acts like a great condenser, and when thus operated on circuits
containing induction motors and similar apparatus for the purpose of
improving the power factor it is called a _synchronous condenser_.
Although the motor performs the duty of a condenser it possesses almost
none of the properties of a stationary condenser other than producing
a leading current, and is free from many of the inherent defects of a
stationary condenser.
The relation of power factor to the size and efficiency of prime
movers, generators, conductors, etc., and the value of synchronous
condensers for improving the power factor is generally recognized.
Induction motors and other inductive apparatus take a component of
current which lags behind the line pressure, and thereby lowers the
power factor of the system, while a non-inductive load, such as
incandescent lamps, takes only current in phase with the voltage and
operates at unity power factor.
Since transformers require the magnetizing current, they may seriously
affect the power factor when unloaded or partially loaded, but when
operating at full load their effect is practically negligible.
The relative effect of fully loaded and lightly loaded induction motors
on the power factor is indicated by the diagram, fig. 2,478. The
magnetizing current is nearly constant at all loads and is wattless,
lagging 90 degrees behind the impressed pressure, or at right angles to
the current which is utilized for power.
In the figure, AB is the magnetizing component, which is always
wattless, and CB the power component. The angle ACB gives the
phase relation between voltage and current; the cosine of this
angle CB ÷ AC is the power factor.
[Illustration: FIG. 2,478.--Diagram showing relative effect of fully
loaded and lightly loaded induction motors on power factor.]
It is evident from the diagram that if the load be reduced, the
side CB is shortened, and as AB is practically constant, the
angle of lag ACB is increased. It therefore follows that the
cosine of this angle, or the power factor is reduced.
The figure clearly shows the reason for the low power factor of
induction motors on fractional loads and also shows that since
the magnetizing current is practically constant in value, the
induction motor can never operate at unity power factor.
With no load, the side CB (real power) is just sufficient to
supply the friction and windage. If this be represented by DB,
since AB remains constant, the power factor is reduced to 10
or 15 per cent. and the motor takes from the line about 30 per
cent. of full load current. It therefore follows that a group
of lightly loaded induction motors can take from the system a
large current at exceedingly low power factor.
The synchronous motor when used as a condenser, as before stated,
has the property of altering the phase relation between pressure and
current, the direction and extent of the displacement being dependent
on the field excitation of the condenser.
It can be run at unity power factor and minimum current input, or
it can be over excited and thereby deliver leading current which
compensates for the inductive load on other parts of the system. The
synchronous condenser, therefore, can supply magnetizing current to
the load on a system while the power component is supplied by the
generators.
[Illustration: FIG. 2,479.--General Electric 400 kw., 550 volt, 600
R.P.M., ~synchronous condenser~ with direct connected exciter installed
in sub-station No. 1 of the Colorado Light & Power Co., Cripple Creek,
Colo. The machine is designed for alternating current starting by means
of a compensator. The field is provided with a standard synchronous
motor winding, and, in addition, an amortisseur winding which assists
in starting and serves as a damping device to minimize hunting.]
[Illustration: FIG. 2,480.--Diagram showing relative location of
alternators and synchronous motors in plant of Witherbee Sherman & Co.,
Mineville N. Y. The distribution system of the Company is provided
with three synchronous motors, as shown. The system includes two
hydro-electric, one turbine driven, and one engine driven generator
plants; from three of these, current is transmitted to the fourth,
which is located in Mineville, at the point "A", the current being
distributed to the motor circuits from the points "A" and "B." The
transmission to the central station at Mineville is over three phase
circuits at 6,600 volts. For operating the mine at Cheever, current
is transmitted direct from the generating station at Port Henry. The
distribution from "A" and "B" is all at 3,300 volts, being stepped down
to 440 volts for the operation of the motors, which have a total rated
capacity of 4,762 horse power. Excepting three synchronous motors,
the load is practically all inductive, there being less than 10 kw.
required for lighting. The actual power demand ranges from 60 to 65
per cent. of the rated motor capacity, and prior to the installation
of the synchronous motors, the power factor was approximately 68 per
cent., the condenser effect of these motors making it possible to
maintain an average of about 90 per cent. power factor in spite of
the fact that a considerable portion of the induction motor load is
very widely distributed. The three synchronous motors are partially
loaded, each motor driving an air compressor through belting. The 180
kva. motor at Cheever takes about 150 kw. for the operation of a 1,250
cubic foot compressor, while the two 360 kva. machines take about 300
kw. each, for the operation of two 2,500 cu. ft. sets. The operation
of these compressors affords a method of utilizing a portion of the
motor capacity mechanically, inasmuch as the load on the motors is
practically constant during the time the mines are in operation, and
thereby permit the motors to be run at approximately 80 per cent. power
factor.]
~Effects of Low Lagging Power Factors.~--Transformers are rated in kva.
output; that is, a 100 kva. transformer is supposed to deliver 100 kw.
at unity power factor at normal voltage and at normal temperatures;
but, if the power factor should be, say .6 lagging, the rated energy
output of the transformer would be only 60 kw. and yet the current
and, consequently, the heating would be approximately the same as when
delivering 100 kw. at unity power factor.
[Illustration: FIG. 2,481.--Field of a synchronous condenser. Note
the ~amortisseur winding~, _erroneously_ called ~_squirrel cage_~
~winding~, consisting of two end rings which serve to short circuit
spokes passing through the pole tips as shown. ~The amortisseur
winding~ _assists in_ ~starting~ _and serves also as a damping device
to minimize_ ~hunting~.]
The regulation of transformers is inherently good, being for small
lighting transformers about 1½ to 2 per cent. for a load of unity
power factor, and about 4 to 5 per cent. at .7 power factor. Larger
transformers with a regulation of 1 per cent. or better at a unity
power factor load, would have about 3 per cent. regulation at .7 power
factor.
Alternators also are rated in kva. output, usually at any value of
power factor between unity and .8.
The deleterious effects of low power factor loads on alternators
are even more marked than on transformers. These are, decreased kw.
capacity, the necessity for increased exciter capacity, decreased
efficiency, and impaired regulation.
Assume the case of a 100 kva. .6 power factor, 60 kw. output.
It is probable that normal voltage could be obtained only with
difficulty, unless the alternator was especially designed for
low power factor service. The lagging power factor current
in the armature sets up a flux which opposes the flux set up
by the fields, and in consequence tends to demagnetize them,
resulting in low armature voltage.
It is often impracticable, without the installation of new
exciters, to raise the alternator voltage by a further increase
of the exciting voltage and current. The field losses, and
therefore the field heating of the alternator, when it is
delivering rated voltage and current, are greater at lagging
power factor than at unity. Increased energy input and
decreased energy output both cause a reduction in efficiency.
[Illustration: FIG. 2,482.--Diagram of a section of the Northern
California Power Co.'s transmission system, showing relative location
of alternators and synchronous condenser. The synchronous condenser is
installed at Kennett, which is served by generating stations at Kilarc
and Volta, located respectively 28 and 38 miles from the point at which
the condenser is operated. The local demand amounts to about 6,500
kw., and before the installation of the synchronous condenser, the
power factor was about 79 per cent. and after installing, about 96 per
cent. while the voltage at the point where the synchronous condenser
is installed is raised approximately 10 per cent. during the change
from no load to full load. In order to obtain close voltage regulation,
a regulator is used in connection with the synchronous condenser
and holds the voltage, at the center of distribution, within 2 per
cent. The regulator is mounted on the side of the control panel and
connected in the field of the synchronous condenser to automatically
change the excitation and compensate for voltage variations. A graphic
demonstration of the improvement in voltage regulation, which has been
secured in this case, is given by the curve drawing voltmeter records
reproduced in fig. 2,483.]
The regulation at unity power factor of modern alternators capable of
carrying 25 per cent. overload, is usually about 8 per cent. Their
regulation at .7 power factor lagging is about 25 per cent. The effect
of low power factor on the lines can best be shown by the following
example:
EXAMPLE.--Assuming a distance of five miles and a load of 1,000
kw. and desiring to deliver this load at a pressure of about
6,000 volts, three phase, with an energy loss of 10 per cent.,
each conductor at unity power factor would have to be 79,200
c.m., at .9 power factor, 97,533 c.m., and at .6 power factor,
218,000 c.m. In other words, at the lower power factor of .6,
the investment in copper alone would be 2.8 times as much.
[Illustration: FIG. 2,483.--Curve drawing ~voltmeter records~ at
Kennett, Cal. ~The upper curve~ shows voltage regulation with
synchronous condenser out of service, and ~the lower curve~, with
synchronous condenser in operation.]
If the same size of wire were used at both unity and .6 power
factor lagging, the energy loss would be about 2.8 times
the loss at unity power factor, or about 28 per cent. Low
lagging power factor on a system, therefore, will generally
mean limited output of prime movers; greatly reduced kilowatt
capacity of generator, transformer and line; and increased
energy losses. The regulation of the entire system will also be
poor.
~Cost of Synchronous Condenser vs. Cost of Copper.~--Referring to the
example given in the preceding paragraph, and calculating the necessary
extra investment in copper with the .6 power factor load, and copper
at 17 cents per pound, the result is that 29,292 pounds more copper is
required than with the power factor of .9 which means a total extra
investment in copper alone of $5,000 (29,292 × $.17). A synchronous
condenser of sufficient capacity to accomplish the same result would
cost about the same amount. It would therefore cost less to install the
condenser because at the same time a considerably increased capacity
would be obtained from the alternators, transformers, etc.
[Illustration: FIG. 2,484.--Diagram showing the ~field current taken
by a synchronous motor~ of normal design when operating at normal kva.
input at various power factors. ~It will be noted~ that a ~slight
departure from unity power factor necessitates a considerable change
in field current~. As the field curves increase with the square of the
current, there is a rapid increase in temperature with leading current.
_This action of leading or lagging current serves automatically to keep
the flux constant in the armature with changes in field excitation._
~When the motor is running at unity power factor~, an increase in field
excitation causes a leading current to flow, and at the same time
this leading current demagnetizes the field until the density of the
armature is restored to its normal value. ~If the field be decreased~
a lagging current flows which in turn magnetizes the field bringing
the density back to its original value. Therefore, with a constant
line voltage, the iron losses in a synchronous motor are approximately
constant irrespective of the field excitations with the exception that
the internal voltage will vary slightly due to the armature I R drop,
the density being a trifle lower at full load than at no load.]
~Synchronous Condenser Calculations.~--In figuring on the installation
of a condenser for correcting power factor troubles, a careful survey
of the conditions should be made with a view of determining just what
these troubles are and to what extent they can be remedied by the
presence of a leading current in the system.
[Illustration: FIG. 2,485.--Diagram showing a set of ~phase
characteristic curves~ taken ~from~ a General Electric _synchronous
motor_. ~These curves show~ _the current input to the motor at various
loads with constant voltage and varying field excitation_. There is a
certain field current at each load that causes a minimum current. Any
increase or decrease of field from the value increases the current and
causes it to lead or lag with respect to the line voltage. By referring
to the _minimum input_ curve, it will be noted that if the machine
be running at full load minimum input current and load is taken off,
the current will be leading or vice versa. In each case the phase
characteristic curve was run back on the lagging side to the break down
point. ~At no load~ and one quarter load the motor still ran in step
when the field was reduced to zero and even taken off altogether, and
it was necessary to reverse the field current in order to back down
the motor. ~The motor runs without slip~, as a synchronous motor, in
this condition, obtaining its excitation from the lagging current and
running as a reaction machine. The amount of load a machine will carry
without field varies with the design, the average being about 40% of
full load. ~It will be noted~ from the _limit of stability_ curve that
the lighter the load on the machine when it breaks down from lack of
sufficient excitation, the greater the current input at this point.
~The no load characteristic~ rises sharply on each side with slight
change in field current, while it flattens out with increase in load
until at overload the current input is practically the same throughout
a large range of field current.]
[Illustration: FIG. 2,486.--Comparison of the ~speed current curves~
and ~speed power factor curves~ of a typical synchronous, and induction
motor. It will be noted that the power factor of the synchronous motor
at start is higher than that of the induction motor owing to the higher
resistance of the squirrel cage winding. ~As the machine approaches
synchronism~, however, the magnetizing current of the induction motor
drops to a very much lower value than in the synchronous motor and
the power factor is consequently much higher. The magnetizing current
of the induction motor at full speed is usually 25 per cent. of full
load current while that of the synchronous motor is from 200 to 250
per cent. of full current, or even higher when running full speed
and normal voltage. This of course is due to the large air gap on
the synchronous machine. ~The current at start~ with full voltage
applied is usually higher in an induction motor owing to the fact
that the total impedance of the stator and rotor are less due to the
greater distribution of the windings and the lower resistance of the
squirrel cage. ~The high magnetizing current~ of a synchronous motor
should not be lost sight of as it is a very important consideration
in starting the machine. Even though the motor can be brought
practically to synchronous speed while still on the compensator, if
line voltage be thrown on, there will be a very heavy rush of current.
The obvious thing to do is to get the field on the motor while still
on the compensator, whenever possible, to avoid the high magnetizing
current. This magnetizing current is obviously equal to the circuit
current of the machine at no load field. In some cases additional
torque near synchronism can be obtained by short circuiting the field
winding through the field rheostat. This has the effect of reducing
the resistance of the rotor winding to some extent and causing the
motor to have less slip with a given load. The gain from this source
is small, however, in most cases, as the self-inductance of the field
winding is so high as to allow very little current to flow even if the
field be short circuited so that the total effective resistance of the
rotor winding is not materially reduced. In some cases where the torque
is nearly sufficient, however, enough gain may be obtained to take
care of the conditions. ~If the field be short circuited~ before the
motor is started there will be a reduction in starting torque and an
increase in current from the line, hence if this method be resorted to,
arrangements must be made to short circuit the field after the motor
has come to constant speed.]
It is necessary to possess a thorough knowledge of the system, covering
the generating capacity in energy and kva., average and maximum load,
and power factor on the alternators, average and maximum load, and
power factor on the feeders, system of distribution, etc.
[Illustration: FIG. 2,487.--Curves showing amount of wattless component
required to raise the power factor of a given kw. load to required
higher value. The wattless components are expressed as percentages of
the original kw. load. ~The numbers at the right~ which indicate the
points of tangency of the power factor curves to the 100 per cent.
line, show the amount of wattless component required to raise a given
kw. load of given lagging power factor to unity power factor. Obviously
the addition of further wattless component in a given case would result
in a leading power factor less than unity.]
The desirable location of a condenser is, of course, nearest the
inductive load in order to avoid the transmission of the wattless
current, but it often happens that a system is so interconnected that
one large condenser cannot economically meet the conditions, in which
case it may be better to install two or more smaller ones.
The question of suitable attendance should also be considered and, for
this reason, it may be necessary to compromise on the location. When
the location of the condenser has been decided upon and the load and
power factor within its zone determined, the proper size of condenser
to raise the power factor to a given value can be found as follows:
The method of procedure can best be explained by reference to
a concrete case. Assume a load of 450 kw. at .65 power factor.
It is desired to raise the power factor to .9. What will be the
rating of the condenser?
[Illustration: FIG. 2,488.--Diagram for synchronous condenser
calculations.]
Referring to the diagram, fig. 2,488, it is necessary to start
with 450 kw. At .65 power factor, or 692 kva., this has a
wattless lagging component of √(692² - 450²) = 525 kva. With
the load unchanged and the power factor raised to .9, there
will be 500 apparent kva., which will have a wattless component
of √(500² - 450²) = 218 kva.
It is obvious that the condenser must supply the difference
between 525 kva. and 218 kva., or 307 kva. A 300 kva. condenser
would, therefore, meet the requirements.
If it be desired to drive some energy load with the condenser
and still bring the total power factor to .9, proceed as
indicated in fig. 2,489. Assume a total load of 150 kw. on the
motor. As before, 450 kw. at .65 power factor, or 692 kva.,
with a wattless component of 525 kva.
The energy load will be increased from 450 to 600 kw. as
indicated, and with the power factor raised to .9 there will be
a kva. of 667 with a wattless component of √(667² - 600²) = 291.
There must be supplied 525 - 291 = 234 in leading kva.
The synchronous motor then must supply 150 kw. energy
and 234 kva. wattless, which would give it a rating of
√(150² + 234²) = 278 kva. at .68 power factor.
[Illustration: FIG. 2,489.--Diagram for synchronous condenser
calculation for cases where it is desired to drive some energy load
with the condenser and still bring the total power factor to .9.]
The standard 300 kva. condenser would evidently raise the power
factor slightly above .9 power factor leading.
By reference to the chart, fig. 2,490, the size of the required
condenser can be obtained direct without the use of the above
calculation. The method of using this curve is as follows:
Assume a load of say 3,000 kw. at .7 power factor and that it
be desired to raise the power factor to .9. Run up the vertical
line at 3,000 kw. to the .7 power factor line, and from there
along the horizontal line to the margin and find a wattless
component at this power factor of 3,000 kva., approximately.
Again run up the 3,000 kw. vertical line to the .9 power factor
line and from there along the horizontal line to the margin
and find a wattless component of 1,500 kva. The rating of the
condenser will then be 3,000 kva. - 1,500 kva. = 1,500 kva.
This table of course can be used for hundreds of kilowatts as
well.
[Illustration: FIG. 2,490.--Curve showing the relation of energy load
to apparent load and wattless components at different power factors.]
For determining the rating of a synchronous motor to drive an
energy load this curve is not so valuable, although it can
be used in determining the wattless component direct in all
cases where the energy component and power factor are known.
Knowing this energy component and power factor or wattless
component, the energy load can obviously be found by referring
to the curved lines on the diagrams, the curve that crosses the
junction of the vertical energy line and the power factor or
wattless component line giving the total apparent kva.