D.C.

Motor Characteristics
Contents:

1. Preface
2. Background From Physics
o 2.1 Torque
o 2.2 Speed
o 2.3 Power
3. Motor Characteristics
o 3.1 Torque/Speed Curves
o 3.2 Power Curves

1. Preface

This site deals with Direct Current permanent magnet motors operated at a constant
voltage. Motor characteristics vary considerably from type to type, and their performance
characteristics can be altered by the way electrical power is supplied. can be quite different
than those covered here.
 2. Background From Physics

Section 2.1: TORQUE

Torque , as defined in 1998 Merriam-Webster, Incorporated

Main Entry: torque

Function: noun
Etymology: Latin torquEre to twist
1 : a force that produces or tends to produce rotation or torsion (an automobile engine
delivers torque to the drive shaft);
also : a measure of the effectiveness of such a force that consists of the product of the force
and the perpendicular distance from the line of action of the force to the axis of rotation
2 : a turning or twisting force

Torque , as defined in University Physics, 8th ed. 1992, by Hugh D. Young:

The quantitative measure of the tendency of a force to cause or change rotational motion is
called torque.

Torque (also called a moment) is the term we use when we talk about forces that act in a
rotational manner. You apply a torque or moment when you turn a dial, flip a lightshwitch,
drill a hole or tighten a screw or bolt.

As shown in the picture of a ratchet, a torque is created by a vertical force applied at the
end of the handle. The force, F, applied to the ratchet as shown causes a tendency to rotate
about point O. The force can be broken down into two components: a radial component,
Frad, parallel to the ratchet handle that does not contribute to the torque, and a tangential
component, Ftan, perpendicular to the handle that does contribute to the torque. The distance
from point O to the point of action of F is described by the direction vector, r. The moment
arm, l is the perpendicular distance between point O and the line of action of F.

If we were to shorten the moment arm by applying the force closer to the head of the
ratchet,the magnitude of the torque would decrease, even if the force remained the same.
Thus, if we change the effective length of the handle, we change the torque (see equation
1).
 UNITS of TORQUE
SI English
newton-meters inch-pounds
{Nm} {inlb}
foot-pounds
{ftlb}
inch-ounces
{inoz}
1 Nm = 0.738 1 inlb = 0.113
ftlb Nm
1 Nm = 0.113 1 ftlb = 1.356
inlb Nm
1 Nm = 141.61 1 inoz = 7.062E-
inoz 03 Nm
Section 2.2: SPEED

Speed (Angular Velocity) , as defined in 1998 Merriam-Webster, Incorporated

Main Entry: angular velocity

Function: noun
he rate of rotation around an axis usually expressed in radians or revolutions per second
or per minute

Motors are devices that convert electrical energy into mechanical energy. The D.C. motors
that we have been dealing with here convert electrical energy into rotational energy. That
rotational energy is then used to lift things, propel things, turn things, etc... When we
supply the specified voltage to a motor, it rotates the output shaft at some speed. This
rotational speed or angular velocity, is typically measured in radians/second {rad/s},
revolutions/second {rps}, or revolutions/minute {rpm}.

o When performing calculations, be sure to use consistent units. In the English

system, calculations should be done in degrees/second, and radians/sec for SI
calculations.

NOTE:

1 revolution = 360
1 revolution = (2* ) radians
1 radian = (180/ )
1 = ( /180) radians

From the angular velocity, , we can find the tangential velocity

of a point anywhere on the rotating body through the equation
tangential velocity, v = r* , where r is the distance from the axis
of rotation. This relation can be used to compute the steady state
(constant speed - no acceleration) speed of a vehicle if the radius
and angular velocity of a wheel is known, or the linear speed of a
rope as it is wound up by a winch.
Section 2.3: POWER

Motive Power , as defined in 1998 Merriam-Webster, Incorporated

Main Entry: 1power

Pronunciation: 'pau(-&)r
Function: noun
Usage: often attributive
Etymology: Middle English, from Old French poeir, from poeir to be able, from (assumed)
Vulgar Latin potEre, alteration of Latin posse
Date: 13th century
1 a : (1) : ability to act or produce an effect
6 a : a source or means of supplying energy; especially : ELECTRICITY
6 b : MOTIVE POWER c : the time rate at which work is done or energy emitted or
transferred

Power in Rotational Motion , as described in University Physics, 8th ed., 1992 by Hugh D.
Young):

When you pedal a bicycle, you apply forces to a rotating body and do work on it. Similar
things happen in real-life situations, such as a rotating motor shaft driving a power tool or
a car engine propelling the vehicle. We can express this work in terms of torque and an
angular displacement...
What about the power associated with work done by a torque acting on a rotating body?

dW/dt is the rate of doing work, or power P. When a torque T (with respect to the axis of
rotation) acts on a body that rotates with angular velocity W, its power (rate of doing work)
is the product of the torque and angular velocity. This is the analog of the relation P = Fv
for particle motion.

Power in rotational motion can be written as:

UNITS of POWER
 SI English
foot-pounds per second {ftlb/s}
Watts {W} horsepower {hp}
newton-meters per second {Nm/s}
1 ftlb/s = 1.818E-03 hp
1 W = 1 Nm/s 1 ftlb/s = 1.356 W
1 W = 0.738 ftlb/s
1 W = 1.341E-03 hp

2. Motor Characteristics
Section 3.1: TORQUE/SPEED CURVES

In order to effectively design with D.C. motors, it is necessary to understand their

characteristic curves. For every motor, there is a specific Torque/Speed curve and Power
curve.

The graph above shows a torque/speed curve of a typical D.C. motor. Note that torque is
inversely proportioal to the speed of the output shaft. In other words, there is a tradeoff
between how much torque a motor delivers, and how fast the output shaft spins. Motor
characteristics are frequently given as two points on this graph:

The stall torque, , represents the point on the graph at which the torque is a
maximum, but the shaft is not rotating.
The no load speed, , is the maximum output speed of the motor (when no torque
is applied to the output shaft).

The curve is then approximated by connecting these two points with a line, whose equation
can be written in terms of torque or angular velocity as equations 3) and 4):
The linear model of a D.C. motor torque/speed curve is a very good
approximation. The torque/speed curves shown below are actual
curves for the green maxon motor (pictured at right) used by
students in 2.007. One is a plot of empirical data, and the other was
plotted mechanically using a device developed at MIT. Note that
the characteristic torque/speed curve for this motor is quite linear.

This is generally true as long as the curve represents the direct

output of the motor, or a simple gear reduced output. If the
specifications are given as two points, it is safe to assume a linear
curve.

Recall that earlier we defined power as the product of torque and angular velocity. This
corresponds to the area of a rectangle under the torque/speed curve with one cornerat the
origin and another corner at a point on the curve (see figures below). Due to the linear
inverse relationship between torque and speed, the maximum power occurs at the point
where = , and = .
Section 3.2: POWER/TORQUE and POWER/SPEED CURVES

By substituting equations 3. and 4. (torque and speed, section 2.1) into equation 2. (power,
section 1.3), we see that the power curves for a D.C. motor with respect to both speed and
torque are quadratics, as shown in equations 5. and 6.

From these equations, we again find that maximum output power occurs at = , and
= repectively.
Torque Speed Curve
General Curve
Using the induction machine power and torque equations it is possible to produce the
torque speed curve shown below.

The torque speed curve is approximately symmetric about synchronous speed. (The
magnitude of the peak negative torque is larger than the positive torque, the difference
being a function of stator resistance)
 Operating Regions
The torque-speed curve brakes down into three operating regions:

1. Braking, n m <0 , s>1

Torque is positive whilst speed is negative. Considering the power conversion equation

P conv =(1s)P gap

it can be seen that if the power converted is negative (from P= ) then the airgap power is
positive. i.e. the power is flowing from the stator to the rotor and also into the rotor from
the mechanical system. This operations is also called plugging.
This mode of operation can be used to quickly stop a machine. If a motor is travelling
forwards it can be stopped by interchanging the connections to two of the three phases.
Switching two phases has the result of changing the direction of motion of the stator
magnetic field, effectively putting the machine into braking mode in the opposite direction.

2. Motoring, 0<n m <n s , 0<s<1

Torque and motion are in the same direction. This is the most common mode of operation.
3. Generating,n m >n s , s<0
In this mode, again torque is positive whilst speed is negative. However, unlike plugging,

P conv =(1s)P gap

indicates that if the power converted is negative, so is the air gap power. In this case, power
flows from the mechanical system, to the rotor circuit, then across the air gap to the stator
circuit and external electrical system.

The torque equation

Using the equation
=3I 2 2 R 2 s s

the torque speed curve can be solved by first solving the circuit model to find the rotor
current. In the absence of a computer, this is a tedious process at best. Multiple solutions of
the above equation for torque at different slips can be made simpler by simplifying the
equivalent circuit model. Consider the diagram below:
The stator part of the equivalent circuit (together with the magnetising branch) can be
replaced by a Thevenin equivalent circuit. In the Thevenin circuit, the stator phase voltage
has been replaced by its Thevenin equivalent,
V TH V TH =jX m R 1 +j(X 1 +X m ) V 1 =|V TH |

and the impedances have been replaced by Thevenin equivalent impedances.

Z TH Z TH Z TH =(R 1 +jX 1 )||jX m =jX m (R 1 +jX 1 )R 1 +jX 1 +jX m =R TH +jX T
H

Incorporating the Thevenin model into the circuit model results in the Thevenin equivalent
circuit model of an induction machine.

In the above circuit, the calculation of rotor current is greatly simplified

I 2 I 2 =V TH Z TH +Z 2 =V TH (R TH +R 2 s )+j(X TH +X 2 )

The above expression for rotor current can be squared and substituted into the torque
equation
I 2 2 =V 2 TH (R TH +R 2 s ) 2 +(X TH +X 2 ) 2 =3V 2 TH (R TH +R 2 s ) 2 +(X TH
+X 2 ) 2 R 2 s s

Using the above equation, the variation of torque with slip can be plotted directly .
Note that if power or efficiency calculations are needed, the full equivalent circuit model
should be used (not the Thevenin version).

Motoring Torque Characteristic

The motoring region of the induction machine torque-speed curve is the region of greatest
interest. The plots below show a number of different torque speed curves, due to
differences in the motor designs. Common features of interest are noted and discussed
below.
 rated . The rated torque of the machine. This is the design operating point.
start . The start torque of the machine, when the machine is at standstill.
max or po . The maximum torque or pull-out torque. Once a machine has reached rated
operating point, this is the maximum torque that can be applied without stopping the
machine (pulling out). In reality, since this torque is significantly more than the design
rated torque, operation at maximum torque is not possible due to thermal heating issues.
(Currents will be above rated values, I2R losses will be excessive.)
pu . The pull-up torque of the machine. In some machines, the lowest point on the torque
speed curve between starting and pullout is not the start torque. In this case it is important
to know the pull-up torque. This is the minimum torque that the motor can accelerate up to
the desired operating speed.

In steady state operation, an induction motor will operate at the speed where the mechanical
load torque equals the torque developed by the motor. At low speeds, the difference
between the motor torque and the load torque accelerates the machine. Normal operation is
to the right of the maximum torque. In this region, an increase in the load torque will cause
the motor to slow, increasing the motor torque until an equilibrium is reached. To the left of
the pullout torque, no such equilibrium can be reached. Mechanically,
motor load =Jd m dt

where J is the rotational inertia of the mechanical system.

 Equivalent Circuit Model Analysis Example
A 480V, 60 Hz, 6-pole, three-phase, delta-connected induction motor has the following
parameters:
R1=0.461 , R2=0.258 , X1=0.507 , X2=0.309 , Xm=30.74
Rotational losses are 2450W. The motor drives a mechanical load at a speed of 1170 rpm.
Calculate the following information:

i. Synchronous speed in rpm

ii. slip
iii. Line Current
iv. Input Power
v. Airgap Power
vi. Torque Developed
vii. Output Power in Hp
viii. Efficiency

Comments
This machine has no iron loss resistance, so the equivalent circuit is as follows:

i. Synchronous speed is given by:

Therefore

ns = 1200 rpm

ii. Slip is given by

Using the rpm equation,

s = (1200-1170)/1200 = 0.025

iii. Now, phase current is given by

 where phase impedance is given by

Using the above equation, Zin = 9.57 + j3.84

And noting that the machine is delta connected, V1 = VLL = 480V

I1 = 43.1 - j17.4 A. |I1| =46.6 A, = -21.9

Therefore IL = 3 46.6 = 80.6 A

iv. Input power is given by:

Therefore:

Pin = 62.2 kW

v. To find airgap power, There are two possible approaches:

a. Airgap power is the input power minus stator losses. In this case the core losses are
grouped with rotational loss. Therefore

Pgap = 62.2 kW - 3 46.62 0.461

Pgap = 59.2 kW

b. Airgap Power is given by

This approach requires rotor current to be found. With no core loss resistance:

Giving I2 = 43.7 A. Substituting into the power equation

 Pgap = 59.2kW

vi. Torque developed can be found from

where synchronous speed in radians per second is given by

giving

= 471 Nm

vii. Output power in horsepower is the output power in Watts divided by 746. (there are 746 W
in one Hp).

and

Therefore output power in Watts is:Pout = 55.3kW

Pout = 74.1 Hp

viii. Efficiency is given by

Therefore

= 55.3/62.2 = 88.9%
 Thevenin Circuit Model Analysis Example
For the machine in the previous example, find:

i. Thevenin circuit parameters and Thevenin voltage

ii. Pullout slip
iii. Pullout Torque
iv. Start Torque

Using Matlab or Excel (or another computer program) plot the torque speed curve for slip
in the range 0 to 1

i. Thevenin circuit parameters and Thevenin voltage:

Thevenin circuit parameters an voltage can be found using the equations provided on the
formula sheet, or from first principles. The Thevenin voltage is the voltage applied to the
rotor assuming that the rotor current is zero. Thevenin impedance is the impedance of the
stator part of the circuit, seen from the rotor, assuming that the stator supply is short
circuited.

Substituting the equivalent circuit parameters in to the above equations gives:

VTH = 475.2 V, RTH = 0.452, XTH = 0.313

ii. Pullout slip

The slip at which maximum torque occurs can be found from maximum power transfer
theory. Maximum torque and maximum airgap power occur at the same slip, therefore
maximum torque occurs when

iii. Pullout Torque

 Pullout torque can be found by substituting the above pullout slip into the Thevenin torque
equation

or from the maximum torque equation directly

Substituting into the above equation:

iv. Start Torque

Start torque can be found by setting s=1 in the above equation for torque.

Either Matlab or Excel can be used to produce a torque speed plot with fairly minimal
effort. The file thevenin_torque_speed.m is a script which may be run to produce the
Matlab plot below
The file excel_trq_speed_example.xls contains an example of how to produce an Excel plot
of the curve (shown below). Note that this file uses complex arithmetic, which is part of the
Excel "Analysis ToolPak" add-in. (Select Tools > Add-ins, Analysis ToolPak

Determining Induction Machine Parameters

Three tests are needed to determine the parameters in an induction machine model. Detailed
testing is quite involved and is specified in IEEE Standard 112.
DC Resistance Test
If a DC voltage is applied to the stator terminals, there will be no potential difference across
any inductance and no induced voltage on the rotor. As a result, the per-phase circuit is
reduced to the stator winding resistance. Applying a DC voltage across two motor terminals
and measuring the current, the stator resistance can be found. Considering the figures
below, it can be seen that the apparent impedance will depend on whether the machine is
wye or delta connected.

R 1Y =V DC 2I DC R 1 =3V DC 2I DC

Locked Rotor Test

In the locked rotor test, the rotor of the machine is prevented from rotating and the supply
voltage gradually increased until rated current is reached. When the rotor is stationary , the
slip, s=1 and the equivalent circuit can be drawn as

Analysing the circuit:

P LR Q LR =3I 2 1 (R 1 +R 2 )=3I 2 1 (X 1 +X 2 ) LR

Therfore, if the stator power, current and reactive power are measured, the resistance and
leakage reactances can be found.
Ideally, the locked rotor test should be performed at a reduced frequency to account for skin
depth and more accurately predict the rotor resistance under load conditions. Typically, this
test is done at 1/4 rated frequency. With the value of stator resistance from the DC test, the
rotor resistance referred to the stator can be found.
If the locked rotor test is done at reduced frequency, then the reactances (which are
proportional to frequency) must be scaled to find the correct value at rated frequency:
(X 1 +X 2 ) rated =f rated f LR (X 1 +X 2 ) LR

At this point, only the total leakage reactance is known. The relative values of stator and
rotor leakage must be found from experience and are summarised in the table below
 Percentages of leakage reactance due to stator or rotor

Motor Type X1 X2
Wound Rotor 50% 50%
Class A 40% 60%
Class B 40% 60%
Class C 30% 70%
Class D 50% 50%

No Load Test
The no-load test of an induction machine enables the magnetizing reactance, core loss
resistance and friction and windage loss to be calculated. To be able to seperate friction and
windage losses from core losses, an adjustable power supply is required. If a fixed voltage
power supply is used, the separation of losses is not possible. In this case, the losses may be
grouped as "rotational losses". The theory behind this test for both cases is presented.

No Load Test - Core Loss combined with Friction and Windage

During the no-load test, the machine is allowed to accelerate up to synchronous speed with
no load applied. The machine will run close to synchronous speed, with s 0 If slip
approaches zero, the rotor current will fall to zero and the equivalent circuit can be drawn
as shown below.

Remembering that this is a per-phase circuit we can write

P nl Q nl =3I 2 1 R 1 +P rotational =3I 2 1 (X 1 +X m )

where rotational losses are defined as

P rotational =P F+W +P core

Note that we assume that stray load losses are zero at no-load
Measuring the stator current, power and reactive power, the remaining paramters can be
found. Using the power, if stator resistance is known (from the DC test), the total rotational
losses can be found
Using the value of X 1 from the locked rotor test, the magnetising reactance, X m , may
be found from the no load reactive power.
No Load Test - Identifying Core Loss and Friction and Windage
Including the core loss resistance,the circuit at locked rotor becomes

In this case, the input power and reactive power are given by:
P nl Q nl =3I 2 1 R 1 +3E 2 R c +P F+W =3I 2 1 X 1 +3E 2 X m

where E is the voltage across the magnetizing branch

.
In order to seperate core loss from friction and windage, it is necessary to measure the input
power at as the input voltage is reduced. Subtracting the stator copper loss, the total core
loss plus friction and windage may be plotted as a function of supply voltage. Extrapolating
to zero volts, when the core loss will be zero, it is possible to estimate the friction and
windage. An example of this plot together with a quadratic fit (in this case an automatic
trendline from Excel) is shown below. From the data, it is possible to estimate the friction
and windage as approximately 200W.

With R 1 and X 1 known, it is possible to calculate E , then find R C and X m from

the input power and reactive power
 Rotor Resistance
Investigating the torque-speed curve it is apparent that the rotor circuit resistance has
significant impact on speed at which maximum torque occurs. The plots below illustrate
two cases, with low rotor resistance on the left and high rotor resistance on the right.

It can be seen from the plots that a high rotor resistance will provide a high starting torque,
leading to rapid acceleration of the mechanical system. This is desirable since during
starting the stator current is significantly above the rated current. Short acceleration times
reduce the stress on the power system caused by high currents.
While high starting torques are desirable, high rotor resistance results in a relatively high
slip during normal running operation. As torque is porportional to rotor joule losses divided
by slip, high resistance causes increased losses and reduced efficiency during normal
operation.
The above points cause a problem. For most applications it is desirable to have:

high starting torque

high efficiency at rated speed

However, designs with high starting torque will have low efficiency at rated speed and
designs with high efficiency will have low starting torque. In order to resolve these
confilicting requirements, two steps must be considered:

1. Careful consideration of the application requirements

2. Designs of motors with variable rotor resistance.

Application Requirements
Many motor applications will not need both high start torque and high rated efficiency.
Alternately, the requirement for either high start torque or efficiency may be so significant
that it over-rides other requirements. Consider two examples:
Fan Load
Fans and rotary pumps typically have a torque requirement that varies as either the square
or cube of mechanical speed. When driving a fan, the motor must provide rated torque at
rated speed, but at lower speeds the torque demand is significantly lower. A fan application
 will therefore not usually require significant starting torque, efficiency during steady
operation at rated speed is the over-riding concern.

Variable-torque high-inertia loads

This type of load typically includes mechanical punches and reciprocating rod pumps used
in oil production ("nodding donkeys"). In the case of a reciprocating pump, the mechanical
laod varies with time, some of the time the motor is working against gravity to lift oil out of
the ground, at other times, it is working with gravity as the rod falls. The speed range of
this system is significant, requiring very high torques at low speeds. When the motor is at
high speed (as the rod falls) the torque (and efficiency) requirement is minimal. In this case
high torque at low speed is the over-riding requirement.

Variable Rotor Resistance

Although it is common to think of low frequency conductors as having a constant
resistance, the resistance of all ac conductors is a function frequency. Induced currents in
conductors act to oppose the originating magnetic field. As a result, the depth of penetration
of the magnetic field into the conductor will vary with frequency. (The magnitude of the
induced voltage is a function of rate of change of flux, and therfore a function of frequency)

Skin Depth
The skin depth of a conductor is defined as the depth at which the magnitude of a magnetic
field has fallen to 1/e of the magnitude of the surface. We can approximate this as the depth
at which currents are actually flowing in the conductor. Skin depth is given by

where

= skin depth
= permeability of the conducting material, = 0 in non-magnetic materials (copper,
aluminum)
= conductivity
f = frequency of the magnetic field, relative to the conductor

Since the frequency of the stator magnetic field seen by the rotor conductors is a function of
slip, the actively conducting region of the rotor bars will be a function of slip. Hence, the
effective roto resistance will be a function of slip. The table below plots skin depth for
aluminum ( 2.9107) in a machine with a 60Hz supply.

Slip Slip Frequency (hz) (mm)

0.025 1.5 76.3
0.05 3.0 54.0
0.083 5.0 42.8
0.167 10.0 29.6
 0.333 20.0 20.6
0.50 30.0 17.1
0.667 40.0 14.8
0.833 50.0 13.2
1.0 60.0 12.0

The table above indicates a number of important points:

1. No matter how deep a rotor bar, only the top 12mm conducts at standstill
2. Medium-large machines with bars deeper than 12mm will have varying rotor resistance
3. Smaller machines with bars less than 12mm deep will have effectively constant rotor
resistance

As an example, a machine with a rectangular bar 72 mm deep will have a resistance 6 times
smaller at low slips than it will at starting.

Power & Torque

Power Flow
The input power to a three-phase induction machine is given by
P in =3 V LL I L cos=3V 1 I 1 cos

Output power can be found by subtracting the losses from the input power
Losses

1. Stator Copper Loss. The stator resistive losses

P SCL =3I 2 1 R 1

2. Rotor Joule Loss. The rotor resistive losses. This is often called rotor copper loss, but since
the rotor conductors are aluminum, rotor joule loss is the more correct terminology.

P RCL =3I 2 2 R 2

3. Core Loss, or Iron Loss. The losses due to eddy current and hysteresis losses in the
laminations. This can be calculated using the resistor R c . At times, core losses are
grouped with friction and windage and stray loss as rotational losses.

Rotor Power
The power transferred to the rotor is called the "Airgap Power". Consider the equivalent
circuit below (the core loss resistance has been removed and core losses grouped into
rotational loss).
 From the above circuit, it can be seen that the total power transfer to the rotor is given by
P gap =3I 2 2 R 2 s

To find the power converted to the mechanical system the rotor joule loss must be
subtracted from the total rotor power
P conv P conv P conv =P gap P RCL =3I 2 2 R 2 s 3I 2 2 R 2 =3I 2 2 R 2 1ss

From the above equations, it can be seen that power converted to the mechanical system is
a function of the airgap power and slip:
P conv =(1s)P gap

Final output power may be obtained by subtracting the rotational loss from P conv .
P out =P conv P rotational

Torque
As with all rotating mechanical systems in steady state, torque can be found from the power
and mechanical speed
=P

In the case of an induction machine, the electromagnetic torque generated by the machine
can be found using
=P conv m =(1s)P gap (1s) s =P gap s

Writing the torque in terms of the rotor current:

=3I 2 2 R 2 s s

Finally, to find the available shaft torque after rotational losses, the output power must be
used.
shaft =P out m

Induction Machine Equivalent Circuit Model

We have seen that induction machines (as you might guess from the name) operate on the
principle of induced currents. There are still two magnetic fields, one from each of the rotor
and stator, but the rotor field is induced by the stator field. Effectively, we can think of the
induction machine as a rotating transformer. The stator is like the primary of a transformer
and creates the initial field, inducing voltages and currents in the secondary rotor winding.
The fundamental differences from a stationary transformer are:

The secondary rotates

There is an airgap, therefore more mmf is needed for a given flux density
 The secondary voltage and frequency depend on speed

As an aside, a wound rotor induction machine can actually be used as a variable frequency
transformer. For instance, a 60Hz system connected to the primary of an induction machine
can transfer power to a 50hz system connected to the rotor if the machine is mechanically
driven at a slip of 5/6.
The per-phase equivalent circuit model for an induction machine in steady state operation
supplied by a balances three-phase supply is based on the transformer model shown below

In the diagram above,

V1 = Phase RMS Voltage

I1 = Stator Phase Current
R1 = Stator Winding Resistance
X1 = Stator Winding Leakage Reactance
Xm = Magnetizing Reactance
Rc = Core Loss Resistance
E1 = Air Gap Voltage
I2 = Rotor Current Referred to Stator
ER = Rotor Induced Voltage (Actual)
IR = Rotor Current Voltage (Actual)
XR = Rotor Leakage Reactance (Actual)
RR = Rotor Resistance (Actual)

Rotor Circuit
We know from the operating principles that induced voltage ER and rotor leakage reactance
XR both depend on slip. To simplify the model we can define them both in terms of their
values when the speed is zero, slip s=1
E R X R =sE R0 =sX R0

where

E R0 is the induced voltage at standstill

X R0 is the rotor leakage reactance at standstill

With the above definitions we can write the equation for actual rotor current as
I R I R I R =E R R R +jX R =sE R0 R R +jsX R0 =E R0 R R s +jX R0

and the transformer model may be re-drawn as

 In the above diagram, the effective turns ratio a eff is constant and equal to the effective
turns ratio at standstill. In a wound rotor machine, a eff , R R and X R0 can be
measured. In a cage machine these parameters cannot be directly determined, there is no
method to directly measure voltages or currents on the rotor. To overcome this difficulty,
the rotor (secondary) circuit can be referred to the stator (primary) side.
Full Equivalent Circuit Model

In the above circuit

R 2 =a 2 eff R R , the rotor resistance referred to the stator

X 2 =a 2 eff X R0 , the rotor leakage reactance referred to the stator

The symbols used in induction machine models vary depending on the text and the context
in which the circuit is being used. R I , R o , R fe , R m can all be found as references
to the iron loss resistance. In some texts (especially from Europe), R 2 . X 2 refer to
actual rotor values with R 2 , X 2 used for referred values. In drives texts, it is
common to find R s , R r for stator resistance and rotor resistance referred to the stator
with X ls , X lr for leakage reactance.
 Maximum Torque and Output Power
Using the Thevenin torque equation:
=3V 2 TH (R TH +R 2 s ) 2 +(X TH +X 2 ) 2 R 2 s s

it is possible to plot the torque-speed curve of an induction machine. However, it is useful

to know the magnitude of the maximum or "pullout" torque without plotting the full torque-
speed curve. From the equation above, the peak torque could be found be differentiating
with respect to slip to find the slip which gives maximum torque. However, a
mathematically simpler and intuitively clearer answer can be found be considering the
power flow in the Thevenin equivalent circuit

Analysing the full equivalent circuit it was observed that

=P gap s

Therefore, since synchonous speed is constant, maximum torque occurs at the same slip as
maximum airgap power. Considering the Thevenin circuit, and applying maximum power
transfer theory, maximum airgap power and maximum torque will occur when
R 2 s =|R TH +j(X TH +X 2 )|

Re-arranging it is possible to obtain the slip for maxiumum torque, or pullout torque. Note
that this should not be called the "maxiumum slip".
s max =s po =R 2 [R 2 TH +(X TH +X 2 ) 2 ] 0.5

Substituting the pullout slip into the Thevenin torque equation give the pullout torque, or
maximum torque:
max =3V 2 TH 2 s (R TH +[R 2 TH +(X TH +X 2 ) 2 ] 0.5 )

Discussion
From the two equations above it can be seen that

The slip at which maximum torque occurs is proportional to rotor resistance

The magnitude of the maxiumum torque is independent of rotor resistance

If all other parameters remain constant, increasing the rotor resistance will:

1. Reduce the speed at which maximum toruqe occurs

2. Increase the starting torque (until spo > 1, then it will reduce the start torque)
3. Increase slip for a given torque
4. Reduce the speed for a given torque
5. Increase the rotor losses at a given torque

The last point above can be shown by considering that the torque equation
=3I 2 2 R 2 s s
 is actually rotor copper loss divided by slip speed in radians per second. If slip increases,
losses must increase to maintain the torque.
The diagram below plots torque speed curves for a motor with the following parameters, R2
is varied.
R1=0.5 , X1=0.75 , X2=0.5 , Xm=100 , f=60Hz, p=6, VLL=230V, Y-connection

Three-Phase Induction Machines

Like other electrical machines, induction machines can be operated as either generators or
motors. However, they are primarily used as induction motors. We will concentrate on
motoring and later consider applications where induction generation is attractive.
Induction machines are by far the most common type of motor used in industrial,
commercial or residential settings. (One notable exception may be in computer equipment,
where there are significant numbers of motors used for hard drives, DVD players etc.)
Depending on location (which affects other electrical energy uses, such as lighting, space
heating) induction motors may consume up to 70% of all electrical energy generated. There
are two underlying reasons why induction motors are the general purpose motor of choice:

Induction motors are cheap

Induction motors are robust
 Rotor Resistance
Investigating the torque-speed curve it is apparent that the rotor circuit resistance has
significant impact on speed at which maximum torque occurs. The plots below illustrate
two cases, with low rotor resistance on the left and high rotor resistance on the right.

It can be seen from the plots that a high rotor resistance will provide a high starting torque,
leading to rapid acceleration of the mechanical system. This is desirable since during
starting the stator current is significantly above the rated current. Short acceleration times
reduce the stress on the power system caused by high currents.
While high starting torques are desirable, high rotor resistance results in a relatively high
slip during normal running operation. As torque is porportional to rotor joule losses divided
by slip, high resistance causes increased losses and reduced efficiency during normal
operation.
The above points cause a problem. For most applications it is desirable to have:

high starting torque

high efficiency at rated speed

However, designs with high starting torque will have low efficiency at rated speed and
designs with high efficiency will have low starting torque. In order to resolve these
conflicting requirements, two steps must be considered:

1. Careful consideration of the application requirements

2. Designs of motors with variable rotor resistance.

Application Requirements
Many motor applications will not need both high start torque and high rated efficiency.
Alternately, the requirement for either high start torque or efficiency may be so significant
that it over-rides other requirements. Consider two examples:
Fan Load
Fans and rotary pumps typically have a torque requirement that varies as either the square
or cube of mechanical speed. When driving a fan, the motor must provide rated torque at
rated speed, but at lower speeds the torque demand is significantly lower. A fan application
will therefore not usually require significant starting torque, efficiency during steady
operation at rated speed is the over-riding concern.
 Variable-torque high-inertia loads
This type of load typically includes mechanical punches and reciprocating rod pumps used
in oil production ("nodding donkeys"). In the case of a reciprocating pump, the mechanical
laod varies with time, some of the time the motor is working against gravity to lift oil out of
the ground, at other times, it is working with gravity as the rod falls. The speed range of
this system is significant, requiring very high torques at low speeds. When the motor is at
high speed (as the rod falls) the torque (and efficiency) requirement is minimal. In this case
high torque at low speed is the over-riding requirement.
Varirable Rotor Resistance
Although it is common to think of low frequency conductors as having a constant
resistance, the resistance of all ac conductors is a function frequency. Induced currents in
conductors act to oppose the originating magnetic field. As a result, the depth of penetration
of the magnetic field into the conductor will vary with frequency. (The magnitude of the
induced voltage is a function of rate of change of flux, and therfore a function of frequency)
Skin Depth
The skin depth of a conductor is defined as the depth at which the magnitude of a magnetic
field has fallen to 1/e of the magnitude of the surface. We can approximate this as the depth
at which currents are actually flowing in the conductor. Skin depth is given by

where

= skin depth
= permeability of the conducting material, = 0 in non-magnetic materials (copper,
aluminum)
= conductivity
f = frequency of the magnetic field, relative to the conductor

Since the frequency of the stator magnetic field seen by the rotor conductors is a function of
slip, the actively conducting region of the rotor bars will be a function of slip. Hence, the
effective roto resistance will be a function of slip. The table below plots skin depth for
aluminum ( 2.9107) in a machine with a 60Hz supply.
Slip Slip Frequency (hz) (mm)
0.025 1.5 76.3
0.05 3.0 54.0
0.083 5.0 42.8
0.167 10.0 29.6
0.333 20.0 20.6
0.50 30.0 17.1
0.667 40.0 14.8
0.833 50.0 13.2
1.0 60.0 12.0
The table above indicates a number of important points:
1. No matter how deep a rotor bar, only the top 12mm conducts at standstill
2. Medium-large machines with bars deeper than 12mm will have varying rotor resistance
3. Smaller machines with bars less than 12mm deep will have effectively constant rotor
resistance

As an example, a machine with a rectangular bar 72 mm deep will have a resistance 6 times
smaller at low slips than it will at starting.

Starting Induction Motors

Induction Motors can usually be started by simply connecting them to the supply line
voltage. This can, however, cause power system problems. Start currents (or "in-rush"
currents) are high, often about 6 times rated current. Starting a motor directly from the
supply line can cause the supply voltage to dip, or "sag", due to the higher than normal
voltage drop across supply line and transformer impedances. In severe cases, this voltage
sag can cause supply instability or the failure of sensitive power electronics connected to
the system.
Wound Rotor Motors
Start behaviour of wound rotor machines can be controlled by adding resistance to the rotor
circuit, as shown below:

In the circuit model above, the actual resistance RE has been referred to the rotor as RE. It
can be seen from the circuit model that the increased impedance will reduce the start
current. In addition, the extra rotor resistance will increase the motor torque during starting,
reducing the start time and overall impact of the in-rush current.
 Cage Machines
Cage motors cannot have additional resistance added to the rotor circuit. A number of
options are available for cage motor starts:

Full Voltage, or direct-on-line start

Wye-Delta Start
Adding additional circuitry to the stator, including
o Autotransformer
o Additional resistance
o Electronic soft-start or variable speed start

Determining Induction Machine Parameters

Three tests are needed to determine the parameters in an induction machine model. Detailed
testing is quite involved and is specified in IEEE Standard 112.
DC Resistance Test
If a DC voltage is applied to the stator terminals, there will be no potential difference across
any inductance and no induced voltage on the rotor. As a result, the per-phase circuit is
reduced to the stator winding resistance. Applying a DC voltage across two motor terminals
and measuring the current, the stator resistance can be found. Considering the figures
below, it can be seen that the apparent impedance will depend on whether the machine is
wye or delta connected.

R 1Y =V DC 2I DC R 1 =3V DC 2I DC

Locked Rotor Test

In the locked rotor test, the rotor of the machine is prevented from rotating and the supply
voltage gradually increased until rated current is reached. When the rotor is stationary , the
slip, s=1 and the equivalent circuit can be drawn as
Analysing the circuit:
P LR Q LR =3I 2 1 (R 1 +R 2 )=3I 2 1 (X 1 +X 2 ) LR

Therfore, if the stator power, current and reactive power are measured, the resistance and
leakage reactances can be found.
Ideally, the locked rotor test should be performed at a reduced frequency to account for skin
depth and more accurately predict the rotor resistance under load conditions. Typically, this
test is done at 1/4 rated frequency. With the value of stator resistance from the DC test, the
rotor resistance referred to the stator can be found.
If the locked rotor test is done at reduced frequency, then the reactances (which are
proportional to frequency) must be scaled to find the correct value at rated frequency:
(X 1 +X 2 ) rated =f rated f LR (X 1 +X 2 ) LR

At this point, only the total leakage reactance is known. The relative values of stator and
rotor leakage must be found from experience and are summarised in the table below

Percentages of leakage reactance due to stator or rotor

Motor Type X1 X2
Wound Rotor 50% 50%
Class A 40% 60%
Class B 40% 60%
Class C 30% 70%
Class D 50% 50%

No Load Test
The no-load test of an induction machine enables the magnetizing reactance, core loss
resistance and friction and windage loss to be calculated. To be able to seperate friction and
windage losses from core losses, an adjustable power supply is required. If a fixed voltage
power supply is used, the separation of losses is not possible. In this case, the losses may be
grouped as "rotational losses". The theory behind this test for both cases is presented.

No Load Test - Core Loss combined with Friction and Windage

During the no-load test, the machine is allowed to accelerate up to synchronous speed with
no load applied. The machine will run close to synchronous speed, with s 0 If slip
approaches zero, the rotor current will fall to zero and the equivalent circuit can be drawn
as shown below.

Remembering that this is a per-phase circuit we can write

P nl Q nl =3I 2 1 R 1 +P rotational =3I 2 1 (X 1 +X m )

where rotational losses are defined as

P rotational =P F+W +P core

Note that we assume that stray load losses are zero at no-load
Measuring the stator current, power and reactive power, the remaining paramters can be
found. Using the power, if stator resistance is known (from the DC test), the total rotational
losses can be found
Using the value of X 1 from the locked rotor test, the magnetising reactance, X m , may
be found from the no load reactive power.
No Load Test - Identifying Core Loss and Friction and Windage
Including the core loss resistance,the circuit at locked rotor becomes

In this case, the input power and reactive power are given by:
P nl Q nl =3I 2 1 R 1 +3E 2 R c +P F+W =3I 2 1 X 1 +3E 2 X m

where E is the voltage across the magnetizing branch

.
In order to seperate core loss from friction and windage, it is necessary to measure the input
power at as the input voltage is reduced. Subtracting the stator copper loss, the total core
loss plus friction and windage may be plotted as a function of supply voltage. Extrapolating
to zero volts, when the core loss will be zero, it is possible to estimate the friction and
windage. An example of this plot together with a quadratic fit (in this case an automatic
trendline from Excel) is shown below. From the data, it is possible to estimate the friction
and windage as approximately 200W.

With R 1 and X 1 known, it is possible to calculate E , then find R C and X m from

the input power and reactive power

Determining Induction Machine Parameters

Three tests are needed to determine the parameters in an induction machine model. Detailed
testing is quite involved and is specified in IEEE Standard 112.
DC Resistance Test
If a DC voltage is applied to the stator terminals, there will be no potential difference across
any inductance and no induced voltage on the rotor. As a result, the per-phase circuit is
reduced to the stator winding resistance. Applying a DC voltage across two motor terminals
and measuring the current, the stator resistance can be found. Considering the figures
below, it can be seen that the apparent impedance will depend on whether the machine is
wye or delta connected.

R 1Y =V DC 2I DC R 1 =3V DC 2I DC

Locked Rotor Test

In the locked rotor test, the rotor of the machine is prevented from rotating and the supply
voltage gradually increased until rated current is reached. When the rotor is stationary , the
slip, s=1 and the equivalent circuit can be drawn as
Analysing the circuit:
P LR Q LR =3I 2 1 (R 1 +R 2 )=3I 2 1 (X 1 +X 2 ) LR

Therfore, if the stator power, current and reactive power are measured, the resistance and
leakage reactances can be found.
Ideally, the locked rotor test should be performed at a reduced frequency to account for skin
depth and more accurately predict the rotor resistance under load conditions. Typically, this
test is done at 1/4 rated frequency. With the value of stator resistance from the DC test, the
rotor resistance referred to the stator can be found.
If the locked rotor test is done at reduced frequency, then the reactances (which are
proportional to frequency) must be scaled to find the correct value at rated frequency:
(X 1 +X 2 ) rated =f rated f LR (X 1 +X 2 ) LR

At this point, only the total leakage reactance is known. The relative values of stator and
rotor leakage must be found from experience and are summarised in the table below

Percentages of leakage reactance due to stator or rotor

Motor Type X1 X2
Wound Rotor 50% 50%
Class A 40% 60%
Class B 40% 60%
Class C 30% 70%
Class D 50% 50%

No Load Test

The no-load test of an induction machine enables the magnetizing reactance, core loss
resistance and friction and windage loss to be calculated. To be able to seperate friction and
windage losses from core losses, an adjustable power supply is required. If a fixed voltage
power supply is used, the separation of losses is not possible. In this case, the losses may be
grouped as "rotational losses". The theory behind this test for both cases is presented.

No Load Test - Core Loss combined with Friction and Windage

During the no-load test, the machine is allowed to accelerate up to synchronous speed with
no load applied. The machine will run close to synchronous speed, with s 0 If slip
approaches zero, the rotor current will fall to zero and the equivalent circuit can be drawn
as shown below.

Remembering that this is a per-phase circuit we can write

P nl Q nl =3I 2 1 R 1 +P rotational =3I 2 1 (X 1 +X m )

where rotational losses are defined as

P rotational =P F+W +P core

Note that we assume that stray load losses are zero at no-load
Measuring the stator current, power and reactive power, the remaining paramters can be
found. Using the power, if stator resistance is known (from the DC test), the total rotational
losses can be found
Using the value of X 1 from the locked rotor test, the magnetising reactance, X m , may
be found from the no load reactive power.
No Load Test - Identifying Core Loss and Friction and Windage
Including the core loss resistance,the circuit at locked rotor becomes

In this case, the input power and reactive power are given by:
P nl Q nl =3I 2 1 R 1 +3E 2 R c +P F+W =3I 2 1 X 1 +3E 2 X m

where E is the voltage across the magnetizing branch

.
In order to seperate core loss from friction and windage, it is necessary to measure the input
power at as the input voltage is reduced. Subtracting the stator copper loss, the total core
loss plus friction and windage may be plotted as a function of supply voltage. Extrapolating
to zero volts, when the core loss will be zero, it is possible to estimate the friction and
windage. An example of this plot together with a quadratic fit (in this case an automatic
trendline from Excel) is shown below. From the data, it is possible to estimate the friction
and windage as approximately 200W.
Modes of operation

As mentioned when considering the induction machine torque-speed curve, the three main
modes of operation are braking, motoring and generating. We have concentrated on
analysis of a machine being used as a motor, but the analysis is general for other modes of
operation.
Motoring: Small Slips
In usual operation of an induction machine, as a motor attached to a fixed frequency
supply, there is little control over the operating point: the motor will operate at the speed
where the load torque is equal and opposite to the motor torque.
Considering again the torque speed equation:

If the slip is small then

and the torque equation can be re-written as

It can be seen that at small slips, torque is proportional to slip, doubling the load will
approximately double the slip.
It is clear from the above analysis that there is very little speed variation in an induction
machine if the synchronous speed is constant. The rotor will rotate at a small slip, slightly
below synchronous speed. If operation at more than one fixed speed is required, it is
necessary to change the synchronous speed. Considering the equation for synchronous
speed

Pole Changing
Changing the number of poles in a machine gives a set of discrete operating speeds. e.g. if a
machine can have 2 or 6 poles, it can operate at approximately 1200 rpm or 3600 rpm (with
a 60Hz supply)
Consequent Poles
The basic idea of consequent poles is to split a single phase winding into two groups of
coils, and to have the capability of reversing the connections of one of the coil groups. It is
best explained with the aid of a diagram as shown below:

In the diagram on the left, a single 2-pole winding is shown. This same flux pattern can be
obtained using two windings connected to operate in parallel, as shown in the centre
diagram. If the number of poles in the machine must be increased from 2 to 4, the supply to
one of the windings can be reversed, resulting in a 4 pole field.

Multiple windings
The idea of consequent poles helps if the speed must change by 2, but if other speeds are
needed, one option is two wind more than one set of stator windings onto a machine, but
only operate one at a time. As an example, it is possible to wind both a 6 pole winding and
a 2 pole winding onto the same stator. If the 6-pole winding is excited, the speed will be 3
times lower than when the 2-pole winding is excited. This idea is used in washing machines
to switch between wash and spin cycles. This approach is expensive as only half of the
 copper in the machine is active at any one time and the stator slots must be made large
enough to fit two sets of windings, rather than just one.

Variable Frequency Operation

If a variable frequency supply is available, the synchronous speed of an induction machine
may be chosen at any desired value. However, caution must be taken when operating an
induction machine at frequencies below the rated value. Consider the induction machine
equivalent circuit in terms of inductances, rather than reactances:

If the machine is operated at reduced frequency the impedance of the circuit will be
reduced. If the voltage is held at the rated value then the currents will be above the rated
value and the machine will overheat. In order to maintain constant current through an
inductor with variable frequency operation, the voltage and frequency must be adjusted
together. This is called constant Volts per Hertz operation.
If the machine is operated at low frequencies (e.g. less than 1/4 rated) the voltage drop
across the stator resistance will be significant. (As reactances get smaller, R1 gets
proportionally bigger.) Unless the stator supply voltage is compensated to allow for
additional stator resistance voltage drop, the available torque will fall.
If the machine is to be operated above rated frequency, the voltage cannot be increased
above rated voltage and the magnetizing current in the machine will be reduced. This mode
of operation is called field weakening operation.
Summary

Below rated frequency, keep the ratio V/f constant at rated values.
At low frequencies, the V/f ratio must be boosted to maintain rated torque.
Above rated frequency, maintain rated voltage.

Constant Volts-per Hertz operation

As an approximation, assume that a constant supply voltage:frequency ratio results in a
constant Thevenin Voltage:frequency ratio. (Due to stator resistance, this assumption
doesn't hold at low frequencies). If slip is small, then it has been shown that the torque
equation

can be approximated as
which, for constant V:f may be written as

i.e. if V/f is constant, torque is approximately proportional to slip speed.

The above statement can be used as a basis for relatively simple speed control.
Example
A 480V, 60Hz, 4-pole motor has rated speed of 1750 rpm and rated torque of 10Nm. If a
torque of 10Nm is needed at a mechanical speed of 1500 rpm, find the synchronous speed,
supply frequency and line-line supply voltage.
At rated torque, the slip speed will be the rated value. For a 4-pole 60hz machine,
synchronous speed is 1800 rpm, therefore rated slip speed =1800-1750=50rpm. When
operating at 1500rpm, 10Nm, slip speed will still be 50rpm and the synchronous speed is
given by

With the synchronous speed, the supply frequency can be found

Finally, if V/f is constant, the supply voltage must be 480 51.67/60 = 413.3V

Variable Voltage Variable Frequency Supply

Most modern drives use Pulse Width Modulation, (PWM) supplies to approximate a
variable voltage and frequency sinusoidal supply. In a PWM supply, a DC voltage is
switched rapidly to approximate the shape of the desired waveform. An example is shown
below.

Variable Frequency Operation

If a variable frequency supply is available, the synchronous speed of an induction machine
may be chosen at any desired value. However, caution must be taken when operating an
induction machine at frequencies below the rated value. Consider the induction machine
equivalent circuit in terms of inductances, rather than reactances:
 If the machine is operated at reduced frequency the impedance of the circuit will be
reduced. If the voltage is held at the rated value then the currents will be above the rated
value and the machine will overheat. In order to maintain constant current through an
inductor with variable frequency operation, the voltage and frequency must be adjusted
together. This is called constant Volts per Hertz operation.

If the machine is operated at low frequencies (e.g. less than 1/4 rated) the voltage drop
across the stator resistance will be significant. (As reactances get smaller, R1 gets
proportionally bigger.) Unless the stator supply voltage is compensated to allow for
additional stator resistance voltage drop, the available torque will fall.

If the machine is to be operated above rated frequency, the voltage cannot be increased
above rated voltage and the magnetizing current in the machine will be reduced. This mode
of operation is called field weakening operation.
Summary

Below rated frequency, keep the ratio V/f constant at rated values.
At low frequencies, the V/f ratio must be boosted to maintain rated torque.
Above rated frequency, maintain rated voltage.

Constant Volts-per Hertz operation

As an approximation, assume that a constant supply voltage:frequency ratio results in a
constant Thevenin Voltage:frequency ratio. (Due to stator resistance, this assumption
doesn't hold at low frequencies). If slip is small, then it has been shown that the torque
equation

can be approximated as

which, for constant V:f may be written as

i.e. if V/f is constant, torque is approximately proportional to slip speed.

The above statement can be used as a basis for relatively simple speed control.

Example
A 480V, 60Hz, 4-pole motor has rated speed of 1750 rpm and rated torque of 10Nm. If a
torque of 10Nm is needed at a mechanical speed of 1500 rpm, find the synchronous speed,
supply frequency and line-line supply voltage.
 At rated torque, the slip speed will be the rated value. For a 4-pole 60hz machine,
synchronous speed is 1800 rpm, therefore rated slip speed =1800-1750=50rpm. When
operating at 1500rpm, 10Nm, slip speed will still be 50rpm and the synchronous speed is
given by

With the synchronous speed, the supply frequency can be found

Finally, if V/f is constant, the supply voltage must be 480 51.67/60 = 413.3V
Variable Voltage Variable Frequency Supply
Most modern drives use Pulse Width Modulation, (PWM) supplies to approximate a
variable voltage and frequency sinusoidal supply. In a PWM supply, a DC voltage is
switched rapidly to approximate the shape of the desired waveform. An example is shown
below.

Generating
If the rotor of an induction machine rotates above synchronous speed, slip is negative, as
are torque, mechanical output power and air gap power. i.e. the machine is operating as a
generator. When considering synchronous machines it is normal to re-define the direction
of positive current when switching between generators and motors (keeping a positive
power flow for both cases). With induction machines we will not do this. If slip is negative,
the "input power" to the electrical terminals will be negative, implying that power is
flowing out of the electrical terminals.
Consider the machine used in the basic examples:
A 480V, 60 Hz, 6-pole, three-phase, delta-connected induction motor with the following
parameters:
R1=0.461 , R2=0.258 , X1=0.507 , X2=0.309 , Xm=30.74
Rotational losses are 2450W.
Now, suppose that the machine is being driven by a mechanical system such that it is
rotating at 1224 rpm
Calculate the following information:

i. slip
ii. Line Current
iii. Power and Reactive Power at the terminals
iv. Airgap Power
v. Torque Developed
vi. Mechanical Power
vii. Efficiency

Much of the theory for this problem is identical to a motor problem:

i. Slip is given by
 Using the rpm equation,

s = (1200-1224)/1200 = -0.02

ii. Now, phase current is given by

where phase impedance is given by

Using the above equation, Zin = -10.3 + j5.29

And noting that the machine is delta connected, V1 = VLL = 480V

I1 = -36.7 - j18.8 A. |I1| =41.4 A,

Therefore IL = 3 41.4 = 71.7 A

Note that the real part of phase current is negative, indicating that real power flows out of
the terminals. Also, the imaginary part of the phase current, which indicates that reactive
power must flow into the terminals. An induction generator cannot operate without a
reactive power supply, either from a power grid or capacitor bank

iii. Using complex notation,

S=3VI*

or

Pelec =3 &times VRe{I} = -53.0kW

Qelec =-3 &times VIm{I} = +27.2kVAR

iv. Airgap Power is given by

This approach requires rotor current to be found. With no core loss resistance:
 Giving I2 = 37.8 A. Substituting into the power equation

Pgap = -55.4kW

v. Torque developed can be found from

where synchronous speed in radians per second is given by

giving

= -441 Nm

vi. Mechanical power can be found using

and

Therefore mechanical power in kW is:Pout = -55.4 (1.0+0.02) - 2.45

Pmech = -58.9 kW

vii. Efficiency is given by

Therefore

= 53.0/58.9 = 89.9 %
Summary
It can be seen from the above analysis that the equations for an induction motor can all be
applied to an induction generator. (As long as output and input power are correctly
described as either electrical or mechanical).. Induction generation used to be relatively
rare. However, it is becoming increasingly common as induction generators are the
generator of choice for large wind turbines. The requirement of an induction generator to
have an independent reactive power supply has caused significant research into the impact
of large wind farms of power system stability.