Mississippi State University

Department of Electrical and Computer Engineering

ECE 3614: Fundamentals of Energy Systems (Fall 2015)
Instructor: Dr. Masoud Karimi ([email protected])
DC Machine

Contents
1 DC Machine: Definitions and Facts

2 Magnetic Flux in a DC Machine

3 Induced Voltage in Rotor Winding

4 Induced Torque in Rotor Winding

5 No-Load Operation of DC Machine

10

6 Motoring Operation of DC Machine

11

7 Generating Operation of DC Machine

11

8 Aspects to Practical DC Machines

12

9 Power Flow and Losses in a DC Machine

13

10 Equivalent Circuit of a DC Machine

14

11 Types of DC Machines

15

12 No-Load Characteristics of a DC Machine

17

13 Terminal Characteristics, Voltage and Speed Regulation

18

14 TC of a Separately Excited DC Generator

19

15 Armature Reaction

20

16 Terminal Characteristics of a Shunt DC Generator

21

17 An Example on DC Generator

23

18 TC of Separately Excited and Shunt DC Motors

25

19 Terminal Characteristics of a Series DC Motor

25

20 Speed Control of DC Motor

26
20.1 Armature Voltage Control . . . . . . . . . . . . . . . . . . . . . . 27
20.2 Field Control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28
20.3 Armature Resistance Control . . . . . . . . . . . . . . . . . . . . 29
21 An Example on Separately Excited DC Motor

30

22 An Example on Shunt DC Motor

31

23 DC Motor Starter

33

DC Machine: Definitions and Facts

A direct current (dc) machine either converts mechanical power to dc
electrical power and is called a dc generator or converts dc electrical power
to mechanical power and is called a dc motor.
Dc machines offer good efficiency and controllability over wide speed
range.
Dc machines are more costly than ac machines and are not widely used in
industry.
Dc machines have been used to perform some tough jobs for example in
steel mills and paper mills.
Dc machines are widely used as actuators for control purposes (for example
in robotic applications), in which context they are called servomotors.
Similar to ac machines, a dc machine comprises a stationary core called
the stator surrounding a rotating core called the rotor.
The magnetic field is established by a permanent magnet or a dc-excited
winding called the field winding placed on the stator.
The main, high-power winding (also called the armature winding) is located on the rotor. Here is where the electrical energy is generated (in
the case of generator) or converted to mechanical energy (in the case of
motor). (Notice the differences with ac machines in terms of field and
armature windings and their locations.)

Magnetic Flux in a DC Machine

Schematic structure of a simple rotating dc machine is shown in Fig. 5.

The rotor is cylindrical and can rotate around its shaft.
2

Figure 1: The stator of a 9 MW, 250 rpm, 1650 V mill motor.

Figure 2: The rotor of a 5 MW, 60/120 rpm, 1750 V converter mill motor.

Figure 3: A 8 MW, 50/100 rpm, 900 V dc reversing mill motor.

Figure 4: Two large dc motors used in the metal industry.

stator

stator
rotor

Figure 5: Schematic structure of a 2-pole dc machine.

The stator has either a permanent magnet providing N-S magnetic poles
or has a winding (supplied by a dc power) to establish the magnetic field.
Multiple number of poles can be used. Fig. 5 shows a two-pole stator.
Assuming that the air gap is narrow, and the stator and rotor cores have
high relative permeability, the magnetic flux is radially towards the rotor
under the N pole; and it is radially out from the rotor under the S pole.
This behavior can be graphically shown in Fig. 6.
B
Bmax

90

270

360

P
2

(degrees)

Bmax

Figure 6: Magnetic flux density versus angle . Negative B means towards the
rotor and positive B means away from the rotor. P is number of magnetic poles.

Induced Voltage in Rotor Winding

Consider a loop of wire located on a carved slot on the outer surface of the
rotor. The loop of wire is shown in Fig. 7.
Assume that the wire is rotating around its axis with a constant rotational
(or mechanical) speed of rad/s in the counterclockwise direction. Note that
5

2 7

3
1

6
8

Figure 7: A loop of wire located on the rotor surface.

= 2 n60m where nm is the rotor speed in rpm. It will be a moving conductor
in the magnetic field. Therefore, according to the Faradays law, a voltage is
induced in the wire. Lets examine this voltage.
For simplicity, assume P = 2. When 0 < < 180 , the section 43 of the wire is
under pole N and section 56 is under pole S. Therefore, the induced voltage on
this section is

e34 = (
v B ). = rB,
where r is the radius of the rotor, and is the height of the rotor (distance
from node 3 to node 4). Notice the direction of vectors, the right hand rule, the
polarity of voltage, and also the fact that v = r.
It is obvious that e56 is also equal to the same value. Moreover, e45 , e23 and e67
are zero. Therefore, the whole voltage induced in the wire is equal to
e = e18 = 2rB = AB =
where A = 2r is the area of the loop and is the magnetic flux. This voltage
is shown in Fig. 8.
Remarks:
The induced voltage is proportional to the flux and the speed of rotation.
In general, when a whole winding is distributed over the surface of the
rotor, the induced voltage can be presented by
e = k
where k is a constant called the machine constant. This constant depends
on the size and construction of the machine (including the cores, air-gap,
and windings).

e
emax

90

270

360

(degrees)

emax

Figure 8: Induced voltage in a rotating loop of wire versus the rotation angle
. Notice that emax = AB where A is the area of the loop, B is the magnetic
field density, and is speed of rotation in rad/s.
The induced voltage is ac! In a dc generator, we would like to have a
dc voltage. How can we make a dc voltage out of the ac voltage? One
challenge is that the armature winding is rotating.
In dc machines, the conversion between ac and dc has been done using
an arrangement which comprises conducting segments (or slip rings) and
brushes, also called commutators (or commutation system), as shown in
Fig. 9. Each terminal of the winding is connected to one of the segments and they rotate with the rotor while the brushes are stationary.
The brushes are made of soft and conducting material such as carbon or
graphite. When rotating, the combination of segments/brushes switches
the connection of the terminals to the output terminals such that the polarity of the output voltage remains unchanged. The brushes are located
such that the commutation from one ring to the other happens at the
zero crossing instant of the voltage. At this location, the two rings are
connected together through the brushes, and a zero-crossing commutation
avoids sparks. The rectified output voltage is shown in Fig. 10.

Induced Torque in Rotor Winding

Consider the same turn of wire (placed on the rotor as discussed before) and
assume that a current i flows through the wire. In the presence of magnetic
field, this current induces a force on the wire. This force for each segment of

the wire can be calculated from F = i( B ). It is easy to observe that the

two forces on the 34 and 56 segments are equal and are given by
F = iB.

brushes

+
slip rings

Figure 9: Slip rings and brushes to rectify the induced voltage.

e
emax

90

180

270

360

(degrees)

Figure 10: Rectified induced voltage versus angle after brushes.

Direction of the forces are such that they generate toques along the same direction. Therefore, the induced torque is equal to
= 2riB = ABi = i
where r is the radius of rotor cylinder and A = 2r is the area of the loop of
wire. If i is dc and flowing into the winding along 3456 direction, then the
torque is counterclockwise for half of the rotation when 34 is under the N pole
(0 < < 180 ) and it is clockwise for another half a cycle (180 < < 360 ).
The net torque is zero over a full cycle, the rotation will not subsist.
Now, it turns out that the same commutation system (slip-rings and brushes)
will address this issue. Lets look at the bigger picture where a dc voltage supply
is connected to the rotor (or armature) winding through the brushes and slip
rings as shown in Fig. 11. And for the moment assume that the current flows
8

into the machine along 3456 direction. The actual direction of current into the
turn of wire changes direction when the wire makes a half-a-cycle rotation. As
a result, the produced torque gets rectified and will be unidirectional during the
entire cycle of rotation.

SW
+

R
E

Figure 11: Connecting a dc supply to the armature winding.

Remarks:
The induced torque on one loop of wire is equal to = i and this means
that it is proportional to the current and to the magnetic flux. This
relationship can be generalized (for a whole winding distributed over the
surface of rotor) as
= ki
where k is a constant called the machine constant. This constant depends
on the physical dimensions and construction of the machine as well as the
number of turns and the distribution of the armature winding.
The current flowing through the wire produces a magnetic field which may
be called rotor field or armature field. Lets denote that with Br . This
field is proportional to the current i and is perpendicular to the plane
of wire turn. Its direction can be found using the right hand rule. The
induced torque is the result of interaction between this field and the main
filed. And the torque is in such a direction that tries to make the two
fields parallel (similar to when two pieces of magnet try to parallel one
another when placed close to each other).
When the commutation takes place, the rotor magnetic field experiences
a leap of 180 degrees and leaps over the opposing-torque half-cycle and
keeps the torque unidirectional.
As far as a single loop of wire is concerned, the induced voltage is equal to
e = 2rB and the induced torque is equal to = 2riB. The electrical

power at its terminals is equal to

Pelec = ei = 2rBi.
The mechanical power corresponding to the torque is
Pmech = = 2riB.
And obviously,
Pelec = Pmech .
This makes perfect sense because we have analyzed a no-loss situation
and exactly the same amount of power (or energy) is converted from one
form to another at the wire location. In practice, there are different losses
involved which well discuss those in detail later.

No-Load Operation of DC Machine

Consider a dc machine and assume that the rotor is initially at rest. Also,
assume that the armature circuit comprises a dc voltage source of E volts, a
series resistance of R ohms and a switch as shown in Fig. 11. Close the switch
at t=0. Neglect all frictions for this analysis. A current
i=

Ee
R

(1)

flows into the wire. The voltage e is the induced voltage on the armature winding. It is also called electromotive force (emf) or sometimes back electromotive
force (back emf). Remember this voltage is proportional to the field and the
speed of rotation. Now, at the start point e = 0 because rotational speed is zero.
The current of (1), in the presence of magnetic field, induces a force (torque) on
the wire (the third principle of magnetism). The rotor starts rotating (as there
is no friction). Once it starts rotating, a voltage e is induced in the wire that
is proportional to the speed. As the rotor speeds up, the voltage goes higher
to the point that it equals E. At this time, the current becomes zero and the
torque also vanishes. The rotor will keep rotating at a constant steady speed
that is called no-load speed. This speed can easily be calculated from
e = E kNL = E NL =

E
k

where k is the machines constant.

In the steady state, the machine draws no current and develops no torque. In
other words,
Pelec = ei = e 0 = 0, Pmech = = 0 = 0
and that is why this scenario is called the no-load condition.

10

Motoring Operation of DC Machine

Assume that the machine is initially running at no-load condition: e = E, i = 0,

E
= 0, = NL = k
. Apply an external torque apposite to the direction
of rotation. This force lowers the speed, lowers the induced voltage e and, as
a result, a current i starts flowing into the machine. Flow of current induces a
force which opposes the external force. This process continues until the machine
settles at a new speed at which the induced force and the external force balance
each other, that is

,
= = ki i =
k
R
.
k

e = E Ri = E
On the other hand,
e = k.
Combining these two equations results in
=

R
k

= NL

R
(k)2

where k is the machines constant.

The current i flows into the machine. This means that an electrical power
enters the machine and is equal to
Pelec = ei = ki.
The machine is developing a mechanical power that is equal to
Pmech = = ki.
Clearly Pmech = Pelec , and this is the motor operation of the machine.

Generating Operation of DC Machine

Assume that the machine is initially running at no-load condition: e = E, i = 0,

E
= 0, = NL = k
. Apply an external torque at the same direction of
rotation. This force increases the speed, increases the induced voltage e beyond
E and a current i starts flowing out from the machine into the battery. Flow of
current induces a force which opposes the external force. This process continues
until the machine settles at a new speed at which the induced force and the
external force balance, that is
= = ki i =

11

,
k

R
.
k

e = E + Ri = E +
On the other hand,
e = k.
Combining these two equations results in
=

E+

R
k

= NL +

R
(k)2

where k is the machines constant.

The current i flows from the machine to the dc voltage source E. This means
that an electrical power is generated by the machine and is equal to
Pelec = ei = ki.
The mechanical power is equal to
Pmech = = ki.
Clearly Pmech = Pelec , and this is the generator operation of the machine.

Aspects to Practical DC Machines

Although the principles of operation remain unchanged, a practical dc machine

differs from the basic/simple structure discussed earlier from the following points
of view.
Large number of turns (wire-loops) are connected to form a coil, and coils
are connected and spread over the surface of rotor to form the armature
winding. There are different techniques and ways as how to sequentially
connect and spread the coils in order to achieve high (and smooth) output
voltage or high (and smooth) output torque, depending on whether it is
desired to operate as a generator or as a motor, respectively.
The stator often has more than two magnetic poles. This way, it is practically easier to achieve higher and smoother magnetic field in the air gap.
The number of magnetic poles P is a factor of 2. Then, a full cycle physical
rotation of rotor corresponds to P2 cycle of the induced ac voltage (before
brushes). In other words, electrical angle (and frequency) is P2 larger than
mechanical angle (and frequency).
The magnetic poles are generated either by permanent magnets or by
proper connection of winding terminals. The poles may be salient (projecting out inside the stator surface towards the rotor) or otherwise.
Real machine is not lossless. Different losses including copper losses (in the
windings), brush losses (friction and copper), core losses (in the magnetic
cores), and the rotational or mechanical losses (caused by friction and
windage) exist.
12

Power Flow and Losses in a DC Machine

The efficiency of the machine is

=

Pout
100.
Pin

The difference between the input power and the output power is the total losses
in the machine. The losses can be listed as
1. Mechanical losses (Pmech ): friction (bearing friction and brushes friction)
and windage (caused by rotating parts and air inside the machines casing).
2. Core losses (Pcore ): hysteresis and eddy
3. Brush loss (Pbrush ): electrical loss at brushes
4. Copper losses (Pcu ): ohmic loss in armature windings (ohmic losses in the
field windings are normally much smaller and are negligible)
5. Stray losses (Pstray ): all other un-modeled losses
Sum of 1, 2 and 5 is often called the rotational losses. It is typically 3 to 5%
of the machine rating for a well designed machine. Moreover, it is relatively
constant from no load to full load operation of the machine. This means that to
approximately estimate the rotational loss, we can run the machine at no load
and measure its input power.
In a dc generator, the input power is the mechanical power provided by
the prime mover. Its relationship with the torque and speed is given by
Pin = in
where in is the input torque and is the speed of rotation in rad/s ( = 2 n60m
where nm is the rotor speed in rpm). The output power is the electrical power
provided by the generator and is given by
Pout = vt it
where vt is the terminal voltage and it is the terminal current.
The converted power and the induced torque satisfy the equation
Pconv = Pin Pstray Pmech Pcore = Pin Prot = ind = ei
where e is the internal emf voltage and i is the armature windings current.
Finally, the output power is
Pout = Pconv Pcu Pbrush .

13

In a dc motor, the input power is the electrical power given to the motor
Pin = vt it
and the output power is the mechanical power obtained from the motor:
Pout = load .
The converted power and the induced torque satisfy
Pconv = Pin Pbrush Pcu = ind = ei
and thus,
Pout = Pconv Pcore Pmech Pstray = Pconv Prot .

10

Equivalent Circuit of a DC Machine

The general equivalent circuit of a dc machine is shown in Fig. 12. It comprises

the field circuit and armature circuit. Each winding is modeled by a resistance
and an inductance.1 The field circuit establishes the magnetic field by means of
current if in the field winding (generated by the field voltage vf ). The induced
magnetic field is proportional to the field current if (neglecting the saturation effect, otherwise, it follows the magnetization curve).
vf = Rf if + Lf
Rf

dif
dt

La

if

Ra

vf

+
field circuit

Lf

armature circuit

va

e = va + Ra ia + La didta (Generator, ia out from machine)

e = va Ra ia La didta (Motor, ia into the machine)
Figure 12: General equivalent circuit of a dc machine
The armature circuit comprises the emf source e that is the generated voltage
explained by the Faradays law (a moving conductor in the magnetic field). In
1 Notice that the inductance voltage is L di and it is practically zero in the steady state
dt
because current is dc. Therefore, it may be ignored for steady state analysis. However, the
inductor voltage is nonzero in transient conditions. The inductance should be considered
especially for control purposes when the dynamic behavior is of interest.

14

the no-load condition, the armature current settles to zero. In the generator
operation, the armature current flows out from the armature and e is larger
than va . In the motor operation, the armature current flows into the winding
and e is smaller than va .

11

Types of DC Machines

Figure 12 shows that the field winding and the armature winding are separate. In practice, they do not have to be separate. These two windings can
be connected together as both of them carry dc currents. Depending on how
the electrical connection between field winding and armature winding is made,
different types of dc machines (yielding different performances and properties)
are obtained. This is one of the major advantages of dc machine. These types
are explained below.
Separately excited dc machines: excitations for field winding and for armature winding are separate.2 Figure 13 shows this connection. In this
type, two dc sources are required when the machine is a motor while only
one dc source is required when the machine is a generator. A rheostat can
be placed in the field in order to control the field current.3
Rf

La

if

vf

Ra
+

Lf

va

Figure 13: Separately excited

Self excited dc machines: there is only one source of excitation4 and both
windings are supplied by this source through different topologies described
below.
Shunt connection: field winding is connected in parallel with the
armature winding (as shown in Fig. 14).
Series connection: field winding is connected in series with the armature winding (as shown in Fig. 15).
Compound
2 The

permanent magnet machine may also be considered among this category.

is true for all machine types discussed here.
4 A source is required for motor operation. In generator operation, the generated voltage
serves as the dc source for the field and no external source is required.
3 This

15

Rf

Ra

vf = va

if
Lf

La

Figure 14: Self excited (shunt connection)

Rf

Lf

Ra

if = ia

La

Figure 15: Self excited (series connection)

Long compound connection: one field winding is connected in
series with the armature winding and the result is connected in
parallel with another field winding (as shown in Fig. 16).
Short compound connection: one field winding is connected in
parallel with the armature winding and the result is connected
in series with another field winding (as shown in Fig. 17).
Remark: In compound connections, the two field windings may connect in two
different ways: cumulative and differential. In the cumulatively compounded
connection, the magnetic fields generated by both windings add up together.
In the differentially compounded connection, the magnetic fields generated by
both windings are in apposite directions.

16

Rf

Lf

Rf

Ra

Lf

La

Figure 16: Self excited (long compound)

Rf

Lf

Rf

Ra

Lf

La

Figure 17: Self excited (short compound)

Remark: Depending on the number of turns of the series and shunt windings, different levels of compound-ness may be achieved resulting in different
performances of the machine.

12

No-Load Characteristics of a DC Machine

The emf voltage satisfies

e = k

(2)

where is the flux, is the speed in rad/sec. Rotation per minute (rpm) is
often used in the context of machines to express the speed. One rpm is 1/60

17

rotation per second. And one rotation per second is 2 rad/sec. Therefore,
n rpm =

2n
rad/sec.
60

Equation (2) shows that the induced emf is proportional to the speed and to
the flux. The flux is produced by the field and it is related to the field current.
The relationship between the field current and the flux is, however, generally
a nonlinear relationship due to the saturation characteristics of ferromagnetic
materials.
e
n = n3 > n2
n = n2 > n1
operating point
n = n1 rpm

if

Figure 18: Relationship among the induced voltage, the rotor speed, and the field
current; also called the no-load curves.
Figure 18 shows curves of e versus field current for different values of speed.
When the machine is operating at no-load condition, e is equal to the armature
voltage va (and the armature current is zero). This characteristics is called the
no-load characteristics of the machine.

13

Terminal Characteristics, Voltage and Speed

Regulation

For a generator, the terminal characteristics (or the load or the output characteristics) is the plot of generated output voltage versus the output current. In
other words, it shows how the generated voltage behaves when different (electrical) loads are connected to the generator.
A generator is expected to offer a regulated voltage for wide range of output
loads. The voltage regulation is, thus, defined as a figure of merit of the generator, as
vNL vFL
VR =
100
vFL
18

and is close to zero for a well-designed generator. In VR equation above, vFL

is the rated voltage of the generator. Assume that a load is connected to the
machine so that it draws the rated power at rated voltage. This is called the
full load condition of generator. Now, take all the load off the output (without changing anything else including rotor speed and magnetic field). In other
words, make the output open circuit. The current becomes zero and machine
will operate at no-load condition. The output voltage in this case is vNL . The
VR quantifies the relative change between these two values.
For a motor, the terminal characteristics (or the load or the output characteristics) is the plot of the rotor speed versus the output torque. In other words, it
shows how the rotor speed behaves when different (mechanical) loads are connected to the rotor.
A motor is expected to offer a regulated speed for wide range of output loads.
The speed regulation is, thus, defined as a figure of merit of the motor, as
SR =

nNL nFL
100
nFL

and is close to zero for a well-designed motor. Notice that nFL is the rated
speed of the motor. Assume that a load is connected to the rotor such that it
delivers the rated power at rated speed. This is called the full-load condition of
the motor. Now, take all the load off the rotor (without changing anything else
including input supply voltage and magnetic field). In other words, make the
output torque (or power) zero. The motor will operate at no-load condition.
The rotor speed in this case becomes nNL . The SR quantifies the relative change
between these two values.

14

TC of a Separately Excited DC Generator

Rf

La

if

vf

Ra

it
+

Lf

vt

RL

Figure 19: Separately excited dc generator feeding a load RL .

Consider a separately excited dc generator whose field current is equal to
if and its rotor turns at a constant speed n rpm. The no-load characteristics,
Fig. 18, gives the value of e. Now, assume that a load equal to RL ohms is
connected to the output terminals of this generator as shown in Fig. 19. Call
the output voltage (across the load) vt where t stands for terminal. Notice
19

that the current flowing through the load (or terminal) is equal to the armature
current ia , i.e. it = ia . In the steady situation, the following equations hold:
vt = e Ra it , vt = RL it
vt

operating point

vt = RL it

vt = e Ra it

Vt

impact of armature reaction

it

It

Figure 20: Terminal characteristics of a separately excited dc generator.

The operating point of the circuit is obtained from the intersection point of
these two lines. This is shown in Figure 20.
Remarks
Decreasing the output resistance RL (increases the current it and) decreases the output voltage. The slope of this change depends, however,
on Ra and is very small. Therefore, the voltage is pretty stable within a
wide range of output loads.
For large values of output current, the output voltage may drastically
fall. This behavior is explained by what is called the armature reaction
(explained below).
Since e = k and it increases by an increase in the magnetic flux or the
rotor speed, the output voltage vt also increases by an increase in the flux
or speed.

15

Armature Reaction

So far we have only talked about the magnetic field generated by the field winding (or permanent magnet) which is located on the stator. A magnetic field is
also generated by the armature winding. This field may be neglected as long
as the armature current is not very high. However, when the armature current
increases, it may no longer be negligible. Contrary to the main field that is
stationary, the armature field is rotating in the air-gap. It rotates half-a-cycle
20

and then leaps the other half-a-cycle due to the switch of current by the commutation system. During a quarter of cycle (which is half of its half-a-cycle
rotation), it is opposing the main filed and during the other quarter it is adding
to the main field. Therefore, it may be thought that the overall effect is zero.
However, this is not true because when it adds to the main field, it can easily
saturate the core (contrary to when it opposes the main field). Therefore, the
overall average effect of the armature field is to decrease the effective field of the
machine. It thus contributes to decrease the emf (e = k) and consequently
the output voltage of the machine.
Another impact of the armature reaction is in shifting the neutral axis of the
main field. Therefore, when the brushes short circuit the slip rings, the voltage
is not zero and this causes arc and sparking at the brushes.
The armature reaction can be reduced by using compensating winding (not discussed here). Also, there are ways to reduce arc/sparking at the brushes (not
discussed here).
In the absence of compensating windings, the armature reaction may be modeled
by a percentage decrease in the main field as the armature current goes up.
Reduction of the main field causes reduction of the generated voltage (in a
generator) and increase of the speed (in a motor).

16

Terminal Characteristics of a Shunt DC Generator

The equivalent circuit of the shunt connected dc generator is shown in Fig. 21.
Lets discuss the no-load characteristics first (i.e. when no RL is connected).
The following equations hold true for no-load situation.
e = Ra ia + Rf if , ia = if
Moreover, the curve for the induced voltage e (as shown in Fig. 18) is also valid.
Therefore, the no-load characteristics of this machine may be drawn as shown
in Fig. 22.
Remarks.
The armature resistance Ra is usually much smaller than the field resistance Rf in this type of motor. Therefore, e and vt are very close in
Rf
this case: vt = Rf +R
e e. Moreover, the line (Rf + Ra )if may be
a
approximated by Rf if if necessary.
If the core has an initial residual magnetization (as shown in Fig. 22),
the circuit builds up its output voltage. This is because if and e interact
in a positive feedback fashion: a small e is generated due to the residual
field (e = k), a small (armature and field) current will flow, this flow of
21

ia

RL

it

vf = va = vt

Rf

Ra

if
Lf

La

Figure 21: A self excited shunt dc generator feeding a load.

e

e = (Rf + Ra )if
magnetization curve at a given speed

operating point

if

Figure 22: No-load characteristics of a shunt dc generator and its voltage buildup property
current in the field winding increases the already existing filed, increased
field increases the induced voltage (e = k) even further. That is a
positive feedback which increases e to reach the operating point as shown
in Fig. 22. This process is symbolically shown by the stair-case path.
If the field resistance is too large, the curves intersect at a point that
corresponds to a very small output voltage. Therefore, the field resistance
should not be above a limit, called the critical field resistance, in order for
the voltage to build up.
Now, lets discuss the terminal characteristics of this generator. When a load
RL is connected to the output, Figure 21, the following equations hold.
22

vt = RL it = Rf if = e Ra ia
ia = if + it
vt = e Ra (it +
vt =

vt
Ra
) = e Ra it
vt
Rf
Rf

Rf
Rf
vt
[N (if ) Ra it ] =
[N (
) Ra it ]
Rf + Ra
Rf + Ra
Rf

where N is the nonlinear magnetization curve showing the relationship between

the emf e and the field current if , i.e. e = N (if ). In addition to the above
analysis, also notice that when the output load (it ) increases, the armature
current also goes up. The armature reaction causes the induced voltage to
drop. This drop will cause a subsequent drop in the field current and the flux.
Thus, the effect of armature reaction becomes even stronger in this type of
connection. The nonlinearity caused by the magnetization phenomenon and
the increased armature reaction result in a peculiar terminal characteristics as
shown in Fig. 23.
vt

operating point

vt = RL it
vt = e Ra it

it

Figure 23: Terminal characteristics of a shunt dc generator.

17

An Example on DC Generator

Consider a 12 kW, 240 V, 1200 rpm, separately excited dc generator with armature and field resistors Ra = 0.2 and Rf = 200 . The field is supplied by
a 120 V dc source.
At no-load condition, the required input power to run the generator at rated
speed is one horse power.
The armature reaction can be modeled as 1% reduction in the field at full load.
There is 2 V voltage drop across each brush at full load operation.
23

(a) Find the field current.

(b) Find approximate rotational loss.
(c) Find the full load current.
(d) Find the terminal voltage at no-load.
(e) What is the voltage regulation of the generator?
(f) Calculate the generators efficiency at the full load?
(g) Calculate the input power and the input torque?
(h) Calculate the converted power and the induced torque?
(i) If the machine has 4 magnetic poles, what is the electrical frequency (in
Hz) of the induced voltage before it gets rectified by the brushes?
Solution:
(a) if =

vf
Rf

120
200

= 0.6 A.

(b) Prot = 1 hp=1 746 = 746 W.

(c) iFL =

12000
240

= 50 A.

(d) At full load, eFL = vt +Ra iFL +vbrush = 240+0.250+22 = 254 V. Now,
e = k; k and are constant in this problem. Due to armature reaction,
increases 1% from FL to NL. Therefore, the no-load (or open-circuit)
terminal voltage is vNL = eNL = 1.01 eFL = 1.01 254 = 256.54 V.
FL
100 =
(e) VR= vNLvv
FL

256.54240
240

100 = 6.7%

(f) Armature copper loss: Ra i2a = 0.2 502 = 500 W

Brush copper loss: 2vbrush iFL = 2 2 50 = 200 W.
Pin = Pout + Ploss = 12000 + 746 + 500 + 200 = 13446 W.
=
(g) Pin = 13446 W, in =

Pout
12000
=
100 = 89.25%
Pin
13446

Pin

13446
2 1200
60

= 107 Nm

12700
(h) Pconv = Pin Prot = 13446 746 = 12700 W, ind = Pconv
= 2
1200 =

60
101 Nm
Another way to calculate converted power: Pconv = eFL i = 254 50 =
12700 W
Another way: Pconv = Pout + Pcu + Pbrush = 12000 + 500 + 200 = 12700 W

(i) The mechanical frequency is 1200 rpm that is 20 Hz. The electrical frequency is P2 times the mechanical frequency which will be 2 20 = 40 Hz.
24

18

TC of Separately Excited and Shunt DC Motors

Consider a separately excited dc motor which is supplied with a given field

voltage vf and a given armature voltage va . Therefore, the field current is
v
if = Rff and this establishes a flux . The following equation holds true for the
speed-torque relationship:
= kia = k

va e
k
=
(va k)
Ra
Ra

therefore
=

va
Ra

.
k (k)2

(3)

Assume that the armature voltage (and resistance) and the field are fixed. Equation (3) shows a straight line in the plane known as the terminal characteristics as shown in Fig. 24.

NL

Figure 24: Terminal characteristics of a separately excited dc motor (also valid for
shunt motor).
When approaches zero (towards no load condition) increases and approaches
va
k and this is the no-load speed. As the torque increases, the speed falls down
Ra
with a rate that is equal to (k)
2 . Since this number is normally a small number,
the motor is expected to have good speed regulation.

19

Terminal Characteristics of a Series DC Motor

Equivalent circuit of a series DC motor is shown in Fig. 25. In this motor,

it = ia = if = i
25

Rf

Lf

it
Ra

vt

if = ia = it

La

Figure 25: Equivalent circuit of a self excited series dc motor

and
vt = (Ra + Rf )i + e.
Also, e = k and
= ki = kk1 i2
(assuming a linear magnetization curve where = k1 i). Therefore,

+ k = (Ra + Rf )
+ kk1

vt = (Ra + Rf )
kk1
kk1
kk1
The speed is calculated as
vt
Ra + Rf
=

kk1
kk1
and the terminal characteristics of this motor is shown in Fig. 26.
Remark 1. A dc motor should not operate at no-load condition because its
speed can become dangerously high.
Remark 2. A dc motor can generate a high torque at zero speed making it
suitable for applications which require a high starting torque such as cranes,
electric trains.

20

Speed Control of DC Motor

DC motors are of interest because they offer good speed control over a wide
range. Equation (3) shows that the speed may be changed (controlled) using
three variables: va , and Ra . Three different methods are obtained as discussed
below.
26

Figure 26: Speed-torque characteristics of a series dc motor.

20.1

Armature Voltage Control

In this method, va is used as the control signal to control the speed. Assume
that motor is operating at a given speed driving a given load. Now, if va is increased, ia tends to go up. Then, = kia goes up and this speeds up the rotor.
Higher speed induces higher e = k. Higher e decreases ia and decreases the
induced torque. The machine will settle at a higher speed.

(va : increasing)

Figure 27: Speed control using armature voltage control

It was said that increasing va increases e. The armature current depends on the
difference between these two voltages. If these two voltages change accordingly
such that the armature current can remain relatively constant, the developed
torque remains relatively constant. In other words, the speed control is performed at constant torque. This is a practically desirable situation. Graphically, the motors speed moves along the red, dashed vertical line on Fig. 27. In
order to realize this method, a mechanism must be implemented to ensure that
ia remains constant despite the fact the va is changing. A feedback control
27

loop should be used in this method to keep the armature current constant.
The armature voltage control can provide a smooth control of speed from zero
to the base speed at a constant torque (constant rate of speed). The speed cannot be increased beyond the base value using this method because the motor
will be subject to over-voltage.
For some applications, it is required to keep the speed constant despite the
changing torque of the load. This is sometimes called adjustable speed control
as opposed to the variable speed control discussed above. This objective can
also be achieved using armature voltage control.
This method of control requires adjustable voltage source that has been conventionally expensive but the technology of power electronics provides an economically feasible solution for building a variable dc voltage source.

20.2

Field Control

In this method, the field serves as the controlling signal. This signal can be
changed either by the field voltage vf or the field rheostat Rf . Assume that va
and Ra is constant. For any given value of , the speed-torque equation is given
by the line:
va
Ra
=

.
k (k)2
This line moves upward (and slightly slopes down) when becomes smaller as
shown in Fig. 28. Therefore, decreasing the field increases the no-load speed.
Thats why when the motor is operating at no-load, the field circuit should
never break as this can cause dangerously high speed.

NL
(: decreasing)

Figure 28: Speed control through field control.

In order to understand what happens in this method of control, assume that
the motor is primarily working at speed . Now, slightly reduce the field .
28

Then the induced voltage e = k slightly reduces. This slight change in e

can cause large increase in ia because Ra is very small. Therefore, the induced
torque = kia may increase significantly causing the rotor to speed up. As
the rotor speeds up, e increases and reduces ia and the machine settles at a new
but higher speed.
The field control is good when it is desired to increase the speed beyond the
base value. Once the speed is brought up to the base value (by the armature
voltage control for example), decrease the field slowly and the speed goes up
(this is also called the field reduction method). Notice that by reducing the
field, the armature current keeps increasing if the torque is constant ( = kia ).
Therefore, in order not to damage the armature circuit, the armature current
must be limited which in turn results in the reduction of the induced torque. In
other words, using the field control method, it is possible to use the motor at
speeds higher than its rated speed but lower torques.

20.3

Armature Resistance Control

Based on the equation

=

va
Ra

,
k (k)2

it is obvious that the armature resistance Ra directly controls the slope of the
curve and controls the speed (assuming that the armature voltage and the field
are fixed). Increasing the armature resistance decreases the speed as shown in
Fig. 29.

NL
(Ra : increasing)

Figure 29: Speed control through armature resistance control.

In order to understand this method, assume that the motor is primarily operating at a given load and given speed. Now, slightly increase the armature
resistance. This will reduce the armature current and thus reduce the induced
torque = kia . Thus, the motor slows down. The induced voltage e = k
goes down and the armature current goes up. The armature current eventually

29

settles down at its original value and the torque goes up and settles back to its
original value and the motor settles at a new but lower value of speed.
The disadvantage of this method is the loss of efficiency because the armature
current often causes big losses; moreover the armature rheostat is much more
expensive than field rheostat because it needs to carry large current.
Remark: Although the above control methods are discussed in an open-loop
framework, it is often required to use a closed-loop feedback in order to have
accurate control despite the system uncertainties and disturbances. Discussion
of closed-loop control techniques for dc machines is beyond the scope of this
course.

21

An Example on Separately Excited DC Motor

Consider a 250 V, 1200 rpm separately excited dc motor with armature resistance of 0.03 . Initially, the motor is supplied at va = 250 V, draws a current
ia = 120 A and its speed is n=1103 rpm. What will be the speed of the motor
if the voltage is reduced to 200 V?
(Note: The load on the shaft of the rotor is unchanged. The field is unchanged.)

Solution: The equivalent circuit is shown below.

Rf

La

if

ia

Ra

vf

+
field circuit

Lf

armature circuit

va

When the applied voltage is 250 V, the emf is equal to

e = vt Ra ia = 250 0.03 120 = 246.4 V.
When the voltage is reduced to 200 V, the emf is equal to
e = vt Ra ia = 200 0.03 120 = 196.4 V.
Notice that the armature current remains unchanged in this situation because
the torque and the field are fixed ( = kia ). And finally,
n1
246.4
1103
e1
=

=
n2 = 879 rpm
e2
n2
196.4
n2

30

22

An Example on Shunt DC Motor

Consider a 75 hp shunt dc motor with the armature resistance of Ra = 0.06

and the field resistance of Rf = 50 . The supply voltage is 250 V. At no load,
the machine draws 15 A current and spins at a speed of 1200 rpm. Ignore the
effect of armature reaction and answer the following.
(a) Draw the equivalent circuit of the machine.
(b) What is the current flowing in the field circuit?
(c) What is the armature current at no load condition?
(d) Calculate the rotational losses.
(e) What percentage of the rated power is the rotational loss?
(f) How much power the motor draws from the supply at no load?
(g) What is the efficiency of the machine at no load?
(h) Assume that some load is applied to the shaft of the motor so that an input
current of 100 A is drawn by the motor.
(h1) Find the speed of the motor in this condition.
(h2) Calculate the SR.
(h3) What is the generated torque at this speed?
(h4) What is the efficiency of the machine at this condition?
(i) Assume that the mechanical load is increased such that the input current
becomes 200 A.
(i1) Find the speed of the motor in this condition.
(i2) Calculate the SR.
(i3) What is the generated torque at this speed?
(i4) What is the efficiency of the machine at this condition?
Solution:
(a) Circuit diagram of the motor is shown in Fig. 30.
(b) if =

vt
Rf

250
50

= 5 A.

(c) ia = it if = 15 5 = 10 A.
(d) Rotational loss is equal to the converted power at no load:
Prot = eia = (vt Ra ia )ia = (250 0.06 10) 10 = 249.4 10 = 2494 W
(e) Prated = 75 746 = 55950 W,
2494
Prot
100 =
100 = 4.46%
Prated
55950
(f) At no load: Pin = vt it = 250 15 = 3750 W

31

it

ia

vf = va = vt = 250 V

Rf

Ra

if
Lf

La

Figure 30: Circuit diagram of the shunt dc motor of Example.

(g) At no load, machine delivers no power: Pout = 0, thus =

Pout
Pin

100 = 0%!

(h1) it = 100 A, thus, ia = 95 A which means e = 250 0.06 95 = 244.3 V.

Notice also that e = k and is constant (because armature reaction is neglected), therefore,
e1
1
249.4
1200
=

=
n2 = 1175.5 rpm.
e2
2
244.3
n2
(h2)

1200 1175.5
100 = 2.08%.
1175.5
(h3) As for the generated toque, notice that the converted power is Pconv =
eia = 244.3 95 = 23209 W. The net output mechanical power is equal to this
minus the rotational loss:
SR =

Pout = 23209 2494 = 20715 W.

Also, Pout = . Therefore,
=

20715
= 168.3 Nm.
1175.5 (2/60)

(h4) The input electrical power is Pin = vt it = 250 100 = 25000 W. Thus,
=

Pout
20715
100 =
100 = 82.86%.
Pin
25000

(i1) it = 200 A, thus, ia = 195 A which means e = 250 0.06 195 = 238.3 V.
e1
1
249.4
1200
=

=
n2 = 1146.6 rpm.
e2
2
238.3
n2
32

(i2)

1200 1146.6
100 = 4.45%.
1146.6
(i3) Pconv = eia = 238.3 195 = 46469 W, Pout = Pconv Prot = 46469 2494 =
43975 W and
SR =

Pout = =

43975
= 366 Nm.
1146.6 (2/60)

(i4) The input electrical power is Pin = vt it = 250 200 = 50000 W. Thus,
=

Pout
43975
100 =
100 = 87.95%.
Pin
50000

Machines efficiency is maximum when it operates at its rated condition.

23

DC Motor Starter

When a dc motor is to start operation from rest, its initial emf is zero, and Ra
is small, therefore a large current flows into the armature circuit:
ia =

vt e
vt

(at the start).

Ra
Ra

The starting current can be limited to safe value by the following methods:
Insert an external resistance in the armature circuit at the start and then
remove it when motor settles. This can even be done in steps and by
gradually decreasing the resistance as the motor speeds up.
Apply a small voltage at the start and increase it as the motor speeds
up. This of course needs a variable source. A controlled rectifier (also
called an AC/DC converter) using power electronic switches can provide
a variable (and controlled) dc source from an ac source. A chopper (also
called a DC/DC converter) is another type of power electronic converter
that provides variable dc voltage from a constant dc source.

33