1

DC machines
 2

Preliminary notes
DC power systems are not very common in the existing engineering
practice.
However, DC motors still have many practical applications, such
automobile, aircraft, and portable electronics, in speed control
applications…
An advantage of DC motors is that it is easy to control their speed in a
wide range.
DC generators are quite rare.
Most DC machines are similar to AC machines: i.e. they have AC voltages
and current within them.
DC machines have DC outputs just because they have a mechanism
converting AC voltages to DC voltages at their terminals.
This mechanism is called a commutator; therefore, DC machines are also
called commutating machines.
 3

The simplest DC machine

The simplest DC rotating machine

consists of a single loop of wire
rotating about a fixed axis. The
magnetic field is supplied by the
North and South poles of the
magnet.
Rotor is the rotating part;
Stator is the stationary part.
 4

The simplest DC machine

We notice that the rotor lies in a slot carved

in a ferromagnetic stator core, which,
together with the rotor core, provides a
constant-width air gap between the rotor and
stator.
The reluctance of air is much larger than the
reluctance of core. Therefore, the magnetic
flux must take the shortest path through the
air gap.

As a consequence, the magnetic flux is perpendicular to the rotor surface

everywhere under the pole faces.
Since the air gap is uniform, the reluctance is constant everywhere under the pole
faces. Therefore, magnetic flux density is also constant everywhere under the pole
faces.
 5

The simplest DC machine

1. Voltage induced in a rotating loop

If a rotor of a DC machine is rotated, a voltage will be induced…

The loop shown has sides ab and cd perpendicular to the figure
plane, bc and da are parallel to it.
The total voltage will be a sum of voltages induced on each
segment of the loop.

Voltage on each segment is:

eind   v×B   l (5.5.1)

 6

The simplest DC machine

1) ab: In this segment, the velocity of the wire is tangential to the path
of rotation. Under the pole face, velocity v is perpendicular to the
magnetic field B, and the vector product v x B points into the page.
Therefore, the voltage is
vBlintopageunderthepoleface
eba   v × B   l  
0beyond the pole edges
2) bc: In this segment, vector product v x B is perpendicular to l. Therefore, the
voltage is zero.
3) cd: In this segment, the velocity of the wire is tangential to the path of rotation.
Under the pole face, velocity v is perpendicular to the magnetic flux density B,
and the vector product v x B points out of the page. Therefore, the voltage is
vBloutof pageunderthepoleface (5.6.2)
edc   v × B   l  
0beyond the pole edges
4) da: In this segment, vector product v x B is perpendicular to l. Therefore, the
voltage is zero.
 7

The simplest DC machine

The total induced voltage on the loop is:

etot  eba  ecb  edc  ead (5.7.1)

2vBlunderthe pole faces

etot   (5.7.2)
0beyond the poleedges
When the loop rotates through 1800,
segment ab is under the north pole
face instead of the south pole face.
Therefore, the direction of the voltage
on the segment reverses but its
magnitude remains constant, leading
to the total induced voltage to be
 8

The simplest DC machine

The tangential velocity of the loop’s edges is

v  r (5.8.1)

where r is the radius from the axis of rotation to

the edge of the loop. The total induced voltage:
2r Blunderthe pole faces
etot   (5.8.2)
0beyond the poleedges
The rotor is a cylinder with surface area 2rl.
Since there are two poles, the area of the rotor
under each pole is Ap = rl. Therefore:
2
 A Bunderthe pole faces
etot    p (5.8.3)
0beyond the poleedges
 9

The simplest DC machine

Assuming that the flux density B is constant everywhere in the air gap under the
pole faces, the total flux under each pole is
  Ap B (5.9.1)
The total voltage is
2
 underthe pole faces
etot    (5.9.2)
0beyond the poleedges

The voltage generated in any real machine depends on the following

factors:

1. The flux inside the machine;

2. The rotation speed of the machine;
3. A constant representing the construction of the machine.
 10

The simplest DC machine

2. Getting DC voltage out of a rotating

loop
A voltage out of the loop is alternatively a
constant positive value and a constant
negative value.
One possible way to convert an alternating
voltage to a constant voltage is by adding a
commutator
segment/brush
circuitry to the
end of the loop.
Every time the
voltage of the
loop switches
direction,
contacts switch
connection.
 11

The simplest DC machine

segments

brushes
 12

The simplest DC machine

3. The induced torque in the

rotating loop
Assuming that a battery is connected
to the DC machine, the force on a
segment of a loop is:

F  i  l ×B  (5.12.1)

And the torque on the segment is

  rF sin  (5.12.2)

Where  is the angle between r and F.

Therefore, the torque is zero when
the loop is beyond the pole edges.
 13

The simplest DC machine

When the loop is under the pole faces:

1. Segment ab: Fab  i  l ×B   ilB (5.13.1)

 ab  rF sin   r ilB  sin 90  rilBccw (5.13.2)

2. Segment bc: Fab  i  l ×B   0 (5.13.3)

 ab  rF sin   0 (5.13.4)

3. Segment cd: Fab  i  l ×B   ilB (5.13.5)

 ab  rF sin   r ilB  sin 90  rilBccw (5.13.6)

4. Segment da: Fab  i  l ×B   0 (5.13.7)

 ab  rF sin   0 (5.13.8)
 14

The simplest DC machine

The resulting total induced torque is

 ind   ab   bc   cd   da (5.14.1)

2rilBunderthe pole faces

 ind  (5.14.2)
0beyond the poleedges
Since Ap  rl and   Ap B (5.14.3)

2
  iunderthe pole faces
 ind   (5.14.4)
0beyond the poleedges

The torque in any real machine depends on the following factors:

1. The flux inside the machine;
2. The current in the machine;
3. A constant representing the construction of the machine.
 15
Commutation in a simple 4-loop
DC machine
Commutation is the process of converting the AC voltages and currents in
the rotor of a DC machine to DC voltages and currents at its terminals.
A simple 4-loop DC machine has four complete loops buried in slots curved in the
laminated steel of its rotor. The pole faces are curved to make a uniform air-gap.
The four loops are laid into the slots in a special manner: the innermost wire in
each slot (end of each loop opposite to the “unprimed”) is indicated by a prime.

Loop 1 stretches
between
commutator
segments a and
b, loop 2
stretches
between
segments b and
c…
 16
Commutation in a simple 4-loop
DC machine
At a certain time instance, when t
= 00, the 1, 2, 3’, and 4’ ends of the
loops are under the north pole face
and the 1’, 2’, 3, and 4 ends of the
loops are under the south pole face.
The voltage in each of 1, 2, 3’, and
4’ ends is given by
eind   v × B  × l  vBl (5.16.1)

 positive, out of the page

The voltage in each of 1’, 2’, 3, and 4 ends is
eind   v×B  ×l  vBl positive, intothe page (5.16.2)

If the induced voltage on any side of a loop is (5.16.1), the total voltage at the
brushes of the DC machine is
E  4eatt  0 (5.16.3)
 17
Commutation in a simple 4-loop
DC machine
We notice that there are two parallel paths for current through the
machine! The existence of two or more parallel paths for rotor current is
a common feature of all commutation schemes.
 18
Commutation in a simple 4-loop
DC machine
If the machine keeps rotating, at t = 450, loops 1 and 3 have rotated into the gap
between poles, so the voltage across each of them is zero. At the same time, the
brushes short out the commutator segments ab and cd.

This is ok since
the voltage across
loops 1 and 3 is
zero and only
loops 2 and 4 are
under the pole
faces. Therefore,
the total terminal
voltage is

E  2eatt  45 (5.18.1)

 19
Commutation in a simple 4-loop
DC machine
At t = 900, the loop ends 1’, 2, 3, and 4’ are under
the north pole face, and the loop ends 1, 2’, 3’, and
4 are under the south pole face. The voltages are
built up out of page for the ends under the north
pole face and into the page for the ends under the
south pole face. Four voltage-carrying ends in each
parallel path through the machine lead to the
terminal voltage of

E  4eatt  90 (5.16.3)

We notice that the voltages in loops 1 and 3 have

reversed compared to t = 00. However, the loops’
connections have also reversed, making the total
voltage being of the same polarity.
 20
Commutation in a simple 4-loop
DC machine
When the voltage reverses in a loop, the connections of the loop are also switched
to keep the polarity of the terminal voltage the same.

The terminal voltage of this 4-loop DC

machine is still not constant over time,
although it is a better approximation to a
constant DC level than what is produced
by a single rotating loop.

Increasing the number of loops on the

rotor, we improve our approximation to
perfect DC voltage.

Commutator segments are usually made out of copper bars and the brushes are
made of a mixture containing graphite to minimize friction between segments and
brushes.
21
 22

Example of a commutator…
 23
Problems with commutation in
real DC machines
1. Armature reaction
If the magnetic field windings of a DC machine are connected to the power
source and the rotor is turned by an external means, a voltage will be
induced in the conductors of the rotor. This voltage is rectified and can be
supplied to external loads. However, if a load is connected, a current will
flow through the armature winding. This current produces its own magnetic
field that distorts the original magnetic field from the machine’s poles. This
distortion of the machine’s flux as the load increases is called armature
reaction and can cause two problems:

1) neutral-plane shift: The magnetic neutral plane is the plane within the
machine where the velocity of the rotor wires is exactly parallel to the
magnetic flux lines, so that the induced voltage in the conductors in the
plane is exactly zero.
 24
Problems with commutation in
real DC machines
A two-pole DC machine: initially, the pole flux is
uniformly distributed and the magnetic neutral plane is
vertical.

The effect of the air gap on the pole flux.

When the load is connected, a current – flowing

through the rotor – will generate a magnetic field from
the rotor windings.
 25
Problems with commutation in
real DC machines
This rotor magnetic field will affect the original
magnetic field from the poles. In some places
under the poles, both fields will sum together, in
other places, they will subtract from each other

Therefore, the net magnetic field will not be

uniform and the neutral plane will be shifted.

In general, the neutral plane shifts in the direction

of motion for a generator and opposite to the
direction of motion for a motor. The amount of the
shift depends on the load of the machine.
 26
Problems with commutation in
real DC machines
The commutator must short out the commutator segments right at the
moment when the voltage across them is zero. The neutral-plane shift may
cause the brushes short out commutator segments with a non-zero voltage
across them. This leads to arcing and sparkling at the brushes!
 27
Problems with commutation in
real DC machines
2) Flux weakening.

Most machines operate at flux densities

near the saturation point.
At the locations on the pole surfaces
where the rotor mmf adds to the pole
mmf, only a small increase in flux occurs
(due to saturation).
However, at the locations on the pole
surfaces where the rotor mmf subtracts
from the pole mmf, there is a large
decrease in flux.
Therefore, the total average flux under
the entire pole face decreases.
 28
Problems with commutation in
real DC machines
In generators, flux weakening
reduces the voltage supplied by a
generator.
In motors, flux weakening leads to
increase of the motor speed.
Increase of speed may increase
the load, which, in turns, results in
more flux weakening. Some shunt
DC motors may reach runaway
conditions this way…

Observe a considerable decrease in

the region where two mmfs are
subtracted and a saturation…
 29
Problems with commutation in
real DC machines
2. L di/dt voltages
This problem occurs in
commutator segments being
shorten by brushes and is called
sometimes an inductive kick.
Assuming that the current in the
brush is 400 A, the current in
each path is 200 A. When a
commutator segment is shorted
out, the current flow through that
segment must reverse.
Assuming that the machine is
running at 800 rpm and has 50
commutator segments, each
segment moves under the brush
and clears it again in 0.0015 s.
 30
Problems with commutation in
real DC machines
The rate of change in current in the shorted loop averages
di 400
  266667 A / s
dt 0.0015

Therefore, even with a small

inductance in the loop, a very large
inductive voltage kick L di/dt will be
induced in the shorted commutator
segment.

This voltage causes sparkling at the

brushes.
 31
Solutions to the problems with
commutation
1. Brush Shifting
𝑑𝑖
• In order to mitigate the sparking at brushes due to neutral plane shift and 𝐿
𝑑𝑡
voltages the most trivial approach is to shift position of brushes.
• So, if the neutral plane shift, then shift the brushes to avoid sparking.
• However, this method is not very effective because
• The position pf the brushes needs to be changed for every change in the
load
• Secondly, it does not by any means rectify the problem of flux weakening
 32
Solutions to the problems with
commutation
2. Commutating poles or interpoles
• To avoid sparkling at the brushes while the machine’s load changes, instead of
adjusting the brushes’ position, it is possible to introduce small poles
(commutating poles or interpoles) between the main ones to make the voltage in
the commutating wires to be zero.
• Such poles are located directly over the conductors being commutated and
provide the flux that can exactly cancel the voltage in the coil undergoing
commutation.
• Interpoles do not change the operation
of the machine
• since they are so small that only affect few
conductors being commutated. Flux
weakening is unaffected.
 33
Solutions to the problems with
commutation
2. Commutating poles or interpoles
• Interpole windings are connected in series with the rotor windings.
• As the load increases and the rotor current increases, the magnitude of
neutral-plane shift and the size of Ldi/dt effects increase too increasing the
voltage in the conductors undergoing commutation.
 34
Solutions to the problems with
commutation
• However, the interpole flux increases too producing a larger voltage in
the conductors that opposes the voltage due to neutral-plane shift.
Therefore, both voltages cancel each other over a wide range of loads.
This approach works for both DC motors and generators.
• The interpoles must be of the same polarity as the next upcoming main
pole in a generator;
• The interpoles must be of the same polarity as the previous main pole in
a motor.
• The use of interpoles is very common because they correct the sparkling
problems of DC machines at a low cost. However, since interpoles do
nothing with the flux distribution under the pole faces, flux-weakening
problem is still present.
 35
Solutions to the problems with
commutation
3. Compensating windings

The flux weakening problem can be very severe for large DC motors.
Therefore, compensating windings can be placed in slots carved in the
faces of the poles parallel to the rotor conductors. These windings are
connected in series with the rotor windings, so when the load changes in
the rotor, the current in the compensating winding changes too…
 36
Solutions to the problems with
commutation

Rotor
Pole and
flux comp.
fluxes

Sum of these three fluxes equals to the

original pole flux.
 37
Solutions to the problems with
commutation
• The mmf due to the
compensating windings is equal
and opposite to the mmf of the
rotor. These two mmfs cancel
each other, such that the flux in
the machine is unchanged.

• The main disadvantage of

compensating windings is
that they are expensive
since they must be
machined into the faces of
the poles. Also, any motor
with compensative windings
must have interpoles to
cancel L di/dt effects.
 38
Solutions to the problems with
commutation
A stator of a
six-pole DC
machine with
interpoles and
compensating
windings.

pole
interpole
 39

Power flow and losses in DC machines

Unfortunately, not all electrical power is converted to mechanical power by a motor

and not all mechanical power is converted to electrical power by a generator…

The efficiency of a DC machine is:

Pout
 100% (5.36.1)
Pin
or
Pin  Ploss
 100% (5.36.2)
Pin
 40

The losses in DC machines

There are five categories of losses occurring in DC machines.

1. Electrical or copper losses – the resistive losses in the armature and field
windings of the machine.

Armature loss: PA  I A2 RA (5.37.1)

Field loss: PF  I F2 RF (5.37.2)

Where IA and IF are armature and field currents and RA and RF are armature and
field (winding) resistances usually measured at normal operating temperature.
 41

The losses in DC machines

2. Brush (drop) losses – the power lost across the contact potential at the
brushes of the machine.

PBD  VBD I A (5.38.1)

Where IA is the armature current and VBD is the brush voltage drop. The voltage
drop across the set of brushes is approximately constant over a large range of
armature currents and it is usually assumed to be about 2 V.

Other losses are exactly the same as in AC machines…

 42

The losses in DC machines

3. Core losses – hysteresis losses and eddy current losses. They vary as B2
(square of flux density) and as n1.5 (speed of rotation of the magnetic field).

4. Mechanical losses – losses associated with mechanical effects: friction

(friction of the bearings) and windage (friction between the moving parts of the
machine and the air inside the casing). These losses vary as the cube of rotation
speed n3.

5. Stray (Miscellaneous) losses – losses that cannot be classified in any of the

previous categories. They are usually due to inaccuracies in modeling. For many
machines, stray losses are assumed as 1% of full load.
 43

The power-flow diagram

On of the most convenient technique to account for power losses in a
machine is the power-flow diagram.

For a DC
motor:

Electrical power is input to the machine, and the electrical and brush losses must be
subtracted. The remaining power is ideally converted from electrical to mechanical
form at the point labeled as Pconv.
 44

The power-flow diagram

The electrical power that is converted is

Pconv  E A I A (5.41.1)

And the resulting mechanical power is

Pconv   ind m (5.41.2)

After the power is converted to mechanical form, the stray losses, mechanical
losses, and core losses are subtracted, and the remaining mechanical power is
output to the load.
 45

Equivalent circuit of a DC motor

The armature circuit (the entire

rotor structure) is represented by
an ideal voltage source EA and a
resistor RA. A battery Vbrush in the
opposite to a current flow in the
machine direction indicates brush
voltage drop.
The field coils producing the
magnetic flux are represented by
inductor LF and resistor RF. The
resistor Radj represents an
external variable resistor
(sometimes lumped together with
the field coil resistance) used to
control the amount of current in
the field circuit.
 46

Equivalent circuit of a DC motor

Sometimes, when the brush drop voltage is small, it may be left out. Also, some
DC motors have more than one field coil…
The internal generated voltage in the machine is

EA  K (5.43.1)

The induced torque developed by the machine is

 ind  K I A (5.43.2)

Here K is the constant depending on the design of a particular DC machine (number

and commutation of rotor coils, etc.) and  is the total flux inside the machine.

Note that for a single rotating loop K = /2.

 47

Magnetization curve of a DC machine

The internal generated voltage EA is directly proportional to the flux in the machine
and the speed of its rotation.
The field current in a DC machine produces a field mmf F = NFIF, which produces
a flux in the machine according to the magnetization curve.

or in terms
of internal
voltage vs.
field
current for
a given
speed.

To get the maximum possible power per weight out of the machine, most motors
and generators are operating near the saturation point on the magnetization curve.
Therefore, when operating at full load, often a large increase in current IF may be
needed for small increases in the generated voltage EA.
 48
Motor types: Separately excited
and Shunt DC motors
Note: when
the voltage to
the field circuit
is assumed
constant,
there is no
difference
between
them…
Separately excited DC motor:
a field circuit is supplied from a Shunt DC motor:
separate constant voltage power a field circuit gets its power from the
source. armature terminals of the motor.
For the armature circuit of these motors:

VT  EA  I A RA (5.45.1)
 49

Shunt motor: terminal characteristic

 50

Shunt motor: terminal characteristic

A terminal characteristic of a machine is a plot of the machine’s output

quantities vs. each other.
For a motor, the output quantities are shaft torque and speed. Therefore, the
terminal characteristic of a motor is its output torque vs. speed.

If the load on the shaft increases, the load torque load will exceed the induced
torque ind, and the motor will slow down. Slowing down the motor will decrease
its internal generated voltage (since EA = K), so the armature current
increases (IA = (VT – EA)/RA). As the armature current increases, the induced
torque in the motor increases (since ind = KIA), and the induced torque will
equal the load torque at a lower speed .

VT RA
   (5.46.1)
K  K  2 ind
 51

Shunt motor: terminal characteristic

Assuming that the terminal voltage and other terms are constant, the motor’s
speed vary linearly with torque.

However, if a motor has an armature reaction, flux-weakening reduces the flux

when torque increases. Therefore, the motor’s speed will increase. If a shunt (or
separately excited) motor has compensating windings, and the motor’s speed
and armature current are known for any value of load, it’s possible to calculate
the speed for any other value of load.
 52
Shunt motor: terminal characteristic
– Example
Example 5.1: A 50 hp, 250 V, 1200 rpm DC shunt motor with compensating windings
has an armature resistance (including the brushes, compensating windings, and
interpoles) of 0.06 . Its field circuit has a total resistance Radj + RF of 50 , which
produces a no-load speed of 1200 rpm. The shunt field winding has 1200 turns per
pole. a) Find the motor speed when its input current is 100 A.
b) Find the motor speed when its input current is 200 A.
c) Find the motor speed when its input current is 300 A.
d) Plot the motor torque-speed characteristic.
 53
Shunt motor: terminal characteristic
– Example
The internal generated voltage of a DC machine (with its speed expressed in rpm):

EA  K
Since the field current is constant (both field resistance and VT are constant) and
since there are no armature reaction (due to compensating windings), we
conclude that the flux in the motor is constant. The speed and the internal
generated voltages at different loads are related as
E A2 K2 n2
 
E A1 K1 n1
EA2
Therefore: n2  n1
E A1
At no load, the armature current is zero and therefore EA1 = VT = 250 V.
 54
Shunt motor: terminal characteristic
– Example
a) Since the input current is 100 A, the armature current is
VT 250
I A  IL  IF  IL   100   95 A
RF 50
Therefore: EA  VT  I A RA  250  95  0.06  244.3V
and the resulting motor speed is:

EA2 244.3
n2  n1  1200  1173rpm
E A1 250
b) Similar computations for the input current of 200 A lead to n2 = 1144 rpm.
c) Similar computations for the input current of 300 A lead to n2 = 1115 rpm.
d) To plot the output characteristic of the motor, we need to find the torque
corresponding to each speed. At no load, the torque is zero.
 55
Shunt motor: terminal characteristic
– Example
Since the induced torque at any load is related to the power converted in a DC
motor:
Pconv  EA I A   ind 
EA I A
the induced torque is  ind 

2443  95
For the input current of 100 A:  ind   190 N - m
2 1173 / 60
2383 195
For the input current of 200 A:  ind   388 N - m
2 1144 / 60
2323  295
For the input current of 300 A:  ind   587 N - m
2 1115 / 60
 56
Shunt motor: terminal characteristic
– Example
The torque-speed
characteristic of the motor is:
 57

Shunt motor: Nonlinear analysis

The flux  and, therefore the internal generated voltage EA of a DC machine are
nonlinear functions of its mmf and must be determined based on the magnetization
curve. Two main contributors to the mmf are its field current and the armature
reaction (if present).
Since the magnetization curve is a plot of the generated voltage vs. field current,
the effect of changing the field current can be determined directly from the
magnetization curve.
If a machine has armature reaction, its flux will reduce with increase in load. The
total mmf in this case will be

Fnet  N F I F  FAR (5.53.1)

It is customary to define an equivalent field current that would produce the same
output voltage as the net (total) mmf in the machine:
FAR
I  IF 
*
F
(5.53.2)
NF
 58

Shunt motor: Nonlinear analysis

Conducting a nonlinear analysis to determine the internal generated voltage in a

DC motor, we may need to account for the fact that a motor can be running at a
speed other than the rated one.
The voltage induced in a DC machine is

EA  K (5.54.1)

For a given effective field current, the flux in the machine is constant and the
internal generated voltage is directly proportional to speed:

EA n
 (5.54.2)
E A0 n0
Where EA0 and n0 represent the reference (rated) values of voltage and speed,
respectively. Therefore, if the reference conditions are known from the magnetization
curve and the actual EA is computed, the actual speed can be determined.
 59
Shunt motor: Nonlinear analysis –
Example
Example 5.2: A 50 hp, 250 V, 1200 rpm DC shunt motor without compensating
windings has an armature resistance (including the brushes and interpoles) of 0.06
. Its field circuit has a total resistance Radj + RF of 50 , which produces a no-load
speed of 1200 rpm. The shunt field winding has 1200 turns per pole. The armature
reaction produces a demagnetizing mmf of 840 A-turns at a load current of 200A.
The magnetization curve is shown.

a) Find the motor speed when its input

current is 200 A.
b) How does the motor speed compare to
the speed of the motor from Example 5.1
(same motor but with compensating
windings) with an input current of 200 A?
c) Plot the motor torque-speed
characteristic.
 60
Shunt motor: Nonlinear analysis –
Example
a) Since the input current is 200 A, the armature current is
VT 250
I A  IL  IF  IL   200   195 A
RF 50
Therefore: EA  VT  I A RA  250  195  0.06  238.3V
At the given current, the demagnetizing mmf due to armature reaction is 840 A-
turns, so the effective shunt field current of the motor is
FAR 840
I  IF 
*
F  5  4.3 A
NF 1200
From the magnetization curve, this effective field current will produce an internal
voltage of EA0 = 233 V at a speed of 1200 rpm. For the actual voltage, the speed is

EA 238.3
n n0  1200  1227rpm
E A0 233
 61
Shunt motor: Nonlinear analysis –
Example
b) A speed of a motor with compensating windings was 1144 rpm when the input
current was 200 A. We notice that the speed of the motor with armature reactance
is higher than the speed of the motor without armature reactance. This increase is
due to the flux weakening.

c) Assuming that the mmf due to

the armature reaction varies
linearly with the increase in current,
and repeating the same
computations for many different
load currents, the motor’s torque-
speed characteristic can be plotted.
 62

Shunt motor: Speed control

There are two methods to control the speed of a shunt DC motor:
1. Adjusting the field resistance RF (and thus the field flux)
2. Adjusting the terminal voltage applied to the armature

1. Adjusting the field resistance

1) Increasing field resistance RF decreases the field current (IF = VT/RF);
2) Decreasing field current IF decreases the flux ;
3) Decreasing flux decreases the internal generated voltage (EA = K);
4) Decreasing EA increases the armature current (IA = (VT – EA)/RA);
5) Changes in armature current dominate over changes in flux; therefore,
increasing IA increases the induced torque (ind = KIA);
6) Increased induced torque is now larger than the load torque load and, therefore,
the speed  increases;
7) Increasing speed increases the internal generated voltage EA;
8) Increasing EA decreases the armature current IA…
9) Decreasing IA decreases the induced torque until ind = load at a higher speed .
 63

Shunt motor: Speed control

The effect of increasing the field

resistance within a normal load
range: from no load to full load.

Increase in the field resistance

increases the motor speed. Observe
also that the slope of the speed-torque
curve becomes steeper when field
resistance increases.
 64

Shunt motor: Speed control

The effect of increasing the field

resistance with over an entire load
range: from no-load to stall.

At very slow speeds (overloaded

motor), an increase in the field
resistance decreases the speed. In this
region, the increase in armature current
is no longer large enough to
compensate for the decrease in flux.

Some small DC motors used in control circuits may operate at speeds close to stall
conditions. For such motors, an increase in field resistance may have no effect (or
opposite to the expected effect) on the motor speed. The result of speed control by
field resistance is not predictable and, thus, this type of control is not very common.
 65

Shunt motor: Speed control

2. Changing the armature voltage

This method implies changing the voltage applied to the armature of the
motor without changing the voltage applied to its field. Therefore, the
motor must be separately excited to use armature voltage control.

Armature
voltage speed
control
 66

Shunt motor: Speed control

1) Increasing the armature voltage VA increases the armature current (IA = (VA -
EA)/RA);
2) Increasing armature current IA increases the induced torque ind (ind = KIA);
3) Increased induced torque ind is now larger than the load torque load and,
therefore, the speed ;
4) Increasing speed increases the internal generated voltage (EA = K);
5) Increasing EA decreases the armature current IA…
6) Decreasing IA decreases the induced torque until ind = load at a higher speed .

Increasing the armature voltage of a

separately excited DC motor does not
change the slope of its torque-speed
characteristic.
 67

Shunt motor: Speed control

If a motor is operated at its rated terminal voltage, power, and field current, it will be
running at the rated speed also called a base speed.
Field resistance control can be used for speeds above the base speed but not
below it. Trying to achieve speeds slower than the base speed by the field circuit
control, requires large field currents that may damage the field winding.
Since the armature voltage is limited to its rated value, no speeds exceeding the
base speed can be achieved safely while using the armature voltage control.
Therefore, armature voltage control can be used to achieve speeds below the base
speed, while the field resistance control can be used to achieve speeds above the
base speed.

Shunt and separately excited DC motors have excellent speed control

characteristic.
 68

Shunt motor: Speed control

For the armature voltage control, the flux in the motor is constant. Therefore, the
maximum torque in the motor will be constant too regardless the motor speed:

 max  K I A,max (5.64.1)

Since the maximum power of the motor is

Pmax   max (5.64.2)

The maximum power out of the motor is directly proportional to its speed.

For the field resistance control, the maximum power out of a DC motor is
constant, while the maximum torque is reciprocal to the motor speed.
 69

Shunt motor: Speed control

• Speed control via changing 𝑅𝐹 involves reduction of flux.

• The speed increase is due to the flux decrease.
• In order to keep 𝐼𝐴 in check the torque has to be decreased as the
speed increases.
• The increase in speed and the decrease in torque is equal hence the
maximum power out of the DC machine is constant.
 70

Shunt motor: Speed control

Torque and power limits as functions of motor speed for a shunt (or separately
excited) DC motor.
 71

Shunt motor: Speed control: Ex

Example 5.3: A 100 hp, 250 V, 1200 rpm DC shunt motor with an armature
resistance of 0.03  and a field resistance of 41.67 . The motor has compensating
windings, so armature reactance can be ignored. Mechanical and core losses may
be ignored also. The motor is driving a load with a line current of 126 A and an initial
speed of 1103 rpm. Assuming that the armature current is constant and the
magnetization curve is

a) What is the motor speed if the field

resistance is increased to 50 ?
b) Calculate the motor speed as a function
of the field resistance, assuming a
constant-current load.
c) Assuming that the motor next is
connected as a separately excited and is
initially running with VA = 250 V, IA = 120 A
and at n = 1103 rpm while supplying a
constant-torque load, estimate the motor
speed if VA is reduced to 200 V.
 72

Shunt motor: Speed control: Ex

shunt separately-excited
For the given initial line current of 126 A, the initial armature current will be
250
I A1  I L1  I F 1  126   120 A
41.67
Therefore, the initial generated voltage for the shunt motor will be

EA1  VT  I A1RA  250  120  0.03  246.4V

 73

Shunt motor: Speed control: Ex

After the field resistance is increased to 50 Ω, the new field current will be
250
IF 2   5 A
50
The ratio of the two internal generated voltages is
E A 2 K22 2 n2
 
E A1 K11 1n1
Since the armature current is assumed constant, EA1 = EA2 and, therefore
1n1
n2 
2
The values of EA on the magnetization curve are directly proportional to the flux.
Therefore, the ratio of internal generated voltages equals to the ratio of the fluxes
within the machine. From the magnetization curve, at IF = 5A, EA1 = 250V, and at
IF = 6A, EA1 = 268V. Thus:
 74

Shunt motor: Speed control: Ex

1n1 E A1n1 268

n2    1103  1187rpm
2 EA2 250

b) A speed vs. RF characteristic

is shown below:
 75

Shunt motor: Speed control: Ex

c) For a separately excited motor, the initial generated voltage is

EA1  VT 1  I A1RA
E A 2 K22 2 n2
Since  
E A1 K11 1n1
E A 2 n1
and since the flux  is constant n2 
E A1
Since the both the torque and the flux are constants, the armature current IA is
also constant. Then

VT 2  I A 2 RA 200  120  0.03

n2  n1  1103  879rpm
VT 1  I A1 RA 250  120  0.03
 76
Shunt motor: The effect of an
open field circuit

If the field circuit is left open on a shunt motor, its field resistance will be
infinite. Infinite field resistance will cause a drastic flux drop and, therefore,
a drastic drop in the generated voltage. The armature current will be
increased enormously increasing the motor speed.

A similar effect can be caused by armature reaction. If the armature

reaction is severe enough, an increase in load can weaken the flux
causing increasing the motor speed. An increasing motor speed increases
its load, which increases the armature reaction weakening the flux again.
This process continues until the motor overspeeds. This condition is called
runaway.
 77
Motor types: The permanent-magnet
DC motor
A permanent magnet DC (PMDC) motor is a motor whose poles are
made out of permanent magnets.
Advantages:
1. Since no external field circuit is needed, there are no field circuit copper
losses;
2. Since no field windings are needed, these motors can be considerable
smaller.
Disadvantages:
1. Since permanent magnets produces weaker flux
densities then externally supported shunt fields,
such motors have lower induced torque.
2. There is always a risk of demagnetization from
extensive heating or from armature reaction
effects (via armature mmf).
 78
Motor types: The permanent-
magnet DC motor
Normally (for cores), a ferromagnetic However, a maximally large residual
material is selected with small residual flux Bres and large coercive
flux Bres and small coercive magnetizing intensity HC are desirable
magnetizing intensity HC. for permanent magnets forming the
poles of PMDC motors…
 79
Motor types: The permanent-
magnet DC motor
A comparison of magnetization
curves of newly developed
permanent magnets with that of a
conventional ferromagnetic alloy
(Alnico 5) shows that magnets made
of such materials can produce the
same residual flux as the best
ferromagnetic cores.

Design of permanent-magnet DC
motors is quite similar to the design
of shunt motors, except that the flux
of a PMDC motor is fixed.
Therefore, the only method of speed
control available for PMDC motors
is armature voltage control.
 80

Motor types: The series DC motor

A series DC motor is a DC motor whose field windings consists of a

relatively few turns connected in series with armature circuit. Therefore:

VT  EA  I A ( RA  Rs ) (5.75.1)
 81

Series motor: induced torque

The terminal characteristic of a series DC motor is quite different from that of the
shunt motor since the flux is directly proportional to the armature current
(assuming no saturation). An increase in motor flux causes a decrease in its
speed; therefore, a series motor has a dropping torque-speed characteristic.
The induced torque in a series machine is

 ind  K I A (5.76.1)

Since the flux is proportional to the armature current:

  cI A (5.76.2)

where c is a proportionality constant. Therefore, the torque is

 ind  KcI A2 (5.76.3)

Torque in the motor is proportional to the square of its armature current. Series
motors supply the highest torque among the DC motors. Therefore, they are used
as car starter motors, elevator motors etc.
 82

Series motor: terminal characteristic

Assuming first that the magnetization curve is linear and no saturation occurs, flux
is proportional to the armature current:

  cI A (5.77.1)

Since the armature current is

 ind
IA  (5.77.2)
Kc
and the armature voltage EA  K (5.77.3)

The Kirchhoff’s voltage law would be

 ind
VT  EA  I A ( RA  RS )  K   RA  RS  (5.77.4)
Kc
K 2
Since (5.77.1), the torque:  ind  KcI  
2
A
(5.77.5)
c
 83

Series motor: terminal characteristic

Therefore, the flux in the motor is

c
  ind (5.78.1)
K
The voltage equation (5.77.4) then becomes

c  ind
VT  K  ind    RA  RS  (5.78.2)
K Kc
which can be solved for the speed:

VT 1 RA  RS
  (5.78.3)
Kc  ind Kc
The speed of unsaturated series motor inversely proportional
to the square root of its torque.
 84

Series motor: terminal characteristic

One serious disadvantage of

a series motor is that its
speed goes to infinity for a
zero torque.

In practice, however, torque

never goes to zero because
of the mechanical, core, and
stray losses. Still, if no other
loads are attached, the
motor will be running fast
enough to cause damage.

Steps must be taken to ensure that a series motor always has a load! Therefore,
it is not a good idea to connect such motors to loads by a belt or other mechanism
that could break.
 85
Series motor: terminal
characteristic – Example
Example 5.4: A 250 V series DC motor with compensating windings has a total
series resistance RA + RS of 0.08 . The series field consists of 25 turns per pole
and the magnetization curve is
a) Find the speed and induced
torque of this motor when its
armature current is 50 A.
b) Calculate and plot its torque-
speed characteristic.
a) To analyze the behavior of a series
motor with saturation, we pick points
along the operating curve and find the
torque and speed for each point. Since
the magnetization curve is given in
units of mmf (ampere-turns) vs. EA for
a speed of 1200 rpm, calculated
values of EA must be compared to
equivalent values at 1200 rpm.
 86
Series motor: terminal
characteristic – Example
For IA = 50 A

EA  VT  I A ( RA  RS )  250  50  0.08  246V

Since for a series motor IA = IF = 50 A, the mmf is
F  NI  25  50  1250 A  turns
From the magnetization curve, at this mmf, the internal generated voltage is
EA0 = 80 V. Since the motor has compensating windings, the correct speed of the
motor will be
EA 246
n n0  1200  3690rpm
E A0 80
EA I A 246  50
The resulting torque:  ind    31.8 N -m
 3690  2 60
 87
Series motor: terminal
characteristic – Example
b) The complete torque-
speed characteristic

We notice severe over-

speeding at low torque
values.
 88

Series motor: Speed control

The only way to control speed of a series DC motor is by

changing its terminal voltage, since the motor speed is
directly proportional to its terminal voltage for any given
torque.
 89

Motor types: Compounded DC motor

A compounded DC motor is a motor with both a shunt and a series field.

Current flowing into a dotted

end of a coil (shunt or
series) produces a positive
mmf. Long-shunt
connection
If current flows into the
dotted ends of both coils, the
resulting mmfs add to
produce a larger total mmf –
cumulative compounding.

If current flows into the dotted end of

one coil and out of the dotted end of Short-shunt
another coil, the resulting mmfs connection
subtract – differential compounding.
 90

Motor types: Compounded DC motor

The Kirchhoff’s voltage law equation for a compounded DC motor is

VT  EA  I A  RA  RS  (5.85.1)

The currents in a compounded DC motor are

I A  IL  IF (5.85.2)

VT
IF  (5.85.3)
RF
Cumulatively compounded
The mmf of a compounded DC motor:

Fnet FF  FSE  FAR (5.85.4)

Differentially compounded

The effective shunt field current in a compounded DC motor:

N SE FAR
I  IF 
*
F IA  (5.85.5)
NF NF Number of turns
 91
Cumulatively compounded motors:
torque-speed characteristic
In a cumulatively compounded motor, there is a constant component of flux and a
component proportional to the armature current (and thus to the load). These motors
have a higher starting torque than shunt motors (whose flux is constant) but lower
than series motors (whose flux is proportional to the armature current).
The series field has a small effect at light loads – the motor behaves as a shunt
motor. The series flux becomes quite large at large loads – the motor acts like a
series motor.
Similar (to the previously discussed)
approach is used for nonlinear analysis of
compounded motors.
 92
Differentially compounded motors:
torque-speed characteristic
Since the shunt mmf and series mmf subtract from each other in a differentially
compounded motor, increasing load increases the armature current IA and
decreases the flux. When flux decreases, the motor speed increases further
increasing the load. This results in an instability (much worse than one of a shunt
motor) making differentially compounded motors unusable for any applications.

In addition to that, these motors are not

easy to start… The motor typically
remains still or turns very slowly
consuming enormously high armature
current.

Stability problems and huge starting

armature current lead to these motors
being never used intentionally.
 93

Compounded DC motor: Example

Example 5.5: A 100 hp, 250 V compounded DC motor with compensating windings
has an internal resistance, including the series winding of 0.04 . There are 1000
turns per pole on the shunt field and 3 turns per pole on the series windings. The
magnetization curve is shown below.
The field resistor has been adjusted for the
motor speed of 1200 rpm. The mechanical,
core, and stray losses may be neglected.
a) Find the no-load shunt field current.
Find the speed at IA = 200 A if the motor is b)
cumulatively; c) differentially compounded
 94

Compounded DC motor: Example

a) At no load, the armature current is zero; therefore, the internal generated voltage
equals VT = 250 V. From the magnetization curve, a field current of 5 A will produce
a voltage EA = 250 V at 1200 rpm. Therefore, the shunt field current is 5 A.
b) When the armature current is 200 A, the internal generated voltage is

EA  VT  I A ( RA  RS )  250  200  0.04  242V

The effective field current of a cumulatively compounded motor will be
N SE F 3
I F*  I F  I A  AR  5  200  5.6 A
NF NF 1000
From the magnetization curve, EA0 = 262 V at speed n0 = 1200 rpm. The actual
motor speed is

EA 242
n n0  1200  1108rpm
E A0 262
 95

Compounded DC motor: Example

c) The effective field current of a differentially compounded motor will be

N SE FAR 3
I  IF 
*
F IA   5 200  4.4 A
NF NF 1000

From the magnetization curve, EA0 = 236 V at speed n0 = 1200 rpm. The actual
motor speed is

EA 242
n n0  1200  1230rpm
E A0 236
 96
Cumulatively compounded
motors: speed control

The same two techniques that have been discussed for a shunt motor
are also available for speed control of a cumulatively compounded
motor.

1. Adjusting the field resistance RF;

2. Adjusting the armature voltage VA.

The details of these methods are very similar to already discussed for
shunt DC motors.
 97

DC motor starters

In order for DC motors to function properly, they must have some

special control and protection equipment associated with them. The
purposes of this equipment are:

1. To protect the motor against damage due to short

circuits in the equipment;
2. To protect the motor against damage from long-term
overloads;
3. To protect the motor against damage from excessive
starting currents;
4. To provide a convenient manner in which to control the
operating speed of the motor.
 98

DC motor problems on starting

At starting conditions, the motor is not turning, therefore the internal

generated voltage EA = 0V. Since the internal resistance of a normal DC
motor is very low (3-6 % pu), a very high current flows.
For instance, for a 50 hp, 250 V DC motor with armature resistance RA of 0.06 
and a full-load current about 200 A, the starting current is

VT  E A 250  0
IA    4167
 A
RA 0.06
This current is over 20 times the motor’s rated full-load current and may severely
damage the motor.
A solution to the problem of excessive starting current is to insert a starting
resistor in series with the armature to limit the current until EA can build up to limit
the armature current. However, this resistor must be removed from the circuit as
the motor speed is high since otherwise such resistor would cause losses and
would decrease the motor’s torque-speed characteristic.
 99

DC motor problems on starting

In practice, a starting resistor is made up of a series of resistors that can be

successively removed from the circuit as the motor speeds up.

A shunt motor with an extra

starting resistor that can be cut out
of the circuit in segments by
closing the 1A, 2A, and 3A
contacts.

Therefore, two considerations are

needed to be taken into account:
Select the values and the number
of resistor segments needed to
limit the starting current to desired
ranges; Design a control circuit shutting the resistor bypass contacts at the proper
time to remove particular parts of the resistor from the circuit.
 102

DC motor problems on starting: Ex

The resistance R1 must be switched out of the circuit when the armature current
falls to
VT  EA,1
I A,min   I min  350 A
Rtot
After the resistance R1 is out of the circuit, the armature current must increase to
VT  EA,2
I A,max   I max  700 A
Rtot ,1
Since EA = K, the quantity VT – EA must be constant when the resistance is
switched out. Therefore
I min Rtot  VT  E A  I max Rtot ,1
The resistance left in the circuit is
n
I min  I min 
Rtot ,1  Rtot Rtot ,n    Rtot
I max  I max 
 106

DC motor starting circuits

Several different schemes can be used to short contacts and cut out the sections
of a starting resistor. Some devices commonly used in motor-control circuits are
Spring-type push button switches
Fuses: Relay: a
protects main coil
against short and a
circuits number of
contacts

Time delay Overload: a

relay similar to heater coil
ordinary relay and normally
except for closed
having contacts
adjustable
time delay.
 107

DC motor starting circuits

A common DC motor starting circuit:

A series of time delay relays shut contacts

removing each section of the starting resistor at
approximately correct times.

Notice that the relay 1TD is energized at the

same time as the motor starts – contacts of 1TD
will shut a part of the starting resistor after some
time. At the same instance, relay 2TD is
energized and so on…

Observe also 4 fuses protecting different parts of

the circuit and the overload in series with the
armature winding.
 108

DC motor starting circuits

Another type of motor starter:

A series of relays sense the value of armature

voltage EA and cut out the starting resistors as it
riches certain values.

This starter type is more robust to different loads.

FL is the field loss relay: if the field is lost for any

reason, power to the M relay will be turned off.

Armature
current in a
DC motor
during
starting.
 109

DC motor efficiency calculations

To estimate the efficiency of a DC motor, the following losses must be determined:

1. Copper losses;
2. Brush drop losses;
3. Mechanical losses;
4. Core losses;
5. Stray losses.
To find the copper losses, we need to know the currents in the motor and two
resistances. In practice, the armature resistance can be found by blocking the rotor
and a small DC voltage to the armature terminals: such that the armature current
will equal to its rated value. The ratio of the applied voltage to the armature current
is approximately RA.
The field resistance is determined by supplying the full-rated field voltage to the
field circuit and measuring the resulting field current. The field voltage to field
current ratio equals to the field resistance.
 110

DC motor efficiency calculations

Brush drop losses are frequently lumped together with copper losses. If treated
separately, brush drop losses are a product of the brush voltage drop VBD and the
armature current IA.

The core and mechanical losses are usually determined together. If a motor is
running freely at no load and at the rated speed, the current IA is very small and the
armature copper losses are negligible. Therefore, if the field copper losses are
subtracted from the input power of the motor, the remainder will be the mechanical
and core losses. These two losses are also called the no-load rotational losses. As
long as the motor’s speed remains approximately the same, the no-load rotational
losses are a good estimate of mechanical and core losses in the machine under
load.