SECTION 8

DC DRIVES
INTRODUCTION
A major area of power electronic application is the dc and ac motor drives. A variety
of power switching circuits are employed for the drive and control of dc and ac motors. In this
section we see the basic defining equations of the different types of dc motors, their output
characteristics under conventional excitation and the various possible methods of their control
with power electronic controllers.
SEPARATELY EXCITED DC MOTOR
The bulk of the motors used with power converters are either separately excited dc mo-
tors (for constant torque and constant horse power applications in manufacturing industries),
and the series excited dc motors (for traction applications).
Figure 1 shows the schematic diagram of a separately excited dc motor. The excitation
field and the armature field are normal to each other. The field current determines the excita-
tion flux in the machine. The armature current determines the armature flux in the ma-
e

a
chine. The defining equations of the motor are
field circuit equation v
e
R
e
i
e
+ L
e
d i
e
dt
armature circuit equation v
a
R
a
i
a
+ L
a
d i
a
dt
+ e
a
mechanical system equation T
g
T
L
+ B + J
d
dt
back emf equation e
a
k
1
i
e

torque equation T
g
k
2
i
e
i
a
Comparing the back emf and the torque equation, we get , because the electrical input k
1
k
2
will be equal to the mechanical output . The system equations may be put in the e
a
i
a
T
g

following form.

e
di
e
d t

v
e
R
e
i
e

a
di
a
d t

v
a
R
a
i
a

e
a
R
a

m
d
dt

T
L
B
+
T
g
B
E6 201 Power Electronics
EE 185 IISc
Tg
T
L
Ra
La
e Va
ia
ie
Re
Le
Ve
J, B

Fig. 1 Separately Excited DC Motor

and are defined as the field electrical time constant, armature electrical time
e
,
a

m
constant and the load mechanical time constant of the system. The independent inputs to the
system are and . The output states of the system are and . The field circuit v
e
, v
a
T
L
i
e
, i
a

equation is seen to be an independent equation. The armature circuit equation and the
mechanical system equations are seen to be coupled to each other through the back emf and
torque relationships. For any set of inputs and , the above set of differential v
e
, v
a
T
L
equations may be integrated to obtain the dynamic performance of the machine.
Steady state relationships
Under steady state, the above dynamic relationships reduce to
V
e
I
e
R
e
V
a
kI
e
+ I
a
R
a
T
g
T
L
+ B
Armature control
In the armature control of the separately excited dc motor, the field current is kept con-
stant at the rated value . The armature voltage applied to the machine is controlled. Under I
e
such a control, we may combine the above armature and mechanical system equations to obtain
the following.
kI
e
V
a
R
a
k
2
I
e
2
R
a
T
L
+ B
a V
a
b T
L
The above equation defines the relationship between load torque and motor speed for any
given armature input voltage ( ) under a given excitation ( ). The output characteristics ( V
a
I
e
) of the armature controlled separately excited dc motor is shown in Fig. 2. vs T
L
Field control
The field control operation is employed for speeds above rated speed ( ). In V
a
/kI
e
this range of control the armature voltage is maintained at and the field current is reduced V
a
below the rated value of . The steady state equation is then I
e
kI
e
V
a
R
a
k
2
I
e
2
R
a
T
L
+ B
The nature of the output characteristics is the same as that of armature control except that the
maximum torque that can now be delivered is limited to approximately (where is the kI
e
I
a
I
a
rated armature current and is the control current in the excitation). As the field is weakened I
e
more, the peak torque capability proportionately reduces, the speed proportionately increases,
Va1
Va2
Va3
N1
N2
N3
Va1 > Va2 > Va3
Constant Torque Envelope
kIeIa
T
N
Fig. 2 Speed Control through Armature Voltage Control
and the output power remains constant. The torque speed curve in the field controlled region is
shown in Fig. 3.
The armature control region is referred to as the constant torque operation and the field
controlled region is referred to as the constant horse power operation.
Dynamic model of the separately excited dc motor
Under constant excitation the machine model is linear and the defining equations are

di
a
dt
d
dt
1
]
1
1
1

a

kI
e
L
a
kI
e
J

B
J
1
]
1
1
1

i
a

1
]
1
+

1
L
a
0
1
]
1
1
1
v
a
+

1
J
1
]
1
1
1
T
L
The characteristic equation of the system is
s
2
+ s

a
+
1

m
_
,
+

a

m
+
k
2
I
e
2
L
a
J
_
,

0
= natural frequency
o

a

m
+
k
2
I
e
2
L
a
J
_
,

= damping

a
+
1

m
_
,
1
2
o
We may qualitatively see that the motor response is a second order response. It could be under
or overdamped. The response is faster for strong field excitation. The damping is low at strong
field excitation.
Control strategies for separately excited dc motor
The range of armature control is below base speed and that of field control is above
base speed. Below base speed, the excitation is kept constant at , and the armature voltage I
e
is varied in the range of . When the rated armature voltage is reached, a further v
a
tV
a
speed increase is possible by lowering the field current below . This strategy of control is I
e
shown in Fig. 4. Usually during acceleration and deceleration (braking) the motor will be oper-
ated in torque control mode and during running under speed control. The reversal of speed is
possible either by reversing the armature voltage or by reversing the field current. However the
second method of speed reversal (by field reversal) is seldom used, because it requires the
de-excitation and re-excitation of the field. This process is slow on account of the large field
E6 201 Power Electronics
EE 187 IISc
Fig. 3 Speed Control through Field Current Control
Ie1, Va1
Va2
Va3
N1
N2
N3
Va1 > Va2 > Va3
kIeIa
T
N
Ie3
Ie2, Va1
Constant Power
Envelope
circuit time constant. Further the armature current will have to be blocked during the time of
field reversal causing discontinuity in the torque generated. For these reasons, four quadrant
drives are invariably equipped with an armature voltage source of bipolar voltage and current
capability. The advantage is that the field excitation is unchanged and the response could be
faster.
When several motors are fed from a common armature voltage source, as in many con-
tinuous manufacturing industries, trim adjustment of the field is employed to closely match the
speeds of several drives operating sequentially.
SERIES EXCITED DC MOTOR
Figure 5 shows the schematic diagram of a series excited dc motor. On account of their
operating characteristics the series motors are well suited for traction application. The excita-
tion field and the armature field are both determined by the armature current. The de-
e

a
fining equations of the motor are
T
N
Field Control
Field Control
Speed Control
Torque Control
A
r
m
a
t
u
r
e

C
o
n
t
r
o
l
DRIVE BRAKE
DRIVE BRAKE
Fig. 4 Overall Control Strategy for DC Motor Control
Fig. 5 Series Excited DC Motor
Tg
T
L
Ra
La
Va
ia
Re
Le
J, B

e
a
armature circuit equation v
a
Ri
a
+L
a
di
a
dt
+ e
a
; R R
e
+ R
a
L L
e
+ L
a
mechanical system equation T
g
T
L
+ B + J
d
dt
back emf equation e
a
ki
a

torque equation T
g
ki
a
2
The system equations may be put in the following form.

a
d i
a
dt

v
a
R
i
a

e
a
R

m
d w
dt

T
L
B
+
T
g
B
and are defined as the electrical and the mechanical time constants of the system. The
a

m
independent inputs to the system are and . The output states of the system are and . v
a
T
L
i
a

The armature circuit equation and the mechanical system equations are seen to be coupled to
each other through the back emf and torque relationships. For any set of inputs , and , the v
a
T
L
above set of differential equations may be integrated to obtain the dynamic performance of the
machine.
Steady state relationship
Under steady state,
V
a
RI
a
+ kI
a

T
g
T
L
+ B
Speed control
Usually the machine is controlled by controlling the armature voltage (Field control
may be incorporated if field diverter resistors are used). The mechanical and the electrical sys-
tem equations may then be combined to get the following steady state relationship.
k

V
a
R+k
_
,
2
T
L
+ B
The torque speed characteristics of the motor is shown in Fig. 6. The speed torque curve is
steep at light loads. This is a desirable feature for traction applications as well as multi motor
drives. Though field weakening is possible, the range of control in restricted. The effect of the
system parameters on the speed torque curve is shown qualitatively in Fig. 7.
E6 201 Power Electronics
EE 189 IISc
N
T
Va1
Va2
Va3
Va1>Va2>Va3
Fig. 6 Fig. 7
N
T
Increasing R
Increasing Va
Increasing e
Increasing
B
Dynamic model of the series excited dc motor
The series machine model is nonlinear. This may be put in the form of

di
a
d t
v
a
Ri
a
ki
a

d
d t

T
L
J

B
J
+ ki
a
2
The system being nonlinear, the solution may be found through numerical integration. Or a
linear small signal model may be found close to any desired operating point.
Control strategy of a series excited dc motor
By controlling the armature voltage, the motor may be driven at any desired speed
(provided the load is not zero). In order to reverse the torque, a reversal of field winding is
necessary. The machine can thus be driven in the forward or reverse direction by reversing the
field. However, operation in the second and fourth quadrant will have to be through a different
control strategy. This may be appreciated by seeing the electrical equation of the machine. At
, the effective resistance of the machine becomes zero. In other words, the back R/k
i
a
e, Ri
a
Field Resistance Line
Excitation Characteristics
i
a
R L

Fig. 8
emf counters the resistive drop in the armature and the machine may operate without input
voltage (self excitation). The self excited braking operation is shown in Fig. 8. For closer v
a
control usually the series machine is reconnected as a separately excited machine for third and
fourth quadrant operation.
SEPARATELY EXCITED DC MOTOR
We can put together the power converter circuit and the load (in this case a separately
excited dc motor) and study the system.
Single quadrant chopper
Figure 1 shows the separately excited dc motor driven through a dc to dc converter
(chopper). The switch may be realised out of a number of possible devices (BJT and a diode,
SCR and a diode, or MOSFET and a diode). The switch is operated as shown in Fig. 2. The
Fig. 1 Chopper Driven Separately Excited DC Motor
Tg
T
L
Ra
La
e Va
ia
ie
Re
Le
Ve
J, B

Ton
Tof f
Vg
switching period is ( ). During the motor draws energy from the source. T
s
T
on
+ T
off
T
on
During armature current freewheels and no energy is drawn from the source. The voltage T
off
and the current waveforms are shown in Fig. 3. The system equations are
During on time:
v
g
R
a
i
a
+ L
a
di
a
dt
+ k
ki
a
B + J
d
dt
+ T
L
During off time:
0 R
a
i
a
+ L
a
di
a
dt
+ k
ki
a
B + J
d
dt
+ T
L
The system equations may be averaged as
v
g
d R
a
i
a
+ L
a
di
a
dt
+ k
k i
a
B + J
d
d t
+ T
L
= duty ratio d
T
on
T
off
The machine's speed or torque may be controlled by control of the duty ratio . The d
system equations may be written in the following state space form.

d i
a
d t
d
d t
1
]
1
1
1

R
a
L
a

k
L
a
k
J

B
J
1
]
1
1
1

i
a

1
]
1
+

v
g
d
L
a

T
L
J
1
]
1
1
1
1
1
The inputs are: Control input d
E6 201 Power Electronics
EE 191 IISc
Fig. 2 Switch States
Ton Tof f
t
Ton Tof f
t
Vg
t
i
a
i
g
t
t
Fig. 3 Switch States and Voltage and Currents in the Chopper
Power input v
g
Load input T
L
The system states are: Armature current i
a
Rotor speed
The control objectives are: Speed control or Torque control through control of duty ratio d
Desired transfer functions: or for Torque or Current control
T
g
d
i
a
d
for speed control

d
The block diagram of the motor is shown in Fig. 4. The small signal system representation is

d
^
i
a
dt
d
^

dt
1
]
1
1
1
1
1

R
a
L
a

k
L
a
k
J

B
J
1
]
1
1
1

^
i
a
^

1
]
1
1
1
+

V
g
L
a
0
1
]
1
1
1
^
d +

D
L
a
0
1
]
1
1
1
^
v
g
+

1
J
1
]
1
1
1
^
t
L
From this the different transfer functions may be evaluated for each of the inputs while the other
inputs are made zero.
= torque control transfer function ; = load disturbance transfer function
^
i
a
^
d
^

^
t
L
= speed control transfer function ;
^

^
d
= power disturbance transfer function ;

^
v
g
Discontinuous conduction
The above converter when realised for single quadrant operation with an IGBT switch
and a diode is as shown in Fig. 5. This is a single quadrant chopper since the output current and
voltage are both restricted to be positive. The power flow can be only in one direction - from
source to load. When such a chopper is used to drive a separately excited dc motor, and for
conditions when the motor current becomes negative, the chopper cannot absorb this negative
Ra
1/La K
K
1/J
B

L
d
Vg
Fig. 4 Block Diagram of Chopper Driven Separately Excited DC Motor
current. Then the load current becomes discontinuous. Such a mode of operation is called the
discontinuous current mode and is shown in Fig. 6. The system defining equations are
During on time ( ): d T
s
v
g
R
a
i
a
+ L
a
d i
a
d t
+ k
k i
a
B + J
d
d t
+ T
L
During off time ( ): d
2
T
s
0 R
a
i
a
+ L
a
d i
a
d t
+ k
k i
a
B + J
d
d t
+ T
L
During idle time [ ]: (1 d d
2
)T
s
0 R
a
i
a
+ L
a
d i
a
d t
+ k
0 B + J
d
d t
+ T
L
The system equations may be averaged as
v
g
d R
a
(d + d
2
) i
a
+ L
a
d i
a
d t
+ k (d + d
2
)
k (d + d
2
) i
a
B + J
d
d t
+ T
L
= duty ratio d
T
on
T
s
E6 201 Power Electronics
EE 193 IISc
Vg
Ra
La i
a
D
S
Fig. 5 Separately Excited DC Motor Driven by a Single Quadrant Chopper
i
a
t
Ts
e
a
t
dTs d2Ts
Vg
Fig. 6 Armature Current and Armature Voltage in Discontinuous Conduction
Notice that is a dependent variable. The machines speed or torque may be controlled by d
2
control of the duty ratio d. The system equations may be written in the following state space
form.

d
^
i
a
d t
d
^

d t
1
]
1
1
1
1
1

R
a
(d+d
2
)
L
a

k(d+d
2
)
L
a
k(d+d
2
)
J

B
J
1
]
1
1
1
1
1

^
i
a
^

1
]
1
1
1
+

v
g
d
L
a

T
L
J
1
]
1
1
1
1
1
The inputs are: Control input d
Power input v
g
Load input T
L
The system states are: Armature current i
a
Rotor speed
The control objectives are: Speed control or Torque control through control of duty ratio
Desired transfer functions: or for Torque or Current control
T
g
d
i
a
d
for Speed control

d
Steady state solution:
If the motor resistance is neglected, then the steady state equations are R
a
D V
g
(D + D
2
) k
Define ; Then ; M
k
V
g
M
D
D + D
2
D
2

D (1M)
M
From the discontinuous current waveform shown in Fig. 6,
;
k I
p
(D + D
2
)
2
B + T
L
I
p

(V
g
k ) D T
s
L
a

V
g
(1M) D T
s
L
a
;where (D + D
2
)
2 L
a
B
k
2
T
s

1
D (1M)
_
,
[
M + M

] M

k T
L
B V
g
Define K
2 L
a
B
k
2
T
s
;
D
M

K
D (1M)
[
M+ M

]
D
K M (M+M

)
(1M)
; ; D
2

1
M
1
MV
g
K
I
av

(B + T
L
)
k
M
K M

+
(
K M

)
2
+ 4 K D
2
2 K
At the border of discontinuous conduction will be satisfied by both the equations M
; D
K
crit
M (M+M

)
(1M)
M K
crit

D (1D)
(D + M

)
From this the criterion for operation on the border of continuous and discontinuous conduction
is found as shown in Fig. 7.
operation in continuous conduction K
2 L
a
B
k
2
T
s
K
crit
operation in discontinuous conduction K
2 L
a
B
k
2
T
s
K
crit
Dynamic model in DCM
The motor equations (when is negligible) are R
a

d i
a
d t

k d
L
a

k d
2

L
a
+
d v
g
L
a

d
d t

k d i
a
J
+
k d
2
i
a
J

t
L
J

B
J
Steady State
; ;
D V
g
k (D + D
2
)
D
2

D V
g
(1M)
k
I
B + T
L
k(D + D
2
)
Dynamic equations (small signal)

d
^
i
a
d t
d
^

d t
1
]
1
1
1
1
1

D+D
2
_
,
k
L
a

D+D
2
_
,
k
J

B
J
1
]
1
1
1
1
1
1
1

^
i
a
^

1
]
1
1
1
+

V
g
k
L
a
kI
a
J
1
]
1
1
1
1
1
^
d
+

k
L
a
k I
a
J
1
]
1
1
1
^
d
2
+

1
J
1
]
1
1
1
^
t
L
Relationship between and d d
2
E6 201 Power Electronics
EE 195 IISc
0 1 D
Kcrit
K
CCM
DCM
Fig. 7 Continuous and Discontinuous Conduction Operation
V
g
D
k
D D
2
^
d
2

V
g
k
_
,
k
^
d +
D
k
^
v
g

D V
g
k
2
^

substituting for in the dynamic equation, we get

^
d
2
d
^
i
a
d t
0

d
^

d t

(D+D
2
) k
J
^
i
a
+
k I
a
J
(
^
d +
^
d
2
)
B
J
^

1
J
^
t
L
Relationship between and and
^
i
a
^
d
^
d
2
I
a

V
g
k
_
,
D T
s
2 L
a

^
i
a

T
s
2 L
a

'

V
g
k
_
,
^
d + D
^
v
g
k D
^

D + D
2
_
,

D V
g
k

^
d
2
+
^
d
D
k
^
v
g
+
V
g
k
^
d
D V
g
k
2
^

on substitution for all the dependent variables,

d
^

d t

^

'

B
J

D
2
B
K M
2
J

;
+
^
d

'
2 B V
g
(1M)
k MK J

;
+
^
v
g

'
D
2
B (2M)
k K J M

;
Notice that the system order has reduced in DCM.
Two quadrant chopper
The chopper may be made into a two quadrant chopper by simply making the switches
in the chopper bidirectional. Such a two quadrant chopper is shown in Fig. 8. The current and
voltage waveform when power is flowing from the motor to the source is shown in Fig. 9. The
two quadrant chopper is capable of driving and braking the motor in one direction.
To Motor
Vg
Fig. 8 Bidirectional Chopper
Va
Ton Tof f
Vg
i
a
t
Fig. 9 Chopper Voltage and Current During Braking Operation
Four quadrant chopper
A four quadrant chopper is capable of driving and braking the motor in both the direc-
tions. A four quadrant chopper made up of IGBTs is shown in Fig. 10. The power circuit does
not show the snubber elements. It may be seen that the four quadrant chopper consists of two
numbers of two quadrant choppers with the load connected between the two choppers in differ-
ential fashion. The switch on/off timing signals are shown in Fig. 11.
SERIES EXCITED DC MOTOR
We can develop the chopper driven dc series motor model as follows.
E6 201 Power Electronics
EE 197 IISc
Fig. 10 Four Quadrant Chopper
Vg
To Motor
P
P
N
N
t
Va < 0
t
Va = 0
t
Va > 0
P
N
P
N
P
N
Fig. 11 Switch Enable Signals for Four Quadrant Chopper
Fig. 1 Chopper Driven Series Excited Motor
Vg
La i
a Tof f
Ton
Le
Va
ig
J, B

Tg
T
L
Single quadrant chopper
Figure 1 shows the series excited dc motor driven through a dc to dc converter
(chopper). The switch is realised using SCRs. In practice the dc series motors are used for
traction application where the voltage, current, and power are high (100s of kW). The devices
used are usually GTOs or SCRs. The chopper operation is as before. The defining equations
are
During on time
V
g
R
a
i
a
+ L
d i
a
d t
+ k i
a

k i
a
2
B + J
d
d t
+ T
L
During off time
0 R
a
i
a
+ L
d i
a
d t
+ k i
a

k i
a
2
B + J
d
d t
+ T
L
The system equations may be averaged as
V
g
d R
a
i
a
+ L
d i
a
d t
+ k i
a

k i
a
2
B + J
d
d t
+ T
L
= duty ratio d
T
on
T
s
The machines speed or torque may be controlled by control of the duty ratio . d
The inputs are: Control input Power input d V
g
Load input T
L
The system states are: Current Speed i
a

The control objectives are: Speed Control or Torque Control through duty ratio d
Desired transfer functions: or for torque control
T
g
d
i
a
d
for speed control

d
Fig. 2 Block Diagram of Chopper Driven Series Excited DC Motor

Ra
1/La
K
K 1/J
B

L
d
Vg
The system equations are nonlinear. The block diagram representation of the system equations
is shown in Fig. 2. The dynamic solution can be obtained only by numerical simulation. The
small signal and steady state model may be readily obtained.
Dynamic model
The small signal dynamic model of the chopper driven series motor may be obtained to be as
follows.

d
^
i
a
d t
d
^

d t
1
]
1
1
1
1
1

R+k
L

kI
a
L
2kI
a
J

B
J
1
]
1
1
1

^
i
a
^

1
]
1
1
1
+

V
g
L
0
1
]
1
1
1
^
d +

D
L
0
1
]
1
1
1
^
v
g
+

1
J
1
]
1
1
1
^
t
L
From this the different transfer functions may be evaluated for each of the inputs while the other
inputs are being zero.
= torque control transfer function ; = power disturbance function
^
i
a
^
d
^

^
v
g
= speed control transfer function ; = load disturbance transfer function
^

^
d
^

^
t
L
The dc series motor unlike the separately excited motor does not operate in the discontinuous
conduction mode.
Ramp
V control
Ton (min)
Ton (max)
Ton
M
A
Fig. 4 forced Commutated Chopper Controller
Steady state solution:
If the motor resistance is neglected, then the steady state equations are R
a
D V
g
k I
a
E6 201 Power Electronics
EE 199 IISc
D
Le La
M
A
C
Vg
L
Fig. 3 A Series Motor Driven by a Force Commutated SCR Chopper
k I
a
2
B + T
L
A single quadrant chopper controlling a dc series motor along with the commutation circuit is
shown in Fig. 3. A possible control scheme for the chopper devices is shown in Fig. 4.
Two quadrant chopper
The motor when operated in the second quadrant the torque has to be negative. This
may be obtained by the dynamic braking scheme shown in Fig. 5. During braking the chopper is
off and the armature and the field are shorted through the braking resistor . Notice that the R
b
field current direction is maintained to be the same as in running. The mechanical energy stored
in the moving system is transferred to the electrical circuit and dissipated in the resistor . In R
b
drives where the braking is more frequent (as in suburban trains), one may not like to dissipate
Le
Run
Brake
Vg
Fig. 5 Dynamic Braking of Series Excited Motor
Fig. 6 Regenerative Braking Chopper for DC Series Excited Motor
Vg
START
ACCELERATE
RUN
Fig. 7 Series & Parallel Connection of DC Series Motors in Traction Application
the stored energy during braking. It will be economical to recover the energy in the moving sys-
tem and feed it to the source. Such a braking operation is called regenerative braking. A
scheme to achieve the same is shown in Fig. 6.
Four quadrant chopper
Usually it is not required in series motor drives to obtain fast multiquadrant transitions.
Therefore usually four quadrant operation is achieved by relay change over circuits.
Multiphase chopper
Traction application invariably uses multiple motor drives. In such cases several lower
rated choppers may be used in place of a single higher rated chopper. The individual choppers
may be operated with a phase shift from each other with obvious advantages. The arrangements
are shown in Figs 7 and 8.
CLOSED LOOP CONTROL OF MOTOR DRIVES
Control Equations of the DC motor
In this section we see the closed loop control of the separately excited dc motor. The
state space model of the separately excited motor driven from an armature supply is given as
E6 201 Power Electronics
EE 201 IISc
ACCELERATE
RUN
START
Fig. 8 Multiphase Chopper Applications

di
a
dt
d
dt
1
]
1
1
1

R
a
L
a

k
L
a
k
J

B
J
1
]
1
1
1

i
a

1
]
1
+

G
L
a
0
1
]
1
1
1
v
c
+

1
J
1
]
1
1
1
t
L
where is the gain of the power converter used for the armature power source. G

X A X + b
^
v
c
+ h
^
t
L
c X [0 1] X
The control of the motor is done through the control of the control input . The control transfer v
c
function is
G(s)
^
(s)
^
v
c
(s)
c [sI A]
1
b
; ;
k G

BR
a
(1+s
e
)(1+s
m
)+k
2
1
]

e

L
a
R
a

m

J
B
The transfer function is of order 2
When the motor is lossless ( ), the system poles are on the imaginary B 0 R
a
axis.
When the losses are very high ( ), the system poles are real and the BR
a
>> k
2
same as the electrical and mechanical frequencies of the motor.
When the electrical time constant is very much less than the mechanical time
constant of the motor ( ), then the system is effectively a first order
e
<<
m
system with the single pole at equal to .
m1

m
BR
a
k
2
+BR
a
In general if the loss is low and the time constants and are not well
e

m
separated, then the motor is an underdamped second order system with the natural
frequency at .
k
2
+BR
a
L
a
J
Fig. 1
Lossless Motor
Fig. 2
Highloss Motor
Fig. 3
Fig. 4
Underdamped Motor
k/ L
a
J
G
log
log
log
log
G
G
G

m1
o

e
>>
m
These different cases are shown in Figs 1 to 4. Practical machines will have a response as
given in Figs 3 or 4. For those in Fig. 3, a proportional controller may be used. For those in
Fig. 4, a lead-lag controller may be used
Proportional Controller
Consider case shown in Fig. 3 above. The motor transfer function is
G(s)
K

1 + s
m1
_
,
;
m1

m
BR
a
BR
a
+k
2
We may introduce a proportional controller with a gain as shown in Fig. 5. The H
c
K
c
closed loop transfer function is
; G
cl
(s)
KK
c

1 + s
m1
_
,
1 +
KK
c

1 + s
m1
_
,

KK
c
1 + s
m1
+ KK
c

1
1+s


m1
KK
c
The closed loop speed response for step change in speed will be

1 e

_
,

The closed loop current transfer function is

G
i
(s)
BKK
c
(1 + s
m
)
k

1 + s
m1
_
,
1 +
KK
c

1 + s
m1
_
,

B
k
(1 + s
m
)
1 + s
The closed loop response in current to step change in speed reference is given by
i
a
(t)

B
k

1 e

_
,

B
k

_
,
e

The first part eventually leads to the steady state current equal to . The sec- I
a
B

/k
ond part is a transient current and can be several times the steady state current (on account of
the multiplier ). In typical systems will be more than 10 and so this type of con-
m
/
m
/
troller will result in large current transients which stress the motor windings enormously. So as
a rule this kind of speed controllers are not employed in motor control. Instead the motor
E6 201 Power Electronics
EE 203 IISc
Kc k
T
L

Fig. 5 Block Diagram of a DC Drive with Proportional Controller
1
B
1
1+s
m
KB
k
1+s
m
1 +s
m1
current is controlled in closed loop and the speed control loop is realised as an outer loop.
Such a controller is as shown in Fig. 6.
CURRENT
All practical drives incorporate an inner current controller ( ) and an outer speed H
i
controller ( ) as shown in Fig. 6. Consider the current loop as shown in Fig. 7. H
s
G
i
(s)
^
i
a
(s)
^
v
c
(s)

(1 + s
m
)
BG
JL
a

1 +
s
Q
p
+
s
2

p
2
_
,

BR
a
+ k
2
JL
a

p
2

BR
a
+ k
2
JL
a
_
,

; Q

p

R
a
L
a
+
B
J
_
,
The transfer function is shown in Fig. 8. The phase of the transfer function is always inside
90 and hence the design of a closed loop compensator is simple. Either a proportional or a
proportional plus integral (PI) current compensator may be used. Consider a PI compensator of
the form
H
i
(s)
1+
s

p
s

p
The loop gain is shown in Fig. 8. The closed loop bandwidth may be seen to be higher G
i
H
i
than and and so the closed loop gain in the range of will be practically unity.
p

m

p
This is shown in Fig. 7b.
SPEED CONTROLLER
Fig. 6 Closed Loop Controller with Nested Control Loops

H
s
H
i
G
i
(s) k
T
L
1
B
1
1+s
m
I *
I
1
Fig. 7 Closed Loop Current Controller and Its Low Frequency Model
H
i
G
i
(s)
I

I
Fig. 8 Motor Current Transfer Function
G
log

m

p
With the current controller , the current gain may be taken as 1, and the system is H
i
(s)
simplified as shown in Fig. 10.
The speed transfer function is now first order

(s)
I

(s)

k
B (1 + s
m
)
and is shown in Fig. 11. The design of closed loop compensator for speed is quite simple.
Select
H
s
(s) K
s
1 +
s

m
s

m
The loop gain is shown in Fig. 12. The bandwidth of the overall system is .
m
kK
s
/B
E6 201 Power Electronics
EE 205 IISc
Fig. 9 Loopgain Transfer Function
G
i
H
i
log

m

p
Fig. 10 Block Diagram of the Current Controlled DC Motor
1
Current Controller
i i
*

H
s
k
T
L
1
B
1
1 +s
m
Fig. 11 Current Speed Transfer Function
G
log

m
Fig.12 Loopgain of the Speed Controlled Motor Drive
G

(s)H
s
(s)
log
m

m
kK
s
/B
All practical motor drives (separately excited dc motor, series dc motor, ac motors) in-
corporate the above type of speed control. The speed controller ( ) output may be clamped H
s
to and . Then during speed transients, the speed loop will not be effective and +I
max
I
max
the motor will be controlled to draw constant current and deliver constant torque correspon-
ding to . Thus the motor current will be within safe limits during transients also. tI
max
FEED FORWARD CONTROLLERS
A Simple Example
Consider the following circuit shown in Fig. 1. The circuit consists of an RL
circuit terminated to a current load. The control objective is to maintain the current through the
inductor constant.
The circuit equation is
L
d i
dt
u (i I
L
)R
i
u+I
L
R
(sL+R )
The above system may be represented by the following block diagram shown in Fig. 2.
Following the classical method we may design a PI controller of the form
; H
i
(s) K
1+sT
a
sT
a
T
a

L
R
The controller loop gain is
T K
1+sT
a
T
a
1
R
1
1+sT
a

K
s L
Such a design will lead to a steady state error of zero and a control bandwidth of . Such a
K
L
controller is shown in Fig. 3. The response in current for step change in command is given by
; i I

1 e

_
,

L
K
u
i L
R
I
L
Fig. 1 A Simple RL Circuit
u
RI
L
i
Fig.2 Block Diagram of the Circuit in Fig. 1
1
R+s L
Fig. 3 Block Diagram with Classical PI Controller
I

i
K(1+sT
a
)
sT
a
1
R+s L
i
u
RI
L
For faster response, the gain must be made large. The performance in the response of K
inductor current will be quite satisfactory for changes in the command input. However, if there
are disturbances in the load current , the performance is not quite satisfactory. It can be I
L
shown that for step changes in the load current the response in the inductor current is given I
L

by
i I
L

T
a

_
,

t
T
a
_
,

This response consists of a slow component , and is not satisfactory. Further

t
T
a
_
,

if either the resistance or the inductor is a function of the current through them then the sys- R L
tem equations become non-linear. In such a case a design based on the classical approach in-
troduces further limitations by way of small signal validity. The following explains the strategy
of feedforward control. The system equation is given by
L
d i
d t
u v
If the desired response in current is required to be first order with a time constant of , the T
i
same may be expressed mathematically as
T
i
d i
d t
+ i I

When the above two equations are combined, we get

u
L
T
i
(I

i) + v
If the input is made up according to the above equation then the resulting response in u
current will be first order with a time constant of . Such a controller is shown in Fig. 4. The T
i
controller is a simple proportional control (of gain ) followed by the additional feed for- L/T
i
ward term ( ). v
Advantages of feed forward control:
The response is same for command tracking as well as disturbance rejection.
The method works for non-linear systems also.
Disadvantages of the feed forward control:
It requires extra signals to be sensed and fed forward (in this case ). v
A knowledge of system parameter is needed ( in this example). L
Application of Feed Forward to Speed Control of DC Motor
The feed forward control is applicable to the speed controller. Consider the dc motor
for which a current controller has been realised. When the current bandwidth is large, the
E6 201 Power Electronics
EE 207 IISc
Fig. 4 Feedforward Current Control in a Simple RL Circuit
i L
R
I
L
I
*
i
v
L
T
i
current controller may be assumed to be a simple gain of magnitude , for frequencies less G
i
than the bandwidth of the current controller. The motor may be represented by
I

G
i
K
T
m
d
J
d
d t
+ B + m
L
The mechanical system is represented in Fig. 5.
Let the desired response be with a time constant of . This may be expressed mathematically T
w
as
T
w
d
d t
+

We may combine this to the system mechanical equation to obtain the following relationship on
the current reference . I

1
G
i
K
T

m
L
+ B +
J
T
w
[

]
_
,
The above controller is shown in block diagram in Fig. 6. It consists of a proportional current
controller and the feed forward terms corresponding to the frictional torque term and the B
load torque term . m
L
In realising the above control apart from the speed signal for feedback, load torque m
L
also is required to be sensed. Further the quantities and are all also to be known. J, B, K
T
G
i
Error in speed error gain will result in a difference in the response time, but will not cause
steady state error in speed. Error in feed forward terms will result in an error in the steady
state response without causing any error in response time. This steady state error may be cor-
rected by an integral controller operating across the proportional controller of speed error op-
erating very slowly.
The following schematic in Fig. 7 shows a feed forward controller for the cur-
rent control in dc motor.
B
I
*
m
d
m
L

Fig. 5 Mechanica System of a Motor Drive

1
J

K
T
G
i

f b

f b m
L
I
*
Fig. 6 Feedforward Speed Controller
1
BT
W
B
1
G
i
K
T
R
I
*
e
iR
i
v
a
Fig. 7 Feedforward Current Controller
L
a
T
a
SPEED CONTROL OF DC MOTOR
(Symmetrical Optimum Method)
In Section 1 it was seen that the speed controller may be realised through a PI
controller around the current controlled dc motor. In such a case it was mentioned that the
speed amplifier PI time constant is chosen to be the same as the mechanical time constant of the
motor. Even though this is a satisfactory method for small machines, it will result in inconven-
iently large circuit elements for the speed amplifier in the case of large motors which have
large mechanical time constant. In such applications a method known as the symmetrical opti-
mum method has become popular.
Machine Model
The model of the current controlled dc machine is shown in Fig. 1
Closed Loop Controller
With such a system we may employ a PI speed controller. The closed loop control of
speed with such a system is shown in Fig. 2.
The various parameters of the system are
: Current Controller Gain. K
i
: Torque Constant of the Motor. K
T
: Gain of the Mechanical system . K
m

K
m

1
J
_
,
: Speed Measurement Gain. K
s
: Response Time Constant of the Current Controller (typically 2 to 5mS). T
i
: Filter Time Constant of Speed Measurement (typically about 1 to 5mS). T
s
The design consists of evaluating the proportional time constant and the proportional gain T
p
, for a given phase margin of design specification and evaluate the corresponding K
p

m
bandwidth of the design.
E6 201 Power Electronics
EE 209 IISc
I
*
m
L

fb
Fig. 1 Current Controlled DC Machine
K
i
K
T
1 +s T
i
1
J s
K
s
1 +s T
i
Fig. 2 Current Controlled DC Machine
I
*
m
L

fb

K
i
K
T
1 +s T
i
K
m
J s
K
s
1 +s T
i
K
p
(1+sT
p
)
s T
p
Design Method
Figure 3 shows the simplified equivalent block diagram of the system. The cur-
rent controller time constant and the speed measurement blocks have been combined and repre-
sented by their low frequency equivalent.
(T
i
+ T
s
+ T
c
)
: Average delay in the converter (1.67 mS for a 6 pulse converter). T
c
db
1/Ti

1/
log

degree
log

m
Fig. 4 Asymptotic Bode Plot of Loopgain
The asymptotic bode plot of the overall loop gain is shown in Fig. 4. The proportional time
constant is chosen to be a multiple of the equivalent natural time constant of the system. T
p

T
p
a
2

In such a case in order to obtain the best phase margin the proportional gain has to be K
p
chosen such that the gain cross over frequency is the geometric mean of the two characteristic
frequencies of the system. Phase margin is then given by

1/T
p
and 1/
_
,

c

1
T
p


1
a

a
T
p

m
180 +

180 + tan
1

c
T
p
_
,
tan
1
(
c
)
_
,

m
tan
1
(a) tan
1

1
a
_
,
The above expression may be simplified to and put in the form of a table for
convenience.
Fig. 3 Current Controlled DC Machine
I
*

fb

K
i
K
T
K
s
1+s
K
m
J s
K
p
(1+sT
p
)
s T
p

m
tan
1

1
2

a
1
a
1
]
_
,
a 1.3 1.4 1.5 1.6 1.7 1.8 1.9 2 2.1 2.2 2.3 2.4 2.5

m
27.9 34.4 39.8 44.3 48 51.2 53.9 56.3 58.4 60.2 61.8 63.2 64.5
For the desired phase margin corresponding value for may be selected from the above a
Table. The proportional time constant is then selected. A suitable value of is 2. T
p
a
The magnitude of the loop gain is unity at the crossover frequency. From this the
proportional gain is evaluated. K
p
G 1
K
p
K
i
K
T
K
s
K
m
T
p
1+j
c
T
p
(
c
)
2
1+j
c

1 K
p
K
i
K
T
K
s
K
m

1+a
2 _
,

1+
1
a
2
_
,

K
p

1
K
i
K
T
K
s
K
m

1+a
2_
,

1+
1
a
2
_
,

Problem Set
1 Consider a permanent magnet dc servo motor with the following parameters.
; T
rated
10 Nm
; N
rated
3700 rpm
; k
T
0.5 Nm/A
; k
E
53 V/krpm
; R
a
0.37
;
e
4.05 mS
;
m
11.7 mS
; B 0
Calculate the terminal voltage in steady state if the motor is required to develop V
t
a torque of 5 at a speed of 1500 rpm. Nm
(N. Mohan, P 11-1, pp 306)
2 Express the transfer function in the following form. G(s)
(s)
V
t
(s)
G(s)
1
k
E
1+
2sD

n
+
s
2

n
2
E6 201 Power Electronics
EE 211 IISc
Calculate and for the servo-motor parameters given in Problem 1. Plot the D
n
magnitude and phase of by means of a Bode plot. G
1
(s)
(N. Mohan, P 11-2, pp 307)
3 With the parameters given in Problem 1, calculate and plot the change in as a
m
function of time for a step change increase of 10 V in the terminal voltage of the un
controlled unloaded motor.
(N. Mohan, P 11-3, pp 307)
4 The servo-motor in Problem 1 is driven by a full bridge dc to dc converter
operating from a 200 V dc bus. Calculate the peak-to-peak ripple in the motor
current if a PWM bipolar voltage switching scheme is used. The motor is
developing a torque of 5 at a speed of 1500 rpm. The switching frequency is 20 Nm
kHz.
(N. Mohan, P 11-4, pp 307)
5 Repeat Problem 4 if a unipolar voltage switching scheme is used in the dc to dc
converter. (figure out the difference between unipolar & bipolar switching first).
(N. Mohan, P 11-5, pp 307)
6 The speed of a separately excited dc motor is controlled by a single phase full
wave controlled rectifier. The field circuit is controlled by a full converter and the
field current is set to the maximum possible value. The ac supply voltage is 240 V,
single phase, 50 Hz. The armature resistance is 0.5 ohm, the field resistance is 345
ohm, and the emf constant is 0.71 V/A - rad/sec. Friction and no-load losses are
negligible. The armature and field currents are continuous and ripple free. The
delay angle of the armature converter is 45 and the armature current of the motor is
55 A. Determine
(a) torque developed by the motor,
(b) speed of the motor in rpm, and
(c) input power factor.
(M.H. Rashid, P 10-6, pp 336)
7 A dc chopper controls the speed of a series motor. The armature resistance is 0.04
ohm. field resistance is 0.06 ohm, emf constant is 35 mV/A - rad/sec. The dc input
to the chopper is 600 V. If it is required to maintain a constant developed torque of
547 Nm, plot the duty cycle of the chopper against the motor speed in the range of 0
to 1200 rpm.
(M.H. Rashid, P 10-12, pp 336)
8 A dc series motor is powered by a dc chopper from a 600 V dc source. The
armature resistance is 0.03 ohm, field resistance is 0.05 ohm, emf constant is 15.27
mV/A - rad/sec. The average armature current is 450 A without any appreciable
ripple. The duty ratio of the chopper is 75%. Determine
(a) input power from the source,
(b) equivalent input resistance of the chopper,
(c) motor speed in rpm, and (d) developed torque of the motor.
(M.H. Rashid, P 10-13, pp 336)
9 A dc series motor is braked with a chopper to return the power into a dc bus of 600
V. The armature resistance is 0.03 ohm, field resistance is 0.05 ohm, emf constant
is 12 mV/A - rad/sec. The average current through the motor is maintained constant
at 350 A. The armature current is continuous and ripple free. The duty ratio of the
chopper is 50%. Determine
(a) average voltage across chopper,
(b) power regenerated to supply,
(c) equivalent load resistance of the motor acting as a generator,
(d) minimum possible braking speed in rpm,
(e) maximum possible braking speed in rpm,
(f) motor speed in rpm.
(M.H. Rashid, P 10-14, pp 337)
10 A dc series motor is controlled by a two phase chopper. The average armature
current is 250 A with negligible ripple. Each of the chopper is operating with 50%
duty ratio. The individual choppers are appropriately shifted in phase. Each
chopper is operated at a frequency of 250 Hz. Sketch the input current drawn from
the source. Evaluate the average input current and the fundamental component of the
input current. At the input a simple LC filter (L = 0.35 mH, C = 5600 microfarad) is
used. Evaluate the rms fundamental component harmonic current injected into the
source.
(M.H. Rashid, P 10-18, pp 337)
11 A 40 hp, 230 V, 3500 rpm, separately excited dc motor is controlled by a linear
converter of gain 200. The moment of inertia of the load is 0.156 Nm/ rad/sec,
friction is negligible, armature resistance is 0.045 ohm, armature inductance is 730
mH, The emf constant is 0.502 V/A - rad/sec. The field current is maintained
constant at 1.25 A.
(a) Obtain the open-loop transfer functions ( ) and ( ) (s)/V
r
(s) (s)/T
L
(s)
for the motor.
(b) Calculate the steady state speed of the motor if the input voltage is V
r
set at 1 V, and the motor is loaded to 50% of its rated torque.
12 The following are the parameters of a permanent magnet dc motor.
; ; ; T
rated
12 Nm N
rated
3000 RPM K
T
0.6 Nm/A
; ; ; R
A
0.575 L
A
3 mH B 0.003 Nm/(rad/ sec)
; J 0.001 Nm/(rad/ sec
2
)
Evaluate the rated current and rated voltage of the motor.
Evaluate the electrical time constant of the motor.
Evaluate the mechanical time constant of the motor.
Rated output power of the motor.
Electrical and mechanical loss in the machine under rated operating
condition.
Efficiency of the motor under rated operating condition.
Sketch the speed ( ) Vs torque ( ) characteristics rad/ sec T
generated
Nwm
(only in the first quadrant) of the motor for the two conditions of at 200 V
A
and 100 V.
13 The above motor is driven by a chopper with an input dc voltage of 250 Volts. The
operating duty ratio of the chopper is 0.6, while the generated torque of the motor is
9 . Evaluate the speed, load torque, output power, input power and efficiency Nwm
of the motor. Assume the chopper switches to be ideal.
14 Evaluate the three different speed transfer functions for the motor, and the three
different current transfer functions for the motor. Present your results in a
normalised form (dc gain, normalised zeroes, and normalised poles).
^

^
d
^
i
A
^
d
E6 201 Power Electronics
EE 213 IISc
^

^
v
g
^
i
A
^
v
g
^

^
t
L
^
i
A
^
t
L
15 Figure 15 shows a dc series motor driven through a chopper from a voltage source
. The mathematical model of the motor is given by v
g
v
in
L
di
a
dt
+ e
a
t
g
J
d
dt
+ B + t
L
e
a
k
1
i
a

t
g
k
2
i
a
2
a. For a lossless motor show that . k
1
k
2
b. Write down the differential equations governing the system during
i) ii) T
on
T
off
c. Write down the averaged equivalent system differential equations and
state the assumptions in arriving at the same.
d. Express the system equations in the form
where is the system state vector ,

x f

x, , v
g
, d, t
L
_
,
x [i
a
]
T
d
T
on
T
on
+T
off
e. Consider small perturbations in the inputs
; ; v
g
V
g
+
^
v
g
d D +
^
d t
L
T
L
+
^
t
L
and the corresponding perturbations in the motor states
; i
a
I
a
+
^
i
a +
^

Under the above perturbations, derive the steady state and small signal linear ac models
in the form
; X g(V
g
, D, T
L
) X [I
a
]
T
Fig. 15 Chopper Driven Series Excited Motor
Vg
La i
a Tof f
Ton
Le
Va
ig
J, B

Tg
T
L

^

x A
^
x + b
^
v
g
+ f
^
d + h
^
t
L
f. The motor has a rated current of A and rated speed of rpm. The 10 750
machine constants are Volt sec/ Amp rad ; Nwm/Amp
2
. k
1
0.1 k
2
0.1
The friction coefficient is NwM sec/rad. The machine when B 0.006
driven from a chopper supplied with dc power at V is running at 150 750
rpm delivering half the rated torque to the load.
i) Evaluate the steady state current drawn by the motor.
ii) Evaluate the steady state duty ratio of the chopper.
g. From the small signal model evaluate the following transfer functions
i) ii)
^
(s)
^
t
L
(s)
^
v
g
0,
^
d 0
^
(s)
^
d (s)
^
v
g
0,
^
t
L
0
h. For mH and NwM sec
2
/rad, express the above L 10 J 0.003
transfer function in the normalised pole zero form.
16 The series motor shown in Fig. 16 is controlled by a chopper ( switching at S and
_
S
high frequency with on and and off) and is running at 750 rpm, delivering S
2
S
1
S
3
a load of 50% rated torque. The motor details are
A ; rpm ; SI unit ; I
rated
10 N
rated
750 k 0.1
; ; B 0.006 NwMSec/rad J 0.003 NwMSec
2
/rad
; ; ; T
L
50% V
g
150 V L 10 mH
a Evaluate the steady state current. Under this condition at t = 0, the chopper
switches are both turned off and dynamic braking is applied ( both S and
_
S
off and off and both on). S
2
S
1
and S
3
b Write down the dynamic equations of the system.
c Write down the initial conditions of the system dynamic equations.
d Write a program to numerically solve the system equations.
e With the help of the program, evaluate the time taken for the motor speed
to drop to 150 rpm for
i) ii) R
b
0.1 R
b
3.0
f Evaluate the peak current reached in each case.
g Evaluate the energy dissipated in the resistor in each case (till the time
N reaches 150 rpm).
E6 201 Power Electronics
EE 215 IISc
Fig. 5 Dynamic Braking of Series Excited Motor
Le
Vg
S
S
S1
S2
S3
Rb
J, B, Tg, T
L

e
a
i
a
h Establish the energy balance from start to end in either case (i.e. initial
stored energy, final stored energy, and the energy dissipated or delivered in
between).
17 Engineer Lam just graduated from Indian Institute of Science with an M.E. in
CAPSAD (how glad?). He did not want to pursue further education because he was
conned by his teacher into practising what he has studied for ages. He got a job in a
Government Defence Research Laboratory.
The place of posting is Hyderabad. He thought that the government wanted a
genius like him to be kept secure far away from all the borders of the country.
The time is 1995 when the gods have been exhibiting their thirst if not
hunger.
The first assignment for Engineer Lam was to design a position control
system. The purpose was supposed to be, to move a glass of milk and position it at
the mouth of Lord Ganesh. Engineer Lam being smart knew that the real purpose of
the position control system is more sinister.
Engineer Lam did some research (i.e. took out his class notes and copied the
system equations) and wrote down the following equations under constant
excitation.
Electrical Circuit. v
a
L
a
di
a
dt
+ R
a
i
a
+e
b
Mechanical system with no counteracting load T
g
J
d
2

d t
2
+ B
d
d t
Torque relationship T
g
K
T
i
a
Back emf relationship e
b
K
e
d
dt
He took these equations to his boss. His boss is an Ex-Service Veteran - Captain
Sham. The captain was pleased with his new recruit. From an old book which was
meant only for the eyes of the captain he read out the following parameters for
Engineer Lam to note down.
; ; ; ; R
a
2 L
a
0.01 J 1 B 5
; K
e
0.5 K
T
20
Engineer Lam was confused. His book said that and were normally equal. K
T
K
e
Finally he understood why they were not equal.
(1) Can you explain a possible reason why they are not equal.
Then Engineer Lam reasoned that the field current may be kept constant so that life
is simple. He knew that for such a linear set of relationship, Transfer function is a
handy way of presenting the dynamics of the machine.
(2) Evaluate the open loop transfer function .
(s)
v
a
(s)
What Engineer Lam next did was a stroke of sheer genius. He neglected the
armature inductance because it was very small. Unlike what you obtained in (2) L
a
the result he got was of the form
(s)
v
a
(s)


s (s +)
(3) Evaluate and .
Engineer Lam was happy that everything was moving great. He looked at the
openloop transfer function and realised that it will be impossible to control the
position with openloop control.
(4) Do you agree with Engineer Lam?
Engineer Lam realised the need for a closed loop controller for position control. A
little more research (!) indicated that a simple proportional compensator will be
sufficient to get zero steady state error in position.
(5) Can you substantiate the above claim?
Engineer Lam sketched a simple proportional controller with a gain of as shown K
in Fig. 1. He remembered that for second order systems where proportional
controller is used, root locus is a good technique to select with a knowledge of K
openloop poles and the desired closed loop poles.
(6) Sketch the root locus for the closed loop poles as a function of K
varying from 0 to .
(7) Select such that the closed loop response is a second order response K
with a damping factor of 0.7. Evaluate the natural frequency of response
under this condition.
Engineer Lam was by now ecstatic. He decided to impress the captain by throwing
in a simulation also. He had this nice package TUTSIM available (pirated from IISc
lab).
(8) Write a simple program simulating the motor, and the closed loop
controller (the simplified model obtained after the stroke of genius from Er.
Lam).
(9) Verify from the simulation the closed loop response for unit step input in
, the natural frequency of response, and the overshoot corresponding Theta

to the designed damping factor of 0.7. The official Nos. may be obtained
from any control system book.
Er. Lam completed all this and showed his results. Captain was thoroughly pleased.
He asked Er. Lam to build the system.
18 Engineer Lam visited the place where the position control system was going to be
installed (remember the Ganesh temple where the system was supposed to be
installed ?). He checked with the user that the position commands will always be
within 2 units ( is limited to be within 2 units). Er. Lam ran his t Theta

t
position control simulation program to find out the maximum voltage to be applied
to the motor and the maximum current drawn by the motor.
From your program evaluate the maximum current and maximum voltage to
be supplied to the motor if the position command input is restricted as
above.
Er. Lam then selected a power amplifier accordingly, which is capable of meeting
the above requirements. He wanted the position feedback signal to the controller to
be in the range of . He selected a suitable potentiometer for the purpose as t10V
shown in Fig. 1.
What is the value of the feedback gain ? What will be the value of the
amplifier gain so that the response is the same as designed in the previous K
problem?
Er. Lam got hold of an amplifier to serve as the Block K. He checked its working
independently that it gives the required gain, and capable of delivering the required
E6 201 Power Electronics
EE 217 IISc
K Motor

v
a
Fig. 1 DC Motor with Proportional Controller
current. The amplifier also has a knob (remember this knob) at the front which can
be adjusted to get the required gain? He connected the motor to the amplifier, and
the position feedback potentiometer to the motor shaft. He built a small summing
circuit as shown below in Fig. 2. He used a linear power supply giving ( ) to t15V
provide the power required for the summing amplifier. Next he built a small step
signal injection circuit as shown below in Fig. 3.
Indicate how the above may be used to inject a unit step input to position
command.
Er. Lam tested his design by injecting a step input of 2 V to the position command.
Just to be sure he also tested his design with the actual gain one half of the designed
value. He ran his simulation program again to check the peak current and voltage
requirements from the amplifier if the gain were doubled. Then ensuring that his
amplifier can supply the required voltage and current, he tested the system with
double the designed gain.
From your simulation program give the response for the above two cases.
What are the peak currents, voltages and settling time in each of these
cases?
Being satisfied with the design, Er. Lam arranged for a demonstration for the benefit
of his boss. The demonstration went off perfectly. Er. Lam showed the designed
response, response with half the gain, and with double the gain. The captain was
quite impressed. He looked at everything. He wanted to throw the switches S1 and
S2 in different sequences and see for himself the position control system. Then his

10k
10k
10k
To Amplifier +15V
-15V
-
+
Fig. 2
K Motor

v
a
Fi g. 1
10V

+ 10V
-
+
To

S1
S2
5k
10k
+15V
-15V
+15V
-15V
Fig. 3
eyes landed on the gain setting knob of the amplifier. He wanted to know its role in
the overall plan. Er. Lam said that by simply cranking up the gain, it is possible to
make the system respond faster. The captain gleefully cranked up the gain.
It all started with a small whimper from the amplifier. The whimper grew
into a growl. Then someone noticed an imperceptible shiver in the motor shaft. The
shivering grew, the shaft started shaking violently, fumes started to come out of the
motor, and mercifully the fuse in the amplifier blew before more damage was done.
Explain what happened. Take into account the neglected inductance of the
motor, and evaluate the gain threshold which the Captain exceeded.
Add the effect of the motor inductance in your simulation program and
create an 'action replay' of what happened in the system.
E6 201 Power Electronics
EE 219 IISc