www.covenantuniversity.

Raising a new Generation of Leaders

Module 4:
Block Diagrams and Signal-
flow Graphs
Block Diagrams and Signal-flow Graphs

• For modelling purposes, it is often

desirable to break a system down into
readily identifiable individual elements
with known transfer functions.

2
• It is equally desirable to depict these
transfer functions graphically so that a
design engineer can quickly see the
interconnections among the various
parts of the system.

3
• Block diagrams and signal-flow graphs
are two useful graphical models for
showing these interconnections.

4
Block Diagrams
• The block diagram method yields a
graphical layout bearing some obvious
physical relationship to the actual
system.

5
• This is based on the concept that if the
transfer function of a system is given as
Y(s)
W(s) 
X(s)
then the system can be represented by a
rectangular box with input X(s) and
output Y(s), as shown below.
6
X(s) Y(s)
W(s)

7
 Block diagram elements
• The elements of a block diagram are:
 An arrow that indicates the direction
of signal flow.

8
A summing point which could have
any number of inputs, but only one
output.

9
A take-off point which could
have any number of take-offs, but
only one input.

10
 Rules for manipulating block diagrams
• In block diagram synthesis, it is
assumed that the block does not load,
and is not loaded by, its neighbours.
• These conditions can usually be realised
to a good degree by appropriate design
and are assumed to obtain in the
following combinations.

11
 Cascade combination

X(s)
W1(s)
Z(s)
W2(s)
Y(s)
≡ X(s) W1(s) W2(s)
Y(s)

12
Parallel combination

13
 Negative feedback combination

R(s) + E(s) Y(s) R(s) W1 (s) Y(s)

W1(s) ≡
- 1  W1 (s)W2 (s)
Z(s)

W2(s)

14
Forward transfer function:
Y(s)
 W1 (s)
E(s)
Open-loop transfer function:
Z(s)
 W1 (s)W2 (s)
E(s)

Error transform:
E(s)  R(s)  Z(s)

15
Error transfer function:
E(s) 1

R(s) 1  W1 (s)W2 (s)
since
E(s)  R(s)  W1 (s)W2 (s)E(s)

Closed-loop transfer function:

Y(s) W1 (s)

R(s) 1  W1 (s)W2 (s)
16
Summing point combination

17
Summing point movement

18
Take-off point movement

19
Example 3.01
Find Y(s)/X(s) for the system below.

20
Solution
Note that W2 and W4 are in parallel;
and W1 and W3 form a negative
feedback loop. Hence the reduced
diagram is

21
The obvious cascade connection
results in

and finally,
Y ( s) W1 W2  W1 W4

X ( s) 1  W1 W3  W1 W2 W5  W1 W4 W5
22
Example 3.02
Find Y(s)/X(s) for the system below.

The steps for a possible solution

approach are given below. Find out
what rules are applied at each step.
23
• To complete this solution, move the
inner take-off point over to the output
side; the resulting block diagram
contains only two negative feedback
combinations.

25
• The inner feedback loop has a gain of
1/W4 in its feedback path, while the
outer feedback loop has a corresponding
gain of (1+W3W4)/W3W4 in its
feedback path.

26
Signal Flow graphs
• The signal-flow graph (SFG) was
introduced by S.J. Mason for portraying
the cause-and effect representation of
linear systems that are modelled by
algebraic equations.

27
• It is a graphical means of depicting the
input-output relationship among the
variables of a set of linear algebraic
equations.

28
• Although it conveys the same
information as the block diagram, it is
often easier to construct, manipulate and
reduce than the block diagram.

29
•In a signal-flow diagram, the relationship
Y ( s)   W ( s) X ( s)
is depicted as

•Here, the signal transforms are

represented by nodes, and the transfer
function which relates them is
represented by a branch.
30
• An arrow indicates the direction of
signal flow, and branches are always
unidirectional.
• These constitute the only elements for a
complete signal-flow graph description
of a linear system.

31
Signal-flow graph terminologies
Input node (source): An input
node is a node that has only
outgoing branches.
Output node (sink): An output
node is a node that has only
incoming branches.

32
• In certain instances, when this condition is
not easily met, it is necessary to convert a
non-input node into an output node.
• This is done by simply connecting a branch
with unity gain from the existing node to a
new node also designated as, or having the
value of, the existing node.

33
Path: A path is any collection of a
continuous succession of branches
traversed in the same direction. This is a
completely general definition of a path
since it gives room for any node to be
traversed more than once.

34
Forward path: A forward path is a path
that starts at an input node and ends at
an output node, and along which no
node is traversed more than once.

35
Loop: A loop is a path that starts or
originates and ends or terminates on the
same node, and along which no other
node is crossed more than once.

36
Non-touching loops: In an SFG, two
loops are said to be non-touching if they
do not share a common node.

37
Path gain: A path gain is the product of the
branch gains encountered in traversing a
path.
Forward-path gain: The forward-path gain
is the path gain of a forward path.
Loop gain: The loop gain is the path gain of
a loop

38
Example 3.03
Consider the signal-flow graph which
results from the following set of
algebraic equations.
Y2 = G1Y1 + G2Y3
Y3 = G3Y2 +G4Y4
Y4 = G5Y2 + G6Y3 + G7Y4
Y5 = G8Y2 + G9Y4

39
•The SFG is shown below.
G7
G2 G4

Y1 Y2 G3 Y3 G6 Y4 G9 Y5
G1

G5

G8

40
• From this SFG, and keeping the above
definitions in mind, it can be deduced
that:
 Y1 is an input node
 Y5 is an output node. Other
possible output nodes are Y2, Y3,
and Y4
 Y1-Y2-Y3-Y2, Y2-Y3-Y4-Y3, Y1-Y2-
Y3-Y4-Y3-Y2, etc are paths
41
The forward path between Y1 and
Y2 is simply the connecting branch
between the two nodes. There are
two forward paths between Y1 and
Y3: the first one is along the branches
from Y1 to Y2 to Y3, and the second one
is along the branches from Y1 to Y2 to Y4
(through the branch with gain G5) and
then to Y3 (through the branch with gain
G4).
42
‒Also, there are two forward paths
between Y1 and Y4, and three
forward paths between Y1 and Y5.

43
Y2-Y3-Y2, Y3-Y4-Y3, Y4-Y4, and Y2-
Y4-Y3-Y2 are the only four loops in
the SFG
The path gain for the path Y1-Y2-
Y3-Y2 is G1G3G2
The forward path gain for the
forward path Y1-Y2-Y3-Y4-Y5 is
G1G3G6G9
44
The loop gain of the loop Y2-Y4-Y3-
Y2 is G5G4G2
Y2-Y3-Y2 and Y4-Y4 are non-
touching loops

45
Rule for Signal-flow graph manipulation

46
Addition rule
•The value of a variable designated by
a node is the algebraic sum of all
variables entering the node.

47
Y(s) = W1(s)X1(s) + W2(s)X2(s) + W3(s)X3(s)

48
Transmission rule

•The value of a variable designated by a

node is transmitted on every branch
leaving the node.

49
Y1(s) = W1(s)X(s)
Y2(s) = W2(s)X(s)
Y3(s) = W3(s)X(s)

50
Parallel connection

•Parallel branches in the same direction

connecting two nodes can be replaced
by a single branch with gain equal to the
sum of the gains of the parallel
branches.

51
52
Series connection
•A series connection of
unidirectional branches can be
replaced by a single branch with
gain equal to the product of the
branch gains.

53
Y1 G1 Y2 G3 Y3 G6 Y4 G9 Y5

Y1 G1G3G6G9 Y5

54
Example 3.04

Obtain Z1(s) and Z2(s) in terms of X1(s)

and X2(s).

55
Y(s) = W1(s)X1(s) + W2(s)X2(s)
Z1(s) = W3(s)Y(s)
= W3(s)W1(s)X1(s) +
W3(s)W2(s)X2(s)
Z2(s) = W4(s)Y(s)
= W4(s)W1(s)X1(s) +
W4(s)W2(s)X2(s)

56
Example 3.05
Obtain Z(s) in the following
figure.

57
 Y(s) = W1(s)X(s) + W2(s)Y(s)
Hence

W1 (s)
Y(s)  X(s)
1  W2 (s)
W1 (s)W3 (s)
Z(s)  W3(s)Y(s)  X(s)
1  W2 (s)

58
Gain formula for a signal-flow graph
• Computing the input-output relation of
an SFG or block diagram by mere
algebraic manipulation is quite involved.
• A simpler and faster approach is to
employ a general gain formula which is
described as follows.

59
• Given an SFG with P forward paths
and L loops, the gain between an
input node Yin and an output node Yout
is given as

60
 Yin P
Fi  i

Yout i 1 
where
Yin = input-node variable
Yout = output-node variable
P = total number of forward
paths between Yin and Yout
Fi = gain of the ith forward path
between Yin and Yout
61
Δ = 1 - (sum of the gains of all
individual loops) + (sum of products of
gains of all possible combinations of
two non-touching loops) – (sum of
products of gains of all possible
combinations of three non-touching
loops) + . . .

62
Δi = Δ for that part of the SFG which is
non-touching with the ith forward path.

63
• It should be noted that the gain formula
can only be applied between an input
node and an output node

64
Example 3.06
Consider the single-loop feedback
control system shown below.

R(s) + E(s) G(s) Y(s)

- y

H(s)

65
The SFG is shown below.
G(s)
R(s) 1 E(s) Y(s) 1 Y(s)

-H(s)

66
•By inspection of the SFG, the
following results are obtained:

67
1. Since there is only one forward path
between R(s) and Y(s), the closed-loop
transfer function Y(s)/R(s) is

Y ( s) 1
Fi  i F11
 
R( s) i 1  

68
2. The gain, F1, of the forward path is
F1(s) = G(s)
3. There is only one loop, and the gain of
this loop is
L11 = -G(s)H(s)
4. Since there is only one loop,
Δ = 1- L11 = 1 + G(s)H(s)

69
5. Also, there are no non-touching
loops, and the forward path is in
touch with the only loop in the SFG.
Therefore,
Δ1 = 1
6. And finally, the closed-loop transfer
function is
Y ( s) G ( s)  1 G ( s)
 
R( s) 1  G ( s) H ( s) 1  G ( s) H ( s)
70
Class work 3.1

Determine the gain between Y1 and Y5

of the SFG shown below.

71
 G7
G2 G4

Y1 Y2 G3 Y3 G6 Y4 G9 Y5
G1

G5

Fig C4.1 G8

72
Class work 3B

For the system given below, obtain the

overall closed-loop transfer function
using the block diagram reduction
techniques and the SFG gain formula.

Fig C4.2
73
74