SIGNAL FLOW GRAPHS

By:
Prof. K Puttaswamy,
PESCE, Mandya

eNotes for the session on 11.04.05 on VTU

EDUSAT Programme

Control Systems - Prof K Puttaswamy, PESCE, Mandya 1

 SIGNAL FLOW GRAPH
The block diagram is useful for graphically representing control systems. For a complicated system,
however the block diagram reduction process becomes time consuming . An alternate approach for finding the
relationships among the system variables of complicated control system is the signal flow graph approach due
to Mason.
Signal flow graph is a diagram which represents a set of simultaneous linear algebraic equation .When
applying the signal flow graph method to control system we must first transform linear differential equation into
algebraic equation in ‘s’
Signal flow graph consist of a network, in which nodes are connected by directed branches. Each node
represents a system variable and each branch connected between two nodes acts as a signal multiplier. Note that
signal flows only in one direction .The direction of the signal flow is indicated by an arrow placed on the branch
and the multiplication factor is indicated along the branch. The signal flow graph depicts the flow of signals
from one point of a system to another and gives the relationship among the signals.
Signal flow graph contains essentially the same information as a block diagram. The advantage of using
signal flow graph is to represent a control system is that a gain formula or Mason’s gain formula is available
which gives the relationships among the system variables without requiring a reduction of the graph
TERMINOLOGY
Node: A node is a point representing a variable or signals.
Transmittance : is a gain between the two nodes.
Branch : is a directed line segment joining two nodes. The gain of the branch is a transmittance.
Input node or source : is a node which has only outgoing branches. This corresponds to an independent
variable .
Output node or Sink : An output node or sink is a node which has only incoming branches. This corresponds
to a dependent variable.
Mixed node : is a node which has both incoming and outgoing branches.
Path : is a traversal of connected branches in the direction of the branch arrows.
Loop : is a closed path.
Loop gain : is the product of the branch transmittance of a loop.
Non touching loops : Loops are non-touching if they do not posses any common nodes.

Control Systems - Prof K Puttaswamy, PESCE, Mandya 2

Forward path : A forward path is a path from an input node to an output node which does not cross any nodes
more then once.
Forward path gain : is the product of the branch transmittances of a forward path.

Mixed mode x4 input node

Input node
x1 a x2 b x3 1 x3 Output node

C
Properties of signal flow graph

1. A branch indicates the functional dependence of one signal upon another. A signal passes through only
in the direction specified by the arrow of the branch .
2. A node adds the signals of all incoming branches and transmits this sum to all outgoing branches.
3. A mixed node which has both incoming and outgoing branches may be treated as an output node by
adding an outgoing branch of unity transmittance.
4. For a given system, signal flow graph is not unique. Many different signal flow graphs can be drawn for
a given sytem by writing the system equations differently.
Signal flow graph Algebra
The independent and dependent variables of the equations become the input nodes and output nodes
respectively. The branch transmittance can be obtained from the coefficients of the equations. To determine
the input output relationship we may use Mason’s formula or we may reduce the signal flow graph to a
graph containing only input and output nodes we use the following rules
1. The value of a node with one incoming branch as shown in fig. 1 x2 = ax1
2. The total transmittance of cascaded branches is equal to the product of all the branch transmittances.
Cascaded branches can be combined into single branch the multiplying the transmittance as shown in
fig. 2
3. Parallel branches may be combined by adding transmittances as shown in fig. 3

Control Systems - Prof K Puttaswamy, PESCE, Mandya 3

 4. A mixed node may be eliminated as shown in fig. 4
5. A loop may be eliminated as shown in fig. 5

a a b ab
=
x1 x2 x1 x2 x3 x1 x3
x2 = ax1 x3 = abx1
fig.(1) Fig. (2)

x1 x2 a+b x2 = (a + b) s1
= x1 x2

b Fig. (3)

x1 x1
a ac
c x4 x4
b x3 = bc
x2 x2 x4 = (ac)x1
x4 = (bc)x2
Fig. (4)

x1 a x2 b x3 x1 ab x3 ab
= = x1 1 - bc x3

c bc
Fig. (5)

x3 = bx 2 x3 = abx1 + bcx3
x2 = ax1 + cx3 x3 = abx1 + b2cx2
= abx1 + b2c (ax1 + cx3)
= abx1 + b2cax1 + b2c2x3
x3 (1 – b2c2 ) = abx1 (1 + bc)
abx1 (1 + bc)
x3 =
1 – b2c2

Control Systems - Prof K Puttaswamy, PESCE, Mandya 4

 ab (1 + bc )
= X1
( 1 + bc) (1 - bc)

ab
x3 = x1
1 - bc

Signal flow graph representaion of linear systems

Here the graph can be drawn from the system equations or can be drawn by inspection of the physical
system.
Consider system defined by the following set of equations
x1 = a11x1 + a12x2 + a13x3 + b1u1 - (1)
x2 = a21x1 + a22x2 + a23x3 + b2u2 - (2)
x3 = a31x1 + a32x2 + a33x3 - (3)

a11 a22
b1
x2
u1 x1 x2 x3 x1 a21 x3

a12 b2 a23

a13 u2
b
a31

a33

x1 x2 a32 x3

u1, u2 input variables

x1x2 & x3 are output variables

Control Systems - Prof K Puttaswamy, PESCE, Mandya 5

 Final signal flow graph is drawn by combining these three equations of simultanious equations.

a31

a22
a11 a33
b1 a21 a32 1
u1 x1 x2 x3 x3

a12 b2 a23

u2

a13

Mason’s Gain formula for signal flow graph

We want to determine the relationship between n an input and output variable of the signal flow graph.
The transmittance between an input node and an output node is the overall gain or transmittance between the
two nodes.
Mason’s gain formula p = k pk k

Pk = path gain or transmittance of kth forward path

= determinant of the graph = 1 – (Sum of all different loop gains) + (sum of gain products of all
possible combinations of two nontouching loops) – (Sum of the gain products of all possible
combinations of three nontouching loops)

=1– La + LbL c - LdLeLf + …………

a b,c d,e,f

k = Cofactor of kth forward path, determinant of the graph with the loops touching the kth forward path
removed
Or value of for all loops except for those which touch the forward path ‘K’

Control Systems - Prof K Puttaswamy, PESCE, Mandya 6

 OR
Value of above by eliminating all loop gains & associated products which are touching to the kth forward
path.

BLOCK DIAGRAM

R(s) G(s) C(s) R(s) G(s) C(s)

R(s) + G(s) C(s) R(s) 1 E(s) G(s) C(s)

-

H(s) -H(s)

REFERENCE: 1. MODERN CONTROL ENGINEERING BY OGATA

2. AUTOMATIC CONTROL SYSTEMS B.C. KHO
3. CONTROL SYSTEM BY V.A. BAKSHI & U.A. BAKSHI

Control Systems - Prof K Puttaswamy, PESCE, Mandya 7

PROBLEM 1 :

H2

+ +
G1 +
- G2 G3
- +
R(s) C(s)

H1

Signal flow graph for the above Block Diagram is

(2)

R(s) 1 1 G1 G2 G3 1
C(s)
(1)

H1 (3)

-1

In this system there is only one forward path between the input R(s) and output C(s). The forward path gain is
P1 = G1 G2 G3
In this figure there are three individual loops. The gains of these loops are
L1 = G1 G2 H1
L2 = - G2 G3 H2

Control Systems - Prof K Puttaswamy, PESCE, Mandya 8

 L3 = - G1 G2 G3
Since all the three loops have a common branch, there are no nontouching loops. Hence determinant
is given by
= 1 - [L1+L2+L3]
= 1 - [G1G2H1 – G2G3H2 – G1G2G3]
The cofactor of the determinant along the forward path connecting the input node & output node by
removing the loops that touches this path. Since the path p1 touches all the three loops.
We get 1 =1

Overall given c(s) p1 1

=
R(s)
G1G2G3
c(s)
R(s) =
1 - G1G2H1 + G2G3H2 + G1G2G3

2. Use the signal flow graph method determine the gain c/R for the block diagram shown.

G1

1 2
R 3 + 4 5
G1 G2 G3
+ + + C
- +

H2

H1

Control Systems - Prof K Puttaswamy, PESCE, Mandya 9

Signal flow graph for the fig. Shown

k Pk k

G4
loop (3)

R 1 G1 G2 G3 1 C

loop (1)

H2

loop (2)
- H1

P=
Mason’s formula

No. of forward path p1 = G1G2G3

P2 = G1G2G4
There are three feed back loops
L1 = G1G2H2
L2 = - G1G2G3H1
L3 = - G1G2G4H1

= 1 – (L1 + L2 + L3)
= 1 – (G2G3H2 – G1G2G3H1 – G1G2G4H1)

L1 touches L3, L2 touches L3 and L1 touches L2

No combinations of non touching loops can exists for the signal flow graph given, only first two terms exsists.

1 is obtained from
1=1
2= 1

Control Systems - Prof K Puttaswamy, PESCE, Mandya 10

 p1 1 + p2 2
p =

G1G2G3 + G1G2G4
=
1 - [G1G2G3H2 – G1G2G3H1 – G1G2G4H1]

G1G2G3 + G1G2G4
c/R = p =
1 - G1G2H2 + G1G2G3H1 + G1G2G4H1

3. The following equation describes a control system. Construct the signal flow graph for it and obtain
transfer function .
Y2 and Y2
U1 for u2 = 0 U2 for u1 = 0
Where Y2 = out put node and u1 & u2 are the inputs
Y1 = a11 Y1 + a12 Y2 + b1 U1 ------ (1)
Y2 = a21 x a22 Y2 + b2 U2 --------- (2)
Solution
The different node variables are Y1 & Y2 ,U1 & U2 are the inputs.
Consider the equation signal flow graph is
a11
b1
U1 Y1

y2

a12

a22
Similarly considering the equation 2
a21
Signal flow graph is Y1 Y2

b2

u2

Control Systems - Prof K Puttaswamy, PESCE, Mandya 11

Combining these two signal the graphs, the total signal graph is

a11 a22

b1 a21
1
u1 output
Y1 Y2

a12 b2

u2

Transfer function Y2

u1 u2 = 0
Assuming u2 = 0

Control Systems - Prof K Puttaswamy, PESCE, Mandya 12

Signal flow graph is

a11 a22

b1 a21 1
U1 Y1 Y2

a12

k Pk k
Using Mason’s gain formula T.F =

k=1
k = No of forward paths = 1
P1 = b1 a21
L1 = a12 a21
L2 = a11
L3 = a22
L2 & L3 are non touching loops

Control Systems - Prof K Puttaswamy, PESCE, Mandya 13

 ∴ = 1 – [ all individual feed back loop gains]
+ [ gain products of all possible combinations of two non touching loops]

= 1 – [ L1 + L2 + L3 ] + L2 L3
1 = cofactor of the determinant eliminating all loop gains which are touching the first forward path.
1 =1
Y2 P1 1 b1a21
U1 =
1– a12a21 –a11 – a22 + a11a22
Assuming u1 = o
Then the graph is
a11
a22

a21
Y1 Y2

a12 b2

u2
Y2
T.F =
U2 U1 =0

K = No of forward path = 1
P 1 = b2
Individual loops are
L1 = a21 a12
L2 = a11
L3 = a22
nontouching loops are L2 & L3

Control Systems - Prof K Puttaswamy, PESCE, Mandya 14

 = 1 – [ L1 + L2 + L3 ] + L2 L3
= 1 – a21 a12 – a11 – a22 + a11 a22
1= 1 – a11
Y2 p1 1 b2 (1-a11)
= =
U2 1-a21a12 – a11 – a22 + a11 a22

4. Construct the signal flow graph for the following set of system equations
Y2 = G1Y1 + G3Y3 --- --- (1)
Y3 = G4Y1 + G2Y2 + G5Y3 --- --- (2)
Y4 = G6Y2 + G7Y3 --- --- (3)

Find the transfer function Y4

Y1
Solution
Consider the equation 1
Signal flow graph is

Y1 G1 Y2 Y3

G3
G4

Consider the equation 2, signal flow graph is G5

Y1
Y2 G2 Y3

G6

Consider the equation 3, signal flow graph is

Y2 Y3 G7 Y4

Control Systems - Prof K Puttaswamy, PESCE, Mandya 15

Combining all the three, complete signal flow graph is G4
G5

G1 G2
Y2 G7
Y1 Y3 Y4

G3
No of forward path = K = 4
4
Transfer function = pk Ak
K =1 G6

= p1 1 + p2 2 + p3 3 + p4 4

P1 = G1G2G7, P2 = G4G7, P3 = G1G6 & P4 = G4 G3 G6

Individual loops are

L1 = G2 G3 L2 = G5 (Self loop )
= 1 – (L1 + L2 )
= 1 – G2G3 –G5
There are no nontouching loop combinations

1 =1 Booth loops are touching

2 = 1 Booth loops are touching
3 = 1– G5 G5 is non touching the path p3

4 =1

Y4 G1G2G7 + G4G7 + G1G6 (1 – G5 ) + G4G3G6

=
Y1 1 – G2G3 – G5

Control Systems - Prof K Puttaswamy, PESCE, Mandya 16

 5. From the given block diagram, draw the signal flow the graph and find C(s)
R(s)

G4

1 2
6
R(s) 5 + 7 C(s)
+ + G1 + G2 G3 +
- - 3 -
4

H2
H1

Signal Flow graph is

G4

R(s) 1 1 2 1 3 G1 4 G2 5 G3 6 7 1 C(s)

-H2
-H1

-1

Control Systems - Prof K Puttaswamy, PESCE, Mandya 17

No of forward paths = K = 2
Forward path gains are P1 = G1 G2 G3
P2 = G4

Feed back loops are

L1 = - G1G2H1 L4 = - G4
L2 = - G2G3H2 L5 = G2G4H1H2
L3 = - G1G2G3
There are no non touching loops
= 1 – [ L1 + L2 + L3 + L 4 + L5 ]
= 1 + G1G2H1 + G2G3H2 + G1G2G3 + G4 – G2G4 H1H2
considering the forward path G1G2G3
all loops touching ∴ 1=1
Considering the forward path G4 , all the loops are touching the path G4 , 2 =1
C(S) p1 1+ p2 2
=
R (s)
G1G2G3 + G4
=
1+G1G2H1 + G2G3H2 + G1G2G3 + G4 – G2G4 H1 H2

6. Given the Electrical Network

1. Find out the laplace transform of the given network and re draw the network in ‘S’ domain
2. Work down the equations for different branch current and node voltages.
3. Simulate each equation by drawing the corresponding signal flow graph.
4. Combine all the signal flow graphs to get the total signal flow graph.
5. Use Mason’s gain formula to derive the transfer function of the given network.
6. Formal the transfer function for the given network.

Control Systems - Prof K Puttaswamy, PESCE, Mandya 18

 R1 R2

Vi C L Vo

Laplace transform of the given network is as shown

R1 R2
V1(s)

I1(s) 1 I2(s)
Vi(s) LS Vo (s)
SC

Equations are
I1(s) = Vi (s) – V1 (s) …… (1)
R1

[ I1(s) – I2 (s)]
V1(s) = ……. (2)
CS
V1(s) - V0 (s)
I2 (s) = …………. (3)
R2
V0(s) = I2(s) LS

Control Systems - Prof K Puttaswamy, PESCE, Mandya 19

Signal flow graph for the equation (1)

1
Vi R1 I1 V1

- 1
R1

1
Signal flow graph for equation 2 I1 Sc V1 I2

-1
Sc

Signal flow graph for equation 3

1
V1 R2 I2 Vo

-1
R2
Signal flow graph for equation 4
I2 SL Vo

Total signal flow graph for the network is

1 1 1
Vi R1 I1 Sc V1 R2 I2 SL Vo

1 1 1
- R - Sc - R
1 2

Control Systems - Prof K Puttaswamy, PESCE, Mandya 20

Using Mason’s gain formula

Vo PK k
= =
Vi

No of forward paths = 1
V0 P1 1
Vi =

L
Forward path gain Ti =
R1R2C
Feedback loops are
1 1 SL
L1 = - L2 = - L3 = -
SR1C SR2C R2

Non touching Loops are L1 & L3

= 1 – [ L1 + L2 + L3 ] + L1L3
1 1 SL L
=1+ + + +
SR1C SR2C R2 R1R2C

All the loops are touching the forward path L

R1R2C

L
V0 R1R2C SL
= =
Vi 1 1 SL L SR1R2C + R2 + R1 + S2LR1C + SL
1+ + + +
SR1C SR2C R2 R1R2C

Control Systems - Prof K Puttaswamy, PESCE, Mandya 21

 6. Find the transfer function for the given network using Mason’s gain formula

R1
Vi Vo R2

Laplace transformed network is 1 1

CS 1 +SR1C
1
CS
R1
1 + SR1C
Z
R1

Vi(s) R2 Vo (s) = Vi (s) I(s) R2 Vo (s)

R1 x 1/Sc
Z= R1 + 1 /Sc

I (S ) = Vi(s) – Vo(s)
Z Z= R1
R1Sc + 1
= Vi(s) – Vo(s) 1 + R1Sc
R1

Vo (s) = I (s) R2

Control Systems - Prof K Puttaswamy, PESCE, Mandya 22

From the equation (1) signal flow graph is

1+ R1SC
R1
Vi 1 Vo

- 1 + R1Sc
R1

From the equation (2) signal flow graph is

I(s) R2 Vo(s)

Then combined signal flow graph is

1 + R1Sc
R1 1 R2 Vo
Vi

1+R1Sc
- R1

Using Mason’s gain formula, the number of forward path = 1

∴ forward pathg gain P1 = R2 1 + R1 Sc

R1

Loop gown L1 = - R2 1 + R1 Sc
R1

Determinant of the graph = 1 – L1

= 1 + R2 ( 1 + R1Sc )
R1
R1 + R2 ( 1 + R1Sc )
=
R1

Control Systems - Prof K Puttaswamy, PESCE, Mandya 23

As L1 is touching the forward path
1=1
then
Vo (s) P1 1 R2 ( 1 + R1SC ) 1 R2 (1 + R1 Sc)
T(S) = = = =
Vi(s) R1 [ R1 + R2 (1 + R1SC)] R1 + R2 (1 + R1 Sc )
R1

8. BLOCK DIAGRAM FOR THE SIGNAL FLOW GRAPH

Procedure :

1. To obtain the block diagram from the given signal flow graph, it is necessary to write set of
system equations representing the given signal flow graph.
2. Assume suitable node variables, write equations for every node.
3. while writing equations remember that the value of the variable represented by the node is the
algebraic sum of all the signals entering at that node. The outgoing branches have no effect on
the value of the node variable.

(1) Obtain the Block diagram from the signal flow graph, shown below

2S Y 3 2 W 1
Z U
X

-4 -3

-5

Write the equations for various node variables Y, Z, W and U, Input node is X. There are only outgoing
branches.
Y = (2S ) X – 4Z - 5U
Z = 3Y – 3W
W = 2Z
U = 1 W + SZ
The block digram simulation of various equations are

Control Systems - Prof K Puttaswamy, PESCE, Mandya 24

Equation (1)

X
2S + Y
-
-

4
From Z

5
From U

From equation (2)

Z
Y
-
3 +

3 From W

For equation (3)

Z 2 W

For equation (4)

W + U
+

From Z S

Control Systems - Prof K Puttaswamy, PESCE, Mandya 25

Combining all the block diagrams, the complete block diagram is

4 S

X - Y Z W +
2S + 3 + 2 U
- -

REFERENCE: 1. CONTROL SYSTEMS BY V.A. BAKSHI & U.A. BAKSHI

2. CONTROL SYSTEMS BY DR.GANESH RAO

Control Systems - Prof K Puttaswamy, PESCE, Mandya 26