EEE405:

ANALOGUE CIRCUIT DESIGN

LECTURE NOTES
NETWORK SYNTHESIS

By:
Olorunniwo O.
Dept. of Electronic & Electrical Engineering
Obafemi Awolowo University
Ile-Ife
 2

Network synthesis:
• Realizability of driving point impedance;
• Synthesis of two-terminal networks;
• Foster and Cauer Form realizations;
• Minimum phase and non-minimum phase networks.

1. Introduction to Network Analysis and Synthesis

When a network is made of linear circuit elements, the network is aptly described by its system or transfer
function 𝐻(𝑠). The response 𝑅(𝑠) and the excitation 𝐸(𝑠) are related by the equation:

𝑅(𝑠) = 𝐻(𝑠)𝐸(𝑠)

Thus, in the network analysis given 𝐸(𝑠), it is possible to obtain 𝐻(𝑠) directly from the network and
determine 𝑅(𝑠). Conversely, in the network synthesis, given the excitation 𝐸(𝑠) and the response
𝑅(𝑠), it is possible to determine the network transfer function 𝐻(𝑠), such that:

𝑅(𝑠)
𝐻(𝑠) =
𝐸(𝑠)
In a given network, 𝐸(𝑠) and 𝑅(𝑠) are either currents or voltages and 𝐻(𝑠) is generally denoted as
immittance. The driving-point immittance is defined as the function by which the variables measured at the
same port are related to each other (See Figure 1).
Therefore, the driving-point impedance of the circuit in Fig.1 is defined as 𝑍(𝑠), such that:
𝑉(𝑠)
𝑍(𝑠) =
𝐼(𝑠)

Figure 1

Moreover, the driving-point admittance of the circuit is defined as 𝑌(𝑠), such that:
𝐼(𝑠)
𝑌(𝑠) =
𝑉(𝑠)

A major challenge of driving-point synthesis is to ascertain if the transfer function given is realizable. The
realizability of a given function is achievable if such system functions could be represented with
positive resistances, inductances and capacitances. Note: The transfer function or transmittance is a
system function that describes the variables that are measured at different ports. Consider: Fig.2
illustrated a two-port network, if the excitation is 𝐼1 (𝑠) and the output is the response 𝑉2 (𝑠), the
driving-point transfer impedance is given by:
 3

𝑉2(𝑠)
𝑍21 (𝑠) =
𝐼1(𝑠)

While, the transfer voltage ratio is determined to be:

𝑉2 (𝑠)
𝐻21(𝑠) =
𝑉1 (𝑠)

Figure 2

The most important aspect of network transfer function synthesis is filter design.

2. Synthesis of two-terminal networks

In electric network theory, the network function could be described either as one- or two-port
network. A one-port function is completely ‘described’ or specified when the voltage-current
relationship at the terminal is known. Fig. 3 illustrated a one-port network.

Figure 3

Thus, the input or driving point impedance of the one-port is given by,
𝑍𝑖𝑛 = 𝑉
𝐼
Note: If for example the 𝑉 = 10𝑉 and 𝐼 = 2𝐴, then the driving-point impedance is

𝑍𝑖𝑛 = 𝑉 = 10 = 5Ω
𝐼 2

However, the network impedance could be a single5Ω , or two 2.5 Ω resistors in series, or two 10Ω
resistors in parallel. Therefore, the main importance or primary concern is the voltage-current
relationship.
Fig.4 show a two-port network which is generally described by the variables 𝐼1, 𝑉1, 𝐼2 𝑎𝑛𝑑 𝑉2. From
these dependent and independent variables, two pairs of voltage-current relationship could be established.
 4

Figure4

Thus, there are six possible equations which describe a two-port network. Four most useful
descriptions are described by z-parameters, y-parameters, h-parameters, and ABCD-parameters.
Example 1: If the system function of a network is given as H (s ) =
1
. Find the response
(s + 2)(s + 3)
R(s ) if the excitation is E (s ) =
3
s
Solution:
The network system function is defined as:

R(s )
H (s ) =
E (s )

Where, the H (s ) is the driving-point immittance, E (s ) the excitation and R(s ) is the response.

Thus if H (s ) = and E (s ) = , then the response R(s ) is given by

1 3
(s + 2)(s + 3) s

R(s ) =
1 3 3
 =
(s + 2)(s + 3) s s(s + 2)(s + 3)

Example 2: Given the driving-point function Z (s ) = 3 + 2s +

1
, find the simplest network
3s
realization
Solution:
The driving-point impedance of the network suggested a simple RLC network: a resistor of 3 , an
inductor of 2H , and a capacitor of 3F in series with a voltage source (See Fig. 5).

Figure 5
 5

Review Questions:
1. Given the driving-point admittances, find the simplest network realization.
a. Y (s ) = 2s +
3s
,
s+2
b. Y (s ) =
1 2s
+ 2 ,
3s + 2 s + 4

2. Figure P.1 shows a network. Write the mesh equation in terms of

a. Differential equations;
b. Complex-frequency variable, s

Figure P.1

2.1 The z parameters

The z parameters are a particular set of equations that describes a two-port network with dimensions
of impedance. The set of equations are given by,

V1 = z11 I1 + z12 I 2
V2 = z 21 I1 + z 22 I 2
Where V1 and V 2 are the independent variables, while I 1 and I 2 are the dependent variables. Further, the
individual z parameters are defined as:

V1 V1
z11 = (open-circuit driving-point impedance) z12 = (open-circuit transfer impedance)
I1 I 2 =0
I2 I1 = 0

V2 V2
z 21 = (open-circuit transfer impedance) z 22 = ( open-circuit driving-point impedance)
I1 I 2 =0
I2 I1 = 0

Note: Each dimension are derived or specified when either I1 or I 2 , in one of the ports, is zero;
hence, the term open-circuit driving-point impedances. Thus, in Fig.4, for example, z11 and z 22
corresponds to the voltage-current relationship in port 1-1’ and 2-2’. Further, z12 or z 21 relates the
voltage in port1 to the current in the other or vice-versa. Hence, the term open circuits transfer
impedances.
 6

Example 3: Find the open circuit parameters of the T-circuit in Figure 6.

Figure 6

Solution: By inspection the z parameters are

V1 V2 V1
z11 = = Za + Zc , z 22 = = Zb + Zc z12 = z 21 = = Zc
I1 I2 I1 =0
I2 I1 =0
I 2 =0

It is noteworthy that z12 = z21 . Moreover, when the open-circuit transfer impedances of a two-port
network are equal, the network is reciprocal.

2.1.1 Characterization of two-port networks

Consider: A set of node equations of a network (active or passive) with dependent voltages
V1 ,V2 ,V3 ,Vk at ports 1 and 2, while the independent currents are I 1 and I 2 , such that:

I 1 = n11V1 + n12V2 + n13V3 +  + n1k Vk

I 2 = n21V1 + n22V2 + n23V3 +  + n2 k Vk
0 = n31V1 + n32V2 + n33V3 +  + n3k Vk (1)

0 = nk1V1 + nk 2V2 + nk 3V3 +  + nkkVk

1
where, n ij represents the admittance between the ith and jth nodes, that is nij = Gij + sC ij + .
sLij
Further, if the circuit is made up of RLC elements then nij = n ji . In addition, let ijth cofactor of the
determinant be  ij . Thus, if  ij =  ji a condition referred to as reciprocity ensues, such that z12 = z 21 .

Therefore, the simple node equations of a two-port network could be resolved into cofactors, to
obtain
 
V1 = 11 I 1 + 12 I 2
 
 
V2 = 21 I 1 + 22 I 2
 
where,
 7

11 12
z11 = z12 =
 
 21  22
z 21 = z 22 =
 

Note: If a passive network is reciprocal, then z 21 = z 21 .

Example4: Find the z parameters of the π-circuit in Fig. 7.

Figure 7

Solution:
Assume I 2 = 0 , then, current I 1 is due to the voltage drops across V1 and V 2 , that is:
I1 = V1 (YA ) + (V1 − V2 )YB = V1 (YA + YB ) − V2YB
Moreover, if I 1 = 0
I 2 = V2 (YC ) + (V2 − V1 )YB = −V1 (YB ) + V2 (YB + YC )

In matrix form,
 I 1  YA + YB − YB  V1 
I  =  − Y YB + YC  V2 
 2  B

where,  = (Y A + YB )(YB + YC ) − YB2 = Y AYB + Y AYC + YB YC ;

Moreover,
11 = YA + YB 12 = −YB
 21 = −YB  22 = YB + YC

Therefore, the open-circuit parameters of the π-circuit are:

Y A + YB − YB
z11 = z12 =
Y AYB + Y AYC + YB YC Y AYB + Y AYC + YB YC
− YB YB + YC
z 21 = z 22 =
Y AYB + Y AYC + YB YC Y AYB + Y AYC + YB YC

Consider examples 3 and 4, what is the relationship between the immittances of the two
circuits?
 8

2.2 y parameters

The short-circuit admittances of a two-port network defines the y parameters. Suppose a set of node
equations of a network (active or passive) is with independent voltages sources V1 , and V2 at ports 1 and
2, while the dependent currents I 1 I 2 ,... I k , such that:

V1 = m11 I 1 + m12 I 2 + m13 I 3 +  + m1k I k

V2 = m21 I 1 + m22 I 2 + m23 I 3 +  + m2 k I k
0 = m31 I 1 + m32 I 2 + m33 I 3 +  + m3k I k (2)

0 = mk1 I 1 + mk 2 I 2 + mk 3 I 3 +  + mkk I k

Where, mii represent the sum of the impedances in the ith mesh and mij represent admittances
between the ith and the jth mesh. In addition, an RLC networks suggest reciprocity, that is, mij = m ji .
The mesh equation of (2) of a two-port network could be resolved to be:

11 
I1 = V1 + 12 V2
 
 
I 2 = 21 V1 + 22 V2
 
where,
11 12
y11 = y12 =
 
 21  22
y 21 = y 22 =
 

Therefore, the parameters y11 , y12 , y 21 , and y 22 denotes the short-circuit admittance parameters. This is so
derived because: To obtain y11 and y 21 , the voltage source V 2 must be short-circuited, while V1 must
be short-circuited to obtain y 22 and y12 . Hence, the term short-circuit parameters.

Example 5
Determine the y parameters of the π-circuit in Fig.7.

Solution:
If V2 = 0 , y11 = YA + YB
Further, if V1 = 0 y 22 = YB + YC

However, y12 = y 21 = −YB

 9

Example 6
Find the y parameters of the bridge-T circuit in Fig. 8.

Figure 8

Solution:
The mesh equations of loop 1, 2 and 3 are as follows:
Loop 1:
V1 − (I 1 − I 3 ) − (I 1 − I 2 ) = 0
1
s
 1 1
V1 = 1 +  I 1 + I 2 −   I 3
 s s
Loop 2:
V2 − ( I 2 + I 3 ) − ( I 2 + I 1 ) = 0
1
s
 1 1
V 2 = I 1 + 1 +  I 2 +   I 3
 s s
Loop 3:
(I 3 − I 1 ) 1 + (I 3 + I 2 ) 1 + 2 I 3 = 0
s s
1 1  1
−   I 1 +   I 2 + 21 +  I 3 = 0
s s  s

In matrix form,
 1 1 
1 + 1 − 
V1   s s   I1 
V  =  1 1+
1 1 I 2 
 2  s s  
 0   1 1  1   I 3 
− 21 + 
 s s  s 

Thus,
 10

2(2s + 1)
= ;
s2
2s 2 + 4s + 1
11 =  22 = ;
s2
2s 2 + 2s + 1
12 =  21 = − ;
s2

If y parameters are explicitly defined by,

Therefore,
I1 I1
y11 = y12 =
V1 V2 = 0
V2 V1 = 0

I2 I2
y 21 = y 22 =
V1 V2 = 0
V2 V1 = 0

2s 2 + 4s + 1
y11 = y 22 =
2(2 s + 1)
2s 2 + 2s + 1
y12 = y 21 = −
2(2 s + 1)

Note: If y11 = y 22 or z11 = z 22 , the network is symmetrical. A symmetrical network is which remained
unchanged despite the interchange of port designation.

2.3 The h parameters

The h parameters are useful in the description of transistor circuits. The h parameters are defined by
the equations:
V1 = h11 I1 + h12V2
I 2 = h21 I1 + h22V2
where,

V1 V1
h11 = h12 =
I1 V2 = 0
V2 I1 = 0

I2 I2
h21 = h22 =
I1 V2 = 0
V2 I1 = 0

Noticeably, h11 and h21 are short-circuit type parameters, and h12 and h22 are open-circuit type
parameters. Moreover, h11 could be interpreted as the input impedance at port 1 with port 2 shorted
out. Obviously,
1 1
h11 = and h22 = ,
y11 z 22
 11

In addition, h12 and h21 are voltage and current transfer ratios, and for reciprocity h12 = −h21 . Thus, h12
and h21 are defined by,
y12 y
h21 = − , and h12 = 12
y11 y11
Example 7
Consider the hybrid equivalent circuit in Fig.9. Find the input impedance Z in .

Fig.9

Solution:
Observe the voltage-controlled source h21V2 at port 1 and the current-controlled source at h21 I 1 at
port 2. Then,

V1 = h11 I 1 + h12V2
V2 = − I 2 Z L = −(h21 I1 + h22V2 )Z L

Solving for V 2 ,
h21 Z L I 1
V2 = −
1 + h22 Z L
Now, substituting V 2 into V1 = h11 I 1 + h12V2 , therefore

 h h Z I 
V1 =  h11 − 12 21 L 1  I 1
 1 + h22 Z L 
So that,
h h Z I
Z in = h11 − 12 21 L 1
1 + h22 Z L

Example 8
Obtain the h parameters of the π circuit in Fig. 7.

Solution:
 12

V1 1 YA + YB I2 1 YAYB + YAYC + YBYC

h11 = = = h22 = = =
I1 V2 =0
y11 YAYB + YAYC + YBYC V2 V2 =0
z 22 YB + YC

y 21 − YA + YB
h21 = − h12 = =
y11 YB

Example 9
The negative impedance converter (NIC) converts positive load impedance into negative impedance at the
input port. Determine the h parameters of the device shown in Fig. 10.

Figure 10

Solution:
The input impedance,
Z in = − Z L
The above expression is expressed as,
V1 − V 2
=
I1 I2
Further, the voltage and current relationships are expressed by,
V1 = kV2
I 1 = kI2
Compared to,
V1 = h11 I1 + h12V2
I 2 = h21 I1 + h22V2
1
h12 = =k
h21
h11 = h22 = 0
0 k
Therefore, the NIC is nonreciprocal and the h parameters matrix is  1 .
 k 0


2.4 The ABCD parameters

The A, B, C , D parameters, whose transition matrix is known as the transmission matrix, relates the
voltage and current at the input port to the corresponding quantities at the output. That is,
 13

V1   A B   V2 
 I  = C D  − I 
 1   2 
The current I 2 carried a negative sign, since the output current is regarded as coming out of the
output port instead of ‘into’ the port. More so, the ABCD parameters could be expressed as:

z11 1
A= B=−
z 21 y 21
1 y11
C= D=−
z 21 y 21

From observation,
A is an open-circuit voltage ratio;
B is a short-circuit transfer impedance;
C is an open-circuit transfer admittance;
D is a short-circuit current ratio.

To describe the short-circuit transfer function B and D in terms of y parameters, and the open-
circuit transfer functions A and C in terms of z parameters,

V1 V1
A= B=
V2 I 2 =0
I2 V2 = 0

I1 I1
C= D=−
V2 I 2 =0
I2 V2 = 0

Moreover, the reciprocity conditions of the A, B, C, D parameters is expressed by,

 A B
det   = AD − BC = 1
C D

Example10
Obtain the ABCD parameters of an ideal transformer in Fig. 11 defined by the equations:

V1 = nV2

I1 =
1
(− I 2 )
n

Figure 11

Solution:
Let the matrix form be given by,
V1  n 0   V2 
 I  = 0 1  − I 
 1   n   2 
 14

Hence the transmission matrix of the ideal transformer is,

 A B  n 0 
C D = 0 1 
   n 
The ideal transformer terminated in load impedance defined by the set of equations:
V1 = nV2
V2
I1 =
nZ L
The input impedance at port 1 is determined by,
V
Z1 = 1 = n 2 Z L
I1
Therefore, an ideal transformer is an impedance transformer.

2.5 The relationship between two-port parameters

Table 1 provides a conversion table to facilitate the conversion between two-port parameters. The
relationships between two-port parameters are easily obtained due to the simple algebraic nature of
the two-port network equations. For example, the z and y relationship could be obtained with matrices
z and y, such that if

Z  = 
z11 z12 
 z 21 z 22 
and the y matrix as,

Y  = 
y11 y12 
 y 21 z 22 

Then, V  = Z I 

Also, I  = Y V 
Replace I  with Y V  ,
V  = Z Y V 
If the above equation is to hold,
Z Y  = U 

Where, U  is a unit matrix. Thus, the matrices Z  and Y  must be inverses of each other, that is,

Z −1 = Y 

or, Y −1 = Z 
 15

Table 1

Now,

Z −1 = adjZ  = 1  z 22 − z12   y11 y12 

− z =
Z z11 z 22 − z12 z 21  21 z11   y 21 y 22 

Let  z = z11 z 22 − z12 z 21

z 22 z12
y11 = y12 = −
z z
Therefore,
z z11
y 21 = − 21 y 22 =
z z

Review Questions
1. Find the z, y, h and T parameters of the network shown in Fig. P1.2a, b, and c.
 16

2. Find the z, y, h and T parameters of the network shown in Fig. P1.3a and b.

3. Analysis of Ladder Networks

Consider the network shown in Fig. 12, where all the series branches are impedances and all parallel
branches are admittances. If the v denote node voltages and I denote branch currents, therefore, the following
relationships could be determined:

v j +1 = i j + 2 Z j + 2 + v j +3
i j = v j +1Y j +1 + i j + 2

Figure 12

As an illustration consider the ladder network in Fig.13. Note: The analysis is initiated at 2-2’ port
toward the 1-1’.

Figure 13
 17

The analysis
V2 (s ) = V2 (s )
I 2 (s ) = Y4 (s )V2 (s )
Va (s ) = I 2 (s )Z 3 (s ) + V2 (s ) = Y4 (s )V2 (s )Z 3 + V2 (s ) = V2 (s )Y4 (s )Z 3 (s ) + 1
I 1 (s ) = Y2 (s )Va (s ) + I 2 (s ) = Y2 (s )1 + Z 3 (s )Y4 (s ) + Y4 (s )V2 (s )
V1 (s ) = I 1 (s )Z1 (s ) + Va (s ) = Z1 Y2 (1 + Z 3Y4 ) + Y4  + (1 + Z 3Y4 )V2 (s )

Example 11:
Find the transfer impedance V2 I 1 for the circuit in Fig. 14.

Figure 14

Solution:
V2 (s ) = 1
I 2 (s ) = 1  1 = 1A
s+2
Va = sL2 I 2 + V2 = s ( 12 )(1) + 1 =
2
 s + 2 4 2s 2 + 4s 2s 2 + 4s + 3
I 1 = Va sC + I 2 =  (
3 s ) + 1 = + 1 =
 2  3 3
 2s 2 + 4s + 3  s + 2
V1 = sL1 I 1 + Va = ( s ) 3
2
 + = s 3 + 2s 2 + 2s + 1
 3  2
Hence, the network transfer function V2 I 1 is obtained as,
V2 1 3
z 21 = = 2 = 2
I1 2s + 4s + 3 2s + 4s + 3
3
Further, other various network functions derivable include:
V1  s 3 + 2s 2 + 2s + 1 
1. Z in = = 3 
 2s + 4s + 3 
2
I1
V2 1
2. = 3
V1 s + 2 s + 2 s + 1
2

I2 1 3
3. = 2 = 2
I 1 2s + 4s + 3 2s + 4s + 3
3
 18

Note: Each successive equation takes into account one new immittance. Moreover, each subsequent
equation is obtained by multiplying the equation just preceding it by the immittance that is next down
in line, in addition to the product of the previous equation.

Example 12
Find the short-circuit admittance function y11 and y12 of the network shown in Fig. 15.

Figure 15 (a) (b)

Solution:
To obtain the short-circuit functions port 2-2’ must be short-circuited. In addition, the series
branches must be represented as impedances and the shunt branches as admittances, such that:

The analysis
I2 = 1
Va = 2  1A = 2
I a = Va sC + I 2 = 2( 12 s ) + 1 = s + 1
 1   1 
 + Va = (s + 1)
5
V1 = I a   + 2 =
 2(s + 1)   2(s + 1)  2
 2
5 2s + 1
 12s 2 + 2s + 5
I 1 = V1Y + I a = 2    + (s + 1) =

 s  2s
Thus,
I 12s 2 + 2s + 5 2 12s 2 + 2s + 5 I2 1 2
y11 = 1 =  = ; y 21 = = =
V1 2s 5 5s V1 52 5

Example 13
Find the z parameters of the ladder network of Fig. 16.

Figure 16
 19

Solution:
The analysis
V2 (s ) = 1
I 2 (s ) = 1  1 = 1A
Vb = 0 + V2 = 0 + 1 = 1
 s   s  3s + 2
I a = Vb   + I 2 =   + 1 =
 2(s + 1)   2(s + 1)  2s + 2
 3s + 2  9 s 2 + 8s + 2
Va = sL1 I a + Vb = (3s )  +1 =
 2s + 2  2s + 2
9 s 3 + 8s 2 + 5s + 2
I 1 = VaY (s ) + I a (s ) =
2s + 2
9 s + 8s + 5s + 2 3
3 2
9s 2 + 8s + 2 36s 3 + 41s 2 + 25s + 8
V1 = I 1 Z (s ) + Va =  + =
2s + 2 s +1 2s + 2 2( s + 1) 2

Thus, the z-parameters of the ladder networks are:

V1 36s 3 + 41s 2 + 25s + 8 V2 2s + 2

z11 = = 4 ; z 21 = = 3
I1 9s + 17s 3 + 13s 2 + 7s + 2 I 1 9 s + 8s + 5s + 2
V1 36s 3 + 41s 2 + 25s + 8 V2
z12 = = z 22 = =1
I2 2(s + 1) 2 I2

Review Questions
1a. Find the voltage V1 V2 ratio of the circuit in Fig. P1.4a.

Figure P1.4a
1b. Find the short-circuit parameters of the network shown in Fig. P1.4b.

Figure P1.4b
 20

3.1 Elementary Synthesis Procedures

The basic philosophy behind the synthesis of driving-point functions is to break up a positive real
function P(s ) into a sum of positive real functions P1 (s ), P2 (s ), Pn (s ) , and then to synthesize these
individual Pi (s ) as elements of the overall network whose driving-point impedance is P(s ) .
That is,
P(s ) = P1 (s ) + P2 (s ) +  + Pn (s )

However, how is P(s ) to be decomposed into individual Pi (s ) ? Suppose P(s ) is given in general
as,
a s n + an−1 s n−1 +  + a1 s + a0 M (s )
P(s ) = n m =
bm s + bm−1 s m−1 +  + b1 s + b0 Q(s )

Note: A function is positive real, if the following conditions are met,

(a) P(s ) is real for real s ;
(b) The real part of Z (s ) is greater than or equal to zero when the real part of s is greater than or
equal to zero, such that Re P(s )  0 for Re s  0
(c) For a positive real function, the right half of the s plane maps onto the right half of the Z (s )
plane.
(d) Further, P(s ) must be rational. A few example of rational functions include: sL, R, and K s ( K
real and positive)

M (s ) D
Let = + R(s ) D0
Q(s ) s
= P1 (s ) + P2 (s )

From previous discussion, P1 (s ) satisfies the positive real function criteria. However, if P2 (s ) is to be
a positive real function, then:
1. P2 (s ) must have no poles in the right-half plane;
2. Poles of P2 (s ) on the imaginary axis must be simple and the residue must be real and positive;
3. Re P2 (s )  0 for all 

3.2 Foster Form Realization—Synthesis of L-C Driving-Point Immittances

The partial fraction expansion of an L-C function is expressed in general terms as,

K0 2K s
P(s ) = + 2 2 2 +  + Ks
s s + 2

The synthesis is accomplished directly from the partial fraction expansion by associated individual
terms to network elements. Hence, if P(s ) is an impedance Z (s ) , the term K 0 s represents a
 21

capacitor 1 K 0 farads, the term K  s would be an inductance of K  henrys, while the term
2 K i s s 2 +  i2 is a parallel tank circuit made up of a capacitor 1 2 K i farads and an inductor 2 K i  i2
henrys, shown in Figure 17.

Figure 17

Note: A useful property of L-C immittances is that the numerator differs by a degree of unity from the
denominator. The L-C networks synthesized by partial fraction expansions are sometimes referred to
as Foster-type networks. Moreover, the L-C impedance form in Fig. 17 is called a Foster I or Foster series
network.

Example 14
Synthesize the L-C driving-point impedance,

Z (s ) =
(s 2
)(
+1 s2 + 8 )
(
s s2 + 4 )
Solution:

The partial fractions of Z (s ) =

(s 2
)(
+1 s2 + 8 ) 5s 2 + 8
=s+ 2
2
=s+ + 2
3s
(
s s +4
2
) s s +4 (
s s +4 )
The synthesized network is obtained in Fig. 18.

Figure 18
 22

Suppose P(s ) is an admittance Y (s ) . Thus, the partial fraction expansion would yield a circuit with
parallel branches shown in Fig. 19.

Figure 19

Note: The admittance form of the network synthesis is referred to as Foster II or Foster parallel
network

Example 15
Synthesize the realizable admittance,

Y (s ) =
( )(
s s2 + 2 s2 + 4 )
( )(
s2 +1 s2 + 3 )
Solution
The partial fraction expansion of Y (s ) is given by,

Y (s ) =
( )(
s s2 + 2 s2 + 4
= +
)1
2s
+
3
2s

( 2
)(
s +1 s + 3
2
s
) s + 3 s +1
2 2

The obtainable network is shown in Fig. 20.

Figure 20

3.3 Cauer Ladder Network

The Cauer ladder network is obtained with the following steps:

1. Consider the case of an L-C impedance Z (s ) with numerator of degree 2n and denominator of
degree 2n-1. That is, Z (s ) has a pole at s =  , and this could be removed by a subtracted
impedance L1 s . Further, the remainder function still remains L-C, such that:

Z 2 (s ) = Z (s ) − sL1
 23

2. The resultant remainder function of Z 2 (s ) has now a numerator of degree of 2n-1, but a
denominator of degree 2n-2. However, an inversion of Z 2 (s ) to yield Y2 (s ) would have a pole
at s =  . A similar process of deducted admittance sC 2 would remove the pole with a remainder
of Y3 (s ) , such that:

Y3 (s ) = Y2 (s ) − sC 2

3. The process continues until the remainder is zero. In addition, the final structure of the network
synthesized is a ladder with series arms as inductors and shunt arms as capacitors.

The Cauer networks realization was published by Wilhem Cauer in 1926. He developed the continued
fraction method of impedance realization and synthesis of passive networks. Furthermore, the
continued fraction method is a series of inversions and divisions, with each successive quotient an
alternate between impedances Z (s ) and admittances Y (s ) .

Example 16
Synthesize the L-C impedance Z (s ) in Cauer network.

Z (s ) =
(s 2
)( )(
+ 1 s 2 + 9 s 2 + 25 )
( )(
s s 2 + 4 s 2 + 16 )
Solution
The continued fraction expansion of Z (s ) is given by,

s 5 + 20s 3 + 64s s 6 + 35s 4 + 259s 2 + 225 (s  Z

s 6 + 20s 4 + 64s 2
15s 4 + 195s 2 + 225 s 5 + 20s 3 + 64s (s 15  Y
s 5 + 13s 3 + 15s
7 s 3 + 49s 15s 4 + 195s 2 + 225 (15s 7  Z
15s 4 + 105s 2
90s 2 + 225 7 s 3 + 49s (7 s 90  Y
7 s 3 + 352 s
63
2 s 90s 2 + 225 ( 207 s  Z
90s 2
225 63
2 s ( 507 s  Y
63
2 s

The final synthesized network is shown in Fig. 21.

 24

Figure 21

Note: The above synthesis method is sometimes referred to as Cauer I network realization.
Suppose, however, both the numerator and the denominator are arranged in ascending order and divide
into each other from the lowest power, with inversions of the remainders—This alternative
realization is called the Cauer II network realization. Thus, the final structure of the Cauer II
network is a ladder with series arms as capacitors and shunt arms as inductors.

Example 17
Synthesize the L-C impedance Z (s ) in Cauer II network.

Z (s ) =
(s 2
)(
+1 s2 + 8 )
(
s s2 + 4 )
Solution
The continued fraction expansion of Z (s ) is given by,

Z (s ) =
(s 2
)( )
+ 1 s 2 + 8 s 4 + 9s 2 + 8
=
(
s s2 + 4 )s 3 + 4s
such that,

s 3 + 4 s 8 + 9 s 2 + s 4 (2 s  Z
8 + 2s 2
7 s 2 + s 4 4 s + s 3 (4 7 s  Y
4 s + 74 s 3
3
7
s 3 7s 2 + s 4 ( 349s  Z
7s 2
s4 3
7
s 3 (3 7 s  Y
3
7
s3
 25

The final synthesized network is shown in Fig. 22.

Figure 22

3.4 Synthesis of R-C Impedances or R-L Admittances

The Foster realization of an R-C impedance could be obtained by a replacement of all inductances
with resistances, such that, a general R-C impedance could be represented as,

Z (s ) =
K0 K1 K2
+ K + + +
s s + 1 s +  2

Where,
1 1 K
C0 = ; R = K  , C1 = , R1 = 1  . Therefore, to represent an R-C impedance, the constants
K0 K1 1
K i and  i must be real and positive. Fig. 23 showed the general R-C driving-point impedance
network.

Figure 23

To summarize, three properties of an R-C impedances are:

(a) Poles and zeros should be on the negative real axis, and are alternately cited;
(b) The singularity nearest to the origin must be a pole;
(c) The residues of the poles must be real and positive.

Further, consider the case when F (s ) is an admittance Y (s ) . Obviously, the admittance is easily
realized as an R-L network. That is, the expansion of the network elements could be obtained with
 26

inductances and resistors. Moreover, all the properties of R-L admittances are same as that of the R-C
impedance. Fig. 24 showed the general admittance network.

Figure 24

Example 18
Obtain the R-C network of the following impedance:
s 2 + 6s + 8
Z (s ) =
s 2 + 4s + 3
Solution:
The partial fraction expansion of the impedance is given by,
s 2 + 6s + 8 2s + 5 1 3
Z (s ) = = 1 + = 1 + 2
+ 2

s 2 + 4s + 3 s 2 + 4s + 3 s + 3 s +1

The R-C network in the series Foster Form is illustrated in Fig. 25.

Figure 25

An alternate representation of the admittance in R-L network is illustrated in Fig. 26.

Figure 26
 27

An alternate method of synthesis could also be based on the continued fraction expansion of
F (s ) shown below:

s 2 + 4 s + 3 s 2 + 6 s + 8 (1  Z
s 2 + 4s + 3
2 s + 5 s 2 + 4 s + 3 (s 2  Y
s 2 + 52s
3
2 s 2 s + 5 ( 43  Z
2s
5 3
2 s (3s 10  Y
3
2 s

Fig. 27 illustrates the corresponding R-C impedance and R-L admittance networks.

Figure 27

Note: Suppose the residues of the poles of the immittance is real and negative, such that the partial
fraction expansion yields,
Ki
−
s +i
Z (s )
To obtain the network realizations, the partial fraction of the impedance could resolve the
s
negative residue. Therefore,
Z (s ) K 0 Ki
= + K + +
s s s +i
Where K 0 , K1 ,, K   0. To obtain Z (s ) , in the desired form of synthesis, both sides should be
multiplied by s .
 28

Example 19
Synthesize in Foster and Cauer network the R-C admittance
2(s + 1)(s + 3)
Y (s ) =
(s + 2)(s + 4)
Solution
The partial fraction expansion is given by,
2(s + 1)(s + 3) 4s + 10
Y (s ) =
1 3
= 2− 2 = 2− −
(s + 2)(s + 4) s + 6s + 8 s+2 s+4

From observation, the residues are negative. However, the partial fraction expansion of Y (s ) s is

Y (s ) 2(s + 1)(s + 3) 34 1 3
= = + 2 + 4
s s (s + 2 )(s + 4 ) s s + 2 s + 4
Multiply both sides by s to obtain,
1 3
s s
Y (s ) = 34 + 2
+ 4
s+2 s+4
The immitance could realized with an R-L impedance or R-C admittance networks in Foster Forms shown in

Figure 28

Further, the continued fraction expansion Y (s ) is,

 29

s 2 + 6s + 8 6 + 8s + 2s 2 ( 34  Z
6 + 92s + 34 s 2
7
2
s + 54 s 2 8 + 6s + s 2 (16 7 s  Y
8+ 20
7
s
22
7
s + s2 7
2
s + 54 s 2 ( 4944  Z
s + 4944s
2
7
2

6s2
44
22
7
s + s 2 (484 21s  Y
22
7
s

s2 6s2
44
(6 44  Z
6s2
44

Fig. 28 shows the admittance Y (s ) synthesized as R-L impedances and R-C admittances in Cauer forms.

Figure 29

3.4.1 Synthesis of R-L-C Functions

Under certain conditions, R-L-C driving-point functions may be synthesized with either partial
fractions or continued fractions.

Example 20

Synthesize Z (s ) =
(s + 2)(s + 4) .
(s + 1)(s + 5)
 30

Solution
The partial fraction of Z (s ) is given by,

Z (s ) =
(s + 2)(s + 4) = 1 + 3 3
= 1+ 4 − 4
3

(s + 1)(s + 5) (s + 1)(s + 5) s +1 s + 5

Since one of the residues is negative, the above expansion cannot be used. Further, the expansion of
Z (s ) s is,
Z (s ) (s + 2)(s + 4 ) 85 3 3
= = − 4 + 20
s s(s + 1)(s + 5) s s + 1 s + 5

Still, the negative residues remain unresolved. However with a bit of manipulation, any negative
term could be resolved such that,

Z (s ) =
(s + 2)(s + 4) = 8 − 34 s + 203 s = 8 − 3 1 − 1  + 203 s
 
s(s + 1)(s + 5) 5 s + 1 s + 5 5 4  s + 1  s + 5

3 3
s
Z (s ) =
17
+ 4 + 20
20 s + 1 s + 5

Fig. 29 illustrated the expanded function.

Figure 29
Review Questions
1. Obtain the synthesis of the following networks with impedance functions:
s 4 + 5s 2 + 6
a. Z (s ) = ;
s3 + s
s 2 + 4s + 3
b. Z (s ) = 2 ;
s + 6s + 8
s 2 + 4s + 3
c. Z (s ) = ;
s 2 + 6s + 8

2. Synthesize by continued fractions the function

 31

s 3 + 2 s 2 + 3s + 1
Y (s ) =
s 3 + s 2 + 2s + 1
3. Synthesize the following function in Cauer form,
s 3 + 2s 2 + s + 1
Z (s ) =
s3 + s2 + s
4. Find the Foster and the Cauer networks of the impedance,

3(s + 1)(s + 4)
Z (s ) =
s+3

4. Minimum Phase and Non-Minimum Phase Networks

The steady steady-state response of a system function could be aptly represented by,
H ( j ) = M ( )e j ( )

where, M ( ) is the amplitude or magnitude response function, and  ( ) represented the phase
response. Moreover, the amplitude and phase response of the system provides valuable information
in the analysis and design of transmission circuits.

Consider the amplitude and phase characteristics of a low-pass filter shown in Fig. 30. The cut-off
frequency of the filter is at  C . This is the frequency at which the amplitude response is 0.707 of the

m aximum amplitude H ( jmax ) .This point is sometimes referred to as the “half-power frequency”, that

is, the frequency at which 20 log H ( jmax ) is down by 3dB to 20 log H ( jC ) . Further, the system

described would not “pass” frequencies that are greater than  C . In addition, if the phase response

 ( ) is linear, then minimum distortion would result (a range between − C    +C ). Often, the shape
of the amplitude and phase responses of a system are dictated by poles and zeros of such function.

Figure 30
 32

For example, a zero at z = −  ji (where  is small compared to  i ) would have a dip in the
amplitude response and a rapid increase change of the phase response at  = i . Similarly, a pole at
p = −  j i with  very small in comparison to  i , would have the amplitude response peaked and
a rapid decrease change of the phase response at  = i . Fig. 31 illustrated the effects of zeros and
poles very near the j axis.

Figure 31

Therefore, when a pole or zero is far away from the j axis, it contributes little to the shaping of the
amplitude and phase characteristics. Consider the amplitude and phase responses of the poles
s1 = −2  j 2 and s 2 = 1  j1 shown in Fig. 32a and b. It observed that the amplitude responses of

Figure 32

the two functions are the same. However, the magnitude phase response of Fig. 32b is greater than
Fig. 32a for all frequencies. This is attributed to the zeros in the RHP which contributed more to the
phase shifts compared to zeros of the LHP. To this end, the following terms are defined:
 33

(a) Minimum phase network: A system that has one or more zeros in the LHP, or on the j -axis only.
(b) Nonminimum network: A system that has one or more zeros in the RHP, or on the j -axis only.
(c) All-pass network: A system function whose zeros in the RHP are mirror images of poles in LHP
about the j -axis. Networks that have all-pass response characteristics are used to correct phase
distortions in a transmission system.

Fig. 33 illustrated the phase responses of the minimum and nonminimum phase functions.

Figure 33

Summary of Network Synthesis Realizations:

34