TWO-PORT NETWORKS

Book: Linear Electric Circuits, By Wallace L. Cassell

Study Objectives
• To understand about two – port networks and its functions.

• To understand the difference among z – parameter, y – parameter, t – parameter,

hybrid-parameter, inverse hybrid-parameter and terminated two port networks.

• To investigate and analysis the behavior of two – port networks.

Definitions
• A pair of terminals through which a current may enter or leave a network is known as
a port.
• Two terminal devices or elements (such as resistors, capacitors, and inductors)
results in one – port network.
• Most of the circuits we have dealt with so far are two – terminal or one – port circuits.
• A two – port network is an electrical network with two separate ports for input and
output.
• It has two terminal pairs acting as access points. The current entering one terminal
of a pair leaves the other terminal in the pair.
I

+
V Linear network
-

I
One – port network

I1 I2
+
+
Linear network V2
V1
-
-

I1 I2
Two – port network

Fig. 1. One-Port and Two-Port networks diagram.

• Two (2) reason why to study two port – network:

 Such networks are useful in communication, control system, power systems

and electronics.

 Knowing the parameters of a two – port network enables us to treat it as a

“black box” when embedded within a larger network.

• From the network, we can observe that there are 4 variables that is I1, I2, V1and V2,
which two are independent.

• The various terms that relate these voltages and currents are called parameters.

Parameters to represent a two-port network

 Z – PARAMETER

 Y – PARAMETER

 T (ABCD, EFGH) – PARAMETER

 Hybrid-PARAMETER

 Inverse-Hybrid PARAMETER

 TERMINATED TWO PORT NETWORKS PARAMETERS

Z – PARAMETER
• Z – parameter also called as impedance parameter and the units is ohm (Ω)

• Impedance parameters is commonly used in the synthesis of filters and also useful in
the design and analysis of impedance matching networks and power distribution
networks.

• The two – port network may be voltage – driven or current – driven.

• Two – port network driven by voltage source.

• Two – port network driven by current sources.

• The “black box” is replace with Z-parameter is as shown below.

• The terminal voltage can be related to the terminal current as:

V1  z11 I1  z12 I 2 (1)

V2  z 21 I1  z 22 I 2 (2)

• In matrix form as:

V1   z11 z12   I1 

V    z z 22   I 2 
 2   21
• The Z-parameter that we want to determine are z11, z12, z21, z22.

• The value of the parameters can be evaluated by setting:

1. I1= 0 (input port open – circuited)
2. I2= 0 (output port open – circuited)
 • Thus,

V1 V1
z11  z12 
I1 I 0 I2 I 0
2 1

V2 V2
z21  z22 
I1 I 0 I2 I 0
2 1

Where

z11 = Open – circuit input impedance.

z12 = Open – circuit transfer impedance from port 1 to port 2.

z21 = Open – circuit transfer impedance from port 2 to port 1.

z22 = Open – circuit output impedance.

Example 1: Find the Z – parameter of the circuit below.

I1 I2
+ +

120Ω 240Ω
V1 V2
_ _

40Ω
Solution:
i) I2 = 0 (open circuit port 2). Redraw the circuit.
 I1 Ia
+ +
240Ω
V1 V2
Ib 120Ω

_ _

40Ω

V1  120 I b .......(1) V2  240 I a .......(3)

280 120
Ib  I1......(2) Ia  I1.......(4)
400 400
sub (1)  (2) sub (4)  (3)
V V2
 Z11  1  84  Z 21   72
I1 I1

ii) I1 = 0 (open circuit port 1). Redraw the circuit.

Iy I2
+ +

V1 120Ω 240Ω V2
Ix
_ _

40Ω
 V2  240 I x .......(1) V1  120 I y .......(3)
160 240
Ix  I 2 .......(2) Iy  I 2 .......( 4)
400 400
sub (1)  (2) sub (4)  (3)
V V
 Z 22  2  96  Z12  1  72
I2 I2
In matrix form:

84 72
Z    
 72 96 
Example 2: Find the Z – parameter of the circuit shown below.

I1 2Ω j4Ω 10Ω I2
+ +
10I2 + -j20Ω V
V1 _ 2
_ _

Solution:
i) I2 = 0 (open circuit port 2). Redraw the circuit.

I1 2Ω j4Ω I2 = 0 V1  I1 (2  j4)
+ + V1
 Z11   (2  j4)
I1
V1 V2 V2  0 (short circuit)
_ _
 Z 21  0
ii) I1 = 0 (open circuit port 1). Redraw the circuit.

I1 = 0 10Ω I2 V1  10I 2
+ + V1
 Z12   10
+ I2
V1 _ 10I2 -j20Ω V
2 V2 V - 10I 2
I2   2
_ _  j20 10
 j 1
2I 2  V2   
 20 10 
V
 Z 22  2  (16 - j8) 
I2
In matrix form;
(2  j4) 0 
Z   
 10 (16 - j8) 
Y – PARAMETER
• Y–parameter also called admittance parameter and the units is siemens (S).

• The “black box” that we want to replace with the Y-parameter is shown below.

I1 I2
+ Y11 Y12 +
V1 V2
- Y21 Y22 -

• The terminal current can be expressed in term of terminal voltage as:

 I1  y11V1  y12V2 (3)

I 2  y21V1  y22V2 (4)

• In matrix form:

 I1   y11 y12  V1 

I    y y22  V2 
 2   21
• The y-parameter that we want to determine are Y11, Y12, Y21, Y22. The values of the
parameters can be evaluate by setting:

i) V1 = 0 (input port short – circuited).

ii) V2 = 0 (output port short – circuited).

• Thus;

I1 I1
Y11  Y12 
V1 V2 0
V2 V1  0

I2 I2
Y 21  Y 22 
V1 V2 0
V2 V1  0

y11 = Short-circuit input admittance

y12 = Short-circuit transfer admittance from port-2 to port-1

y21 = Short-circuit transfer admittance from port-1 to port-2

y22 = Short-circuit output admittance

Example 1: Find the Y – parameter of the circuit shown below.

I1 5Ω I2
+ +
20Ω 15Ω
V1 V2
_ _
Solution:
i) V2 = 0
V 1  20 I a .......( 1)
I1 5Ω I2
5
+ Ia  I 1 .......( 2 )
25
V1 Ia 20Ω sub (1)  (2)
_ I 1
 Y11  1  S
V1 4
V1  5I 2
I2 1
Y21   S
V1 5
ii) V1 = 0
V2  15 I x .......(3)
I1 5Ω I2 5
+ Ix  I 2 .......(4)
25
15Ω Ix V2 sub (3)  (4)
_ I 4
Y22  2  S
V2 15
V2  5 I1
I1 1
Y12   S
V2 5
In matrix form;
 1 1
 
Y    41 5S
4 
 
 5 15 

Example 2 (circuit with dependent source):

Find the Y – parameters of the circuit shown.

I1 2Ω j4Ω 10Ω I2
+ +
10I2 + -j20Ω V
V1 _ 2
_ _

Solution:
i) V2 = 0 (short – circuit port 2). Redraw the circuit.
I1 2Ω j4Ω 10Ω I2 I0
+ V1  (2  j4)I1
+ I1 1
V1 _
10I2  Y11    (0.1 - j0.2) S
_ V1 2  j4
I2
 Y21   0S
V1
ii) V1 = 0 (short – circuit port 1). Redraw the circuit.
I1 2Ω j4Ω 10Ω - 10I 2
I2 I1  ........(1)
2  j4
+
V V - 10I 2
+ I2  2  2
10I2 -j20Ω - j20 10
_ V2
_ 1 1 
2I 2  V2   .......( 2)
 10 - j20 
 I2
 Y22   (0.05  j0.025) S
V2
sub (2)  (1)
I
Y12  1  (-0.1  j0.075) S
V2
In matrix form;
0.1  j0.2  0.1  j0.075
 Y   
0.05  j0.025 
S
 0

T (ABCD) PARAMETER
• T – parameter or ABCD – parameter is another set of parameters relates the variables
at the input port to those at the output port.
• T – parameter also called transmission parameters because this parameter are useful in
the analysis of transmission lines because they express sending – end variables (V1 and
I1) in terms of the receiving – end variables (V2 and -I2).
• The “black box” that we want to replace with T – parameter is as shown below.

I1 I2
+ A11 B12 +
V1 V2
- C21 D22 -

• The equation is:

……………….(5)

.………………(6)

• In matrix form is:

V1   A B   V2 
 I   C D   I 
 1   2 
• The T – parameter that we want determine are A, B, C and D where A and D are
dimensionless, B is in ohm (Ω) and C is in siemens (S).
 • The values can be evaluated by setting
i) I2 = 0 (output port open – circuit)
ii) V2 = 0 (output port short circuit)
• Thus;

V1 V1
A B
V2 I2 0
I2 V2  0

I1 I1
C  D
V2 I2 0
I2 V2  0
• In term of the transmission parameter, a network is reciprocal if;

AD - BC  1 ……….………………(7)
A = Open-circuit voltage ratio
B = Negative short-circuit transfer impedance
C = Open-circuit transfer admittance
D = Negative short-circuit current ratio

Example: Find the ABCD – parameter of the circuit shown below.

I1 2Ω 4Ω I2

+ +

V1 10Ω V2

_ _

Solution: I1 2Ω

i) I2 = 0,
+ +

V1 10Ω V2

_ _
 V2  10 I1
I1
C   0.1S
V2
V1  2 I1  V2
V  6
V1  2 2   V2  V2
 10  5
V
 A  1  1 .2
V2
ii) V2 = 0,

10
I1 2Ω 4Ω I2 I2   I1
14
+
I
V1 10Ω
 D   1  1 .4
I1 + I2 I2
_
V1  2 I1  10I1  I 2 
V1  12 I1  10 I 2
 14 
V1  12  I 2   10 I 2
 10 
V
 B   1  6.8
I2

1.2 6.8
T    
 0 .1 1 .4 
 Inverse T (EFGH) PARAMETER

• Inverse T – parameter or EFGH – parameter is a another set of parameters relates the

variables at the output port to those at the input port.
• Inverse T – parameter, Inverse transmission parameters are useful in the analysis of
transmission lines because they express receiving – end variables (V2 and I2) in terms
of the sending – end variables (V1 and –I1).
• The “black box” that we want to replace with T – parameter is as shown below.

I1 I2
+ E11 F12 +
V1 V2
- G21 H22 -

 The equation is:

………………….(8)

.……...………….(9)

• In matrix form is:

• The T – parameter that we want determine are A, B, C and D where E and H are
dimensionless, F is in ohm (Ω) and G is in siemens (S).
• The values can be evaluated by setting
i) I1 = 0 (input port open – circuit)
ii) V1 = 0 (inputport short circuit)
 • Thus;

• In term of the transmission parameter, a network is reciprocal if;

…………………(10)
EH-FG=1
E = Open-circuit voltage gain
F = Negative short-circuit transfer impedance
G = Open-circuit transfer admittance
H = Negative short-circuit current gain

Example 19.9 (Ref. Book by Alexander):

The ABCD parameters of the two-port network in Fig. 1 are

The output port is connected to a variable load for maximum power transfer. Find RL and
the maximum power transferred.

Fig. 1. For Example 19.9.

Solution:
What we need is to find the Thevenin equivalent (ZTH and VTH) at the load or output port.
We find using the circuit in Fig. 2(a).

(a)
Fig. 2. Solution of Example 19.9: (a) finding ZTh, (b) finding VTh, (c) finding RL for
maximum power transfer.
Our goal is to get ZTh = V2/I2. Substituting the given parameters into Eq. (5) and (6), we
obtain

……..………….(i)

……..………….(ii)

At the input port, V1 = -10I1. Substituting this into Eq. (i) gives

or
……………......(iii)
Setting the right-hand sides of Eqs. (ii) and (iii) equal, we get

Hence,

To find VTh, we use the circuit in Fig. 2(b). At the output port I2 = 0 and at the input port V1
= 50 – 10I1. Substituting these into Eqs. (i) and (ii),we get
 Fig. 2. (b) finding VTh.

……………………….(iv)

.……………………….(v)
Substituting Eq. (v) into Eq. (iv), we get as follows.

Thus

The equivalent circuit is shown in Fig. 2(c). For maximum power transfer,

Fig. 2. (c) finding RL for maximum power transfer.

The maximum power is as follows.

 Hybrid Parameters:
The equations for the hybrid parameters are:

V1   h11 h12   I 1 

 I   h h22  V2 
 2   21

The following is a popular model used to represent a particular variety of transistors.

I1 K1 I2

+ + +

V1 K2V2 K4 V2
_ K3V1
_ _

We can write the following equations:

……………..(11)

……………..(12)
We want to evaluate the h parameters from the above set of equations

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

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

h11 = Short-circuit input impedance

h12 = Open-circuit reverse voltage gain
h21 = Short-circuit forward current gain
h22 = Open-circuit output admittance

Another example with hybrid parameters.

Given the circuit below.

I1 -I2
The equations for the circuit are:

+ +
R1 V1 = (R1 + R2)I1 + R2I2
V1 R2 V2
_
V2 = R2I1 + R2I2
_

The H parameters are as follows.

V1 V1
h11  h12  1
I 1 V =0 = R1 V2 I1=0
=
2

I2 I2 1
h21  = -1 h22  =
I1 V2=0 V2 I1=0 R2
The equations for the hybrid parameters are:

……………..(13)

……………..(14)

V1 = 0
I2 = 0

I2 = 0 V1 = 0

g11 = Open-circuit input admittance

g12 = Short-circuit reverse current gain
g21 = Open-circuit forward voltage gain
g22 = Short-circuit output admittance
Relationship among the Two-Port Parameters:

Going From Y to Z Parameters

For the Y parameters we have: For the Z parameters we have:

I Y V V Z I

From above; V  Y 1 I  Z I

Therefore

 y y 
 22 12

z z  
 Y 
Z Y 1
  11 12   Y 
 y 
 z 21 z 22   21
y
11 
   
 Y Y 

where

Y  det Y
 I1 I2

Z11 Z12
+ +
V1 V2
- -
Z21 Z22

We have the following equation for Z-parameters

………………(1)

………………(2)

From Eq. (1) and (2) we get

…………………………….(3)

………………..……………(4)

Then using Cramer's rule from Eq. (3) and (4) we get the expression of V1 and I2 as follows.
 ………………………….. (5)

………………………….. (6)

But for H-parameters we have the following relationship.

……………………….(7)

…………………………(8)
Comparing Eq.(5) with (7) we get the following relationships.

Where

and comparing Eq. (6) with (8) get the following relationships
The passive two-port is defined as one which has no controlled source or sources. The
passive network has the following significant relationship between the mutual
parameters.

….…………………………..(15)

For passive two-port network Eq. (8) and Eq. (9) becomes

………………(8')

………………(9')

Then we have

E = D, F = B, G = C, and H = A

The A, B, C, D parameters in the passive case are sometimes called the “general circuit
constants” and play and important part in the transmission of electrical energy. These
parameters are important in the cascade connection of two-ports.

Table 1: Significant Information Concerning the Passive Two-Port

z12 = z21 (15) g12 = -g21 (16)

y12 = y21 (17) h12 = -h21 (18)

AD-BC = 1 (19)
E=D (20) F=B (21)
G=C (22) H=A (23)
A common type of passive two-port network with z11 = z22, y11 = y22, and A = D, is termed
“symmetrical”.
-j4Ω -j4Ω

j2Ω

Fig. 2. Diagram of a typical symmetrical two-port network.

Interconnection of Two-Port Networks

Two-Port may be connected in :
(a) series, (b) parallel, (c) series-parallel, (d) parallel-series, or (e) cascade
Note that the two-port parameters were developed on the basis that the currents at the
terminals of a port are identical; that is, if current I1 flows into terminal 1, then I1 flows
out of terminal 1’.

(a) Series

Fig. 3. Series connection of two two-port network.

For network Na,

And for network Nb,

We notice from Fig. 3 that

and that

Thus, the z parameters for the overall network are

Or
For several two-port networks connected in series, in general we have the following
relationship.

….……(26)
Here, I1a = I1b = Iin,
I2a = I2b = Iout,
V1 = Vin,
V2=Vout

(b) Parallel

Fig. 4. Parallel connection of two two-port network.

The parallel connection of two two-port networks is shown in Fig. 4. For the two networks,

and

But from Fig. 4,

Using the expression if I1a, I1b, I2a, and I2b we get:

Thus, the y parameters for the overall network are

or

For several two-port networks connected in parallel, in general we have the following
relationship.

……………………….(27)
Here V1a = V1b = Vin,
V2a = V2b = Vout,
I1 = Iin,
I2=Iout
(c) ) Series-parallel

I1 I1a I2a
+ +
+
V1a Na V2a
‒ ‒ I2
+

V1 I1 V2
I1b I2b ‒
+ +
Nb
V1b V2b
‒ ‒ ‒
Fig. 5. Series-Parallel connection of two two-port network.
From network Na we get as follows.
…………(i)

…………(ii)

From network Nb we get as follows.

………..(iii)

..………(iv)

From Fig. 5 we have the following relationship.

Putting the expressions of Eq. (i), (ii), (iii), (iv) in these equations we get as follows.

………………..(v)

………………..(vi)
But from Fig. 5 we have the following relationships.

Putting these relationships in (v) and (vi) we get as follows.

In matrix form the above equations can be written as follows.

Or

So,

Here V1 = Iin, I1=Iin

V2 = Iout, I2=Iout
[ha] and [hb] are the hybrid parameters matrix for network Na and Nb.
When several two-port networks are connected series-parallel we have the following
relationship.

…………………….(28)
ASSIGNMENT-1
Find the impedance at the input port (Zin) of the circuit in figure below.

ASSIGNMENT-2
Derive the two-parameters matrix for several two-port networks connected in Parallel-
Series.
 (e) Cascade
The connection of two two-port networks in cascade is shown in Fig. 8. For the two networks,

Fig. 7. Cascade connection of two two-port network.

From Fig.7,

Substituting these expression in the above two equations we get:

Thus, the transmission parameters for the overall network are the product of the transmission
parameters for the individual transmission parameters:
or

………………………(29)
Here V1 = Vin,
I1 = Iin,
V2 = Vout,
I2=Iout

ASSIGNMENT-3
Show that the two-port network parameters matrix for cascade connection of network Na and
Nb are as follows.

Example 19.12 ( From the book: by Alexander)

Evaluate V2/Vs in the circuit in figure below.
Solution:
This may be regarded as two two-ports in series. For Nb, we can write

Thus

But

…………..(1)

...…………(2)

Also, at the input port

…………………………………..…………………(3)
and at the output port

..………….(4)

Substituting Eqs. (3) and (4) in Eq. (1) we get

…….…………………..(5)

While substituting Eq. (4) in Eq. (2) gives

 …………..(6)
Substituting Eq. (6) into Eq. (5) we get

And so,
V2/Vs = 1/2.85 = 0.3509 (Answer)
 The Validity Tests for two-port networks connection
Series Validity:

Voltage Vx must be zero for “series validity” on the output side as shown in the figure above.

Parallel Validity:

Voltage Vx must be zero for “parallel validity” on the output side as shown in the figure above.
EXAMPLE: Validity tests for series-series connection
If the voltmeter meter reading is zero in both cases, then the networks can be connected
series-series.

EXAMPLE: Validity tests for parallel-parallel connection

 Reciprocal Two-Port Circuit
Reciprocal : if interchange of ideal voltage source at one port with ideal ammeter at other
port produces same ammeter reading.

Symmetric Two-Port Circuits

Reciprocal two-port circuit is symmetric if its ports can be interchanged without
disturbing values of the terminal currents and voltages.

A reciprocal two-port circuit is symmetric if in addition to the equation of reciprocal

network, the following are true
 TERMINATED TWO-PORT NETWORKS

Fig. 6. A two-port terminated in Z2 ohms at terminals 2.

The symbol Zin1 represents the impedance looking into terminals 1. We have the following
equation for the terminated network.

Then

So,

………………..(30)
Similarly if we terminate port-1 by Z1 we have the following equation.

………………..(31)

Iterative Impedance: If such a impedance is terminated to any port so that the input

impedance looking from other port is exactly same what is terminated is known as iterated
impedance (Zk2 say) looking into the input port (Fig. 7).

Fig. 7. Two-port iterative parameter looking into terminal 1.

………………..(32)

Solution of equation (32) gives us the expression of the iterated impedance looking into port-1 as
follows
 .

………(33)

In the similar way if Zk1 impedance is terminated at port-1 we have the following equation of
iterated impedance looking into port-2.

………..……(34)

Image Impedance:

Fig. 8. Two-port image parameters.

According to the above figure (Fig. 8) Modification of Eq. (30) and (31) we get

………………………………...………………(35)
 …………………………………………..(36)

Substituting ZI2 from Eq. 36 into Eq. 35 we get the following relationship.

………….………………………..(37)

Similarly,

…………………………………(38)

From Eq. (37) and (38) we have the following relationships.

……..……………………………(39)

…………………………………………(40)
Image impedances from Open-Circuited and Short-Circuited Impedances
We will define the following:

(a) Z1o = impedance looking into terminals 1 with terminals 2 open-circuited, that is, with
Z2 = ∞;

(b) Z1s = impedance looking into terminals 1 with terminals 2 short-circuited, that is, with
Z2 = 0.

Under the above two conditions, modifying Eq. (30) gives us the following relationships.

……………………………(41)

…..…………………………(42)

Then from Eq. (41) and (42) we get the following equation.

Then,

…………….(43)

and

………………..(44)
 The Symmetrical Two-Port
Characteristic impedance
The symmetrical two-port is characterized by the following relation

A = D, since Z11 = Z22 in symmetrical network. Then, for the symmetrical case

Likewise, for the symmetrical two-port,

In the symmetrical case, the four general terminal impedance values merge into one, common
called the “characteristic impedance” and commonly represented by the symbol Z0. That is, for
the symmetrical two-port,

………..(45)
Example: Consider the non-symmetrical two-port of figure shown below. Find out the two
iterative impedances and the two image impedances.

Solution: We know the transmission parameters are as follows.

Here,

and
Then

Consequently,

Then,
Using short-circuited and open-circuited impedance we find out image impedances as follows.

and

Then

In the similar way, we get

 Synthesis of a Two-Port with Specified Values of Image
Impedance
The Fig. 9 below shows a two-port in the form of a T. The problem is to find the values of Z1,
Z2, and Z3 which give the specified image impedances.

Fig. 9. Synthesis of Two-port network.

Here,
Then,

………………..(46)

………………..(47)

Substituting the value of Z3 + Z2 from Eq. (47) into (46) to obtain

…………………...…………(48)

Then,

…………(49)
 ……………………….(50)

Unfortunately, Eq. (49) and (50) contains a set of two equations in three unknowns. We may
chose

..…………………….(51)

Then from (49) and (50) we obtain

……………………..(52)

If choose the value of Z2 as in the above way, we can prove that the image impedances satisfy
their requirements.
In fact, a simplified design often results, if either Z1 or Z3 is chosen as zero. For example, a two-
port will be designed to give

and

with

Then,

and

from Eq. (50)

and

Of course, Z2 will be taken as +6 = 6<0°.

 Then, from Eq. (49),

Again, the positive sign is chosen and

The circuit configuration is as follows.

 An L-Type Two-Port with Resistive Image Impedances
Consider the two-port of figure shown in below.

If terminated at terminals 1 in RI1 ohms, the impedance looking into terminals 2 must be RI2
ohms. That is,

…………………..(53)

and

Separating the real and imaginaries gives

………………..(54)
 ……………..…….(55)

From Eq. (55),

…………..(56)

If we put Eq. (56) into Eq. (54) we get as follows.

…………………….(57)

Since X22 must be positive in sign,

Then from Eq. (57) we get

….…….(58)

From Eq. 54,

………………….(59)
 An example of L-type network to match impedance
The following example is the application of two-port (L-type) networks connecting a
transmission line (feeder), an elevated antenna and a vacuum tube (transmitter) at ground level.

The object is take optimum power from the transmitter and to deliver optimum
power to the antenna.

Optimum power indicates a conjugate match of impedances.

For the two-port network between antenna and the feeder,

If X2 is chosen as -221 Ω,

For the two-port between the feeder and the transmitter,

If X2 is chosen as -1310 Ω,

Operation at a frequency of 4 × 106 Hz (a frequency assigned for amateur radio operation in

the united states) necessitates the values of inductance and capacitance shown in the
complete system of Fig. 10.

The required two L-networks are as follows:

Fig. 10. Full network of generator, feeder, and antenna with L-two port networks at each ends of
the feeder for impedance matching. Inductances are in μH and Capacitances are in pF.

The Image Propagation Function

Consider that the two-port of Fig. 11 is loaded at terminals 2 in ZI2 ohms. The voltage
transfer function Tv = V2/V1 is desired.

It is convenient to use

And from Fig. 11 we have

 Fig. 11. A two-port loaded at terminals 2 in ZI2 ohms.

Then the expression of Voltage Transfer Function, Tv is as follows.

……………………………………….…….……(60)

But

So,
 ……………………….………………(61)

If we choose

Then from Eq. (61) we get

Recall that

Then we get
So,

………………………….……………………………(62)

The Current Transfer Function, Ta is

….…………………………………….………………………….(63)

From Fig. 11 we get

and

So Eq. (63) is turned into

……………………………………...……………(64)

The Volt-Ampere Transfer Function, Tva is as follows.

………………...……………(65)
Now suppose Fig. 11 is terminated at terminal 1 with ZI1 we are looking from terminals 2.
Then in the similar above way we obtain the following relationships.

……………….……………(66)

…….……………...……………(67)

………………………….……………(68)

The character of the Image Transfer Function can best be seen in the symmetrical two-port in
which ZI1 = ZI2 as follows.

………………….……….……………(69)

Taking the natural logarithm of each side of Eq. (69) we get as follows.

…………………... ..………(70)
It is customary to write

…………………………….(71)

Then

…………………………….(72)

and
………. ……………………(73)

Example: Find the image parameters (ZI1, ZI2, θ) of the two-port network shown below.

Solution: Using open-circuited and short-circuited case from both the terminals we get
 12

We know the following relationship.

Taking natural logarithm at both the side we get

Then the image parameters are as follows.

Cascade connection of three Two-Port Networks

Fig. 12. The cascade connection of three two-port networks with image match at each junction and
the entire network is terminated by ZI2 ohms.

According to Fig. 12 the following relationship expressions may be written:

, , and

…………(74)

…………(75)

Reflection of voltage and current

Consider as in Fig. 13 a voltage source of impedance Zg ohms feeding a load of Zl ohms. Because of the
mismatch of impedance “reflection” of voltage and of current is said to occur at the load. Since the
mismatch is measured by the impedance term Zg ‒ Zl ohms, it is convenient to bring this mismatch term into
the situation shown in Fig. 14.
Fig. 13. A source feeding a load. Fig. 14. The circuit of Fig. 13 with the impedance
mismatch measured by the term (Zg - Zl ).

Application of the compensation theorem in Fig. 14 permits the impedance mismatch to be incorporated
into the votage source (Zl ‒ Zg)I of Fig. 15. The polarity of this source should be carefully checked.

Fig. 15. The circuit of Fig. 13 with the impedance mismatch incorporated into voltage source (Zl ‒ Zg)I.
Superposition may now be applied to the circuit of Fig. 15, as in Fig. 16 and Fig. 17.

The “incident” components of load voltage and load current are defined as those due to the driving
voltage V in the matched impedance system of Fig. 16. These components are, respectively,

……………………………(76)

and
………………….…………(77)

Fig. 16. The incident components of load voltages and load current, due to the driving voltage V.

Fig. 17. The reflected components of load voltages and load current, due to the mismatch voltage (Zl ‒ Zg)I.
By definition, the “reflected” components of load voltage and load current are those components due to
the mismatch voltage (Zl ‒ Zg)I of Fig. 17 in the same matched system. That is

……………..…………(78)

and

……………………(79)

The “reflection coefficient for voltage” is defined as

……………………………………………(80)

......………………….……………(81)

Similarly, the “reflection coefficient for current” is defined as

…………………….……………(82)

The two reflection coefficients being identical, one may speak of the reflection coefficient

………………….……………(83)

Evidently, the reflection coefficient is zero only if, in Fig. 13, impedance mismatch is absent because Zl =
Zg.
 Reflection of Power

The power dissipated in the load may be said to be composed of the incident component Pi and the
reflected component Pr, as stated in the following equation.
……………(84)

But

and

Consequently,

……………(85)

……………(86)

The first component of Eq. (86) contains the incident values. Consequently

……………(87)

The second component of Eq. (86) contains the reflection terms. Consequently

……………(88)

Note that the reflected power does not equal the real part of the product of the reflected voltage and
the conjugate of the reflected current.
Insertion Loss of a Two-Port Network

Consider in Fig. 18 the direct connection of a voltage source V with internal impedance of Z1 ohms to a
load of Z2 ohms.

Fig. 18. A voltage source directly connected to a load Z2 .

With I2out representing the current flowing into terminal 2 without the presence of a coupling network,
one may write

……………………(89)

If a two-port network is introduced as coupling medium, the situation appears as in Fig. 19.

Fig. 19. A two-port network inserted between the source and the load of Fig. 18.
By following the technique as discussed before, the impedance mismatch at each port may be accounted for
as in Fig. 20.

Fig. 20. The circuit of Fig. 19 with voltage sources representing the impedance mismatch at the ports.

The circuit is now ready for superposition. There are three components of both I1 and I2, as follows:

(a) I1’ and I2’ resulting from V acting alone;

(b) I1” and I2” resulting from the mismatch voltage (ZI1 ‒ Z1)I1 acting alone; and

(c) I1’” and I2’”resulting from the mismatch voltage (ZI2 ‒ Z2)I2 acting alone.

Observe that calculation of these three components of I1 and I2 is relatively easy because of the image
impedance match in each case. The result is

, ,

,
 ,

But

and

Consequently,

……………(90)

and

……………(91)
Solving for I2 using Eq. (90) and (91) we get

..……….…(92)
The influence of certain important terms upon I2 may be pointed out after rearranging Eq. (92) as follows:

……………(93)

The Eq. (93), the first term represent the current I2out, which would have existed in the load if the
coupling network had been absent, as per Fig. 18 and Eq. 89. The next three terms are fundamentally
similar; they become unity for equality between Z1 and Z2, ZI1 and Z1, and ZI2 and Z2 respectively. These
three terms may be expressed as

……………………………………….........………….(94)

(inequality factor between Z1 and Z2) ; …………….(95)

(reflection factor for ZI1 and Z1) ; ………………….(96)

and

(reflection factor for ZI2 and Z2) ; ………………….(97)

The sixth term of Eq. (93) contains the reflection coefficient for port 1, expressed as

………………………….. .………………..(98)

and the reflection coefficient for port 2, given by

……………………………..………………..(99)

Thus, the sixth term, called the interaction factor, may be written as

…..……………..……………(100)

The interaction factor F4 is unity if there is an image match at either port, or if, in the absence of
an image match, θ is very large.

It is convenient to express “the current insertion ratio” as

…………………….(101)

and the “insertion loss” as

……….........................(102)
 Introduction to Filter
Why Use Filters?

Filters allow some signal frequencies applied at their input terminals to pass through to their output
terminals with little or no reduction in signal level.

Computers use filters: to reduce EMI (electro-magnetic interference) emissions from their power lead; to
smooth the output of the switched-mode power supply: to limit the video bandwidth of signals going
to the display.

Fig. 1 shows the relationship between time and frequency domains for a simple sinusoidal signal.

Fig. 1 A simple sinusoidal signal in time and frequency domains.

 What is Filter?

A filter is a circuit that is designed to pass a specified band of frequencies while attenuating all signals
outside this band.

Filter networks may be either active or passive.

Passive filter networks contain only resistors, inductors, and capacitors.

Active filters, which are the only type covered in this text, employ operational amplifiers (op-amps) as
well as resistors and capacitors.

The Transfer Function:

Both analog and digital filters can be considered a "black box." Signals are input on one side of the
black box and output on the other side. The amplitude of the output signal voltage (or its equivalent digital
representation) depends on the filter design and the frequency of the applied input signal. The output
voltage can be found mathematically by multiplying the input voltage by the transfer function, which is a
frequency-dependent equation relating the input and output voltages. The transfer function is
illustrated in Fig. 2.

Fig. 2. The input, Vin and output, Vout relationship with the transfer function, F(ω) .

The relationship between input and output will be a function of frequency ω (omega), given in terms of
radians per second.

The transfer function, F(ω), is frequency dependent. For example, suppose that at ω = 0.5, F(ω) is equal
to 1 and hence Vin = Vout.
 Now suppose that at ω = 2, F(ω) is equal to 0.01, hence Vout is Vin ÷ 100. In
decibels, the gain is -40dB, since it is 20log(Vout /Vin); since the gain is negative, this can be referred to as
a (positive) attenuation, or signal loss, of 40dB.

For the most common filter types, the transfer function is often presented in graphical form. The graph
has a number of curves showing signal gain (loss) versus frequency.

An imaginary "brick wall" lowpass filter, illustrated in Fig. 3, is ideal in that it has an infinitely steep change
in its frequency response at a certain cutoff frequency. It passes all signals below the cutoff frequency
with a gain of 1. That is, signals below the cutoff frequency have their amplitude multiplied by 1 (it., they are
unchanged) as they pass through the filter.

Fig.3. The Ideal "Brick Wall" Filter.

BASIC FILTER TYPES

1. The first type is the low-pass filter (LPF). As might be expected, an LPF passes low-frequency
signals, and rejects signals at frequencies above the filter’s cutoff frequency (Fig. 4).

Fig.4. Low-pass frequency response.

2. The inverse of the low-pass is the high-pass filter, which rejects signals below its frequency (Fig. 5).

Fig.5. High-pass frequency response.

3. Band-pass filters pass frequencies within a specified band and reject components outside the band
(Fig. 6).

Fig.6. Band-pass frequency response.

 4. Band-reject filters reject frequencies within a specified band and pass components outside this band
(Fig. 7).

Fig.7. Band-reject frequency response.

Filter Terminology

Passband:

The range of signal frequencies that are allowed to pass through a filter, with little or no change to the
signal level, is called the passband.

Cutoff frequency:

The passband cutoff frequency (or cutoff point) is the passband edge where there is a 3 dB reduction
in signal amplitude (the half-power point).

Stopband:

The range of signal frequencies that are reduced in amplitude by an amount specified in the design, and
effectively prevented from passing, is called the stopband.
Skirt response:

In between the passband and the stopband is a range of frequencies called the skirt response, where the
reduction in signal amplitude (also known as the attentuation) changes rapidly.

Frequency Response:

The simple figures in (a), (b), (c), and (d) depict the four types of filter characteristics with respect to
frequency.

(a) Lowpass

(b) Highpass
 (c) Bandpass

(d) Bandstop

The designer must determine the cutoff frequencies, the stopband attenuation, and whether a lowpass,
highpass, bandpass, or bandstop filter is required. Sometimes this specification will be supplied by the
system designer, but this may be left to the filter designer to decide for him or herself.

Phase Response:

Radians/sec are used as the unit of frequency measure because in an analog filter this gives a value for
reactive impedance, which is directly proportional to the frequency.

For AC measurements with loads that include reactive elements like capacitors and inductors, the current
can be expressed as the ratio of voltage to load impedance. If there is some reactance in the load, the
current through the load is not in phase with the voltage across it.
The power dissipated at a resistive load is the product of voltage and current averaged over one sine
wave cycle. This is the r.m.s. voltage times the r.m.s. current.

No power is dissipated in a purely reactive load because over one complete sine wave cycle the product
of voltage and current is zero.

Instead, energy is stored in capacitors and inductors, which is the reason for the phase difference
between voltage and current at a reactive load.

Simple Lowpass and highpass RC networks are illustrated in Fig. 8.

Fig.8. Lowpass and Highpass RC Networks.

The ideal filter has a rectangular shape, indicating that the boundary between the passband and the
stopband is abrupt and that the rolloff slope is infinitely steep.

This type of response is ideal because it allows us to completely separate signals at different
frequencies from one another. Unfortunately, such an amplitude response curve is not physically
realizable.

We will have to settle for the approximation that will still meet our requirements for a given application.
Deciding on the best approximation involves making a compromise between various properties of the
filter’s transfer function.

The important properties are the following.

Filter order: The order of a filter has several effects. It is directly related to the number of components in
the filter and, therefore, to its price and the complexity of the design task.
Therefore, higher-order filters are more expensive, take up more space, and are more difficult to
design.

The primary advantage of higher-order filters is that they will have steeper rolloff slopes than similar
lower-order filters.

Rolloff rate: Usually expressed as the amount of attenuation in dB for a given ratio of frequencies. The
most common units are “dB/decade” or “dB/octave.”

From Fig. 4, four parameters are of concern:

Amax is the maximum allowable change in gain within the passband. This quantity is also often called the
maximum passband ripple.

Amin is the minimum allowable attenuation (referred to the maximum passband gain) within the
stopband.

f1 is the cutoff frequency or passband limit.

fs is the frequency at which the stopband begins.

These four parameters define the order of the filter.

Fig. Low Pass Filter Characteristics (a) Actual (b) Ideal

Fig. High Pass Filter Characteristics (a) Actual (b) Ideal

Fig. Band Pass Filter Characteristics

Fig. Band Stop Filter Characteristics

 Passive Low Pass Filter (LPF)

 LPF ideally allows lower frequencies and attenuates higher frequencies.

 One of the simplest form of LPF

Transfer Function:

is the cut-off frequency.

It is a frequency at which ( ) drops to 70.7% of

|( )max | or becomes of | ( )|max.

So, here, can be calculated as:

 …………………………….(103)

Fig. Ideal and Actual Lowpass Filter characteristics.

Passive High Pass Filter (HPF)

 Ideally, HPF attenuates lower frequencies and allows higher frequencies.

 One of the simplest form of HPF

Transfer Function:

Fig. Ideal and Actual Highpass Filter characteristics.

Passive Band Pass Filter (BPF)

 BPF allows frequencies of a particular range and eliminates other frequencies.

 Typical example of BPF

Transfer function :

 Here, H(0) = 0 and H(∞) = 0

 How it is BPF ?

 Resonance Frequency, 0!!!!!

 Zeq= R ⇒Filter allows 0means it is a BPF

Here, 1 and 2 are half power frequencies i.e. power dissipated is half of the maximum power.

Taking, |H(ω)| = 1 for the center frequency (ω0) within the passband and get

…………………(104)

Taking, |H(ω)| =

We get the expressions of the upper cutoff frequency (ω2) and lower cutoff frequency (ω1) as follows.
Bandwidth of BPF = 2- 1

Quality Factor,

Fig. Ideal and Actual Bandpass Filter characteristics.

Passive Band Stop Filter

 It rejects a particular range of frequencies and allows rest of the frequencies.

Example of band stop filter :

Transfer Function

Here, H(0) = 1 and H(∞) = 1.

But at resonance frequency,

0=0⇒Filters does not allow 0

Fig. Ideal and Actual Bandstop Filter characteristics.

FILTER ANALYSIS
Characteristic Impedances of T- and –π Sections

T-Section:

π-Section:
 Fig. 9. Symmetrical T-section filter.

Fig. 10. Symmetrical π-section filter.

Characteristic impedance of T-section in terms of series and shunt arms:

Looking into the network in Fig. 9:

 In order for Zin to equal Z0T it follows that:

The above equation may be solved for Z0T and the result stated in terms of Z1 and Z2 . Thus it can be
shown that the characteristic impedance of the T-section is:

……………………………………(105)

In the similar way, using Fig. 10 the characteristic impedance of the π-section is:

………………………………….……(106)

From Eq. 105 and 106 it is seen that

………………...……………………(107)
 and

………………………………………(108)

The next basic relationship to be establihed is that the ratio of input current to output current,
namely, I1/I2, is completely defined by the series arm impedance (Z1 ) and the shunt arm impedance (Z2) out
of which the symmetrical T- or π-section is composed.

For the T-section shown in Fig. 9 it is plain from Kirchhoff's emf law that

Whence,

………………………………(109)

Referring to Fig. 10 for the π-section and remembering that V1 = I1Z0πand that V2 = I2Z0T , the
current Iseries in the series arm is:

From which
 and

……………………(110)

Transmission Constant of a Filter Section

Assuming that the filter section is terminated on an image impedance basis and that we wish to
specify a measure of the attenuation and phase shift of the filter itself, we employ the following definition
of the transmission constant:

…………………………..(111)

where Z’T is the transfer impedance from the input terminals of the filter section to the output
terminals, namely, V1/I2,

ZI1 is the image impedance seen looking to the right of the input terminals, namely, V1 /I1

α is called the attenuation of the filter section is called the phase-shift constant of the filter section.

β is called the phase-shift constant of the filter section.

Actually the α and β defined in Eq. 111 apply to any four-terminal network which is terminated on an
image impedance basis as shown in Fig. 2.

The attenuation, α, is a measure of the ratio of the power input to the power output of a filter
section which is terminated in its characteristic impedance, since under these conditions the real part of Eq.
111 may be written as:
 …………………………..(112)

where R0 is the resistive component of Z0

W1 is the power entering the input terminals

W2 is the power leaving the output terminals.

From Eq. 111 it is plain that

Where
…………………(113)

angle of lead of I1 with respect to I2

As applied to a series or cascade arrangement‘of filter sections like those shown in Fig. 11:

……………………(114)
 Fig. 11. Three symmetrical T-sections terminated on a characteristic impedance basis.

Units of Attenuation or Transmission Loss

The Neper:

Attenuation in Neper ………………………(115)

or

Attenuation in Neper = …….………………..(116)

In reference to Eq. 112, for a filter section which is terminated in its characteristic impedance, the
output power W2 is employed as the reference power level and we can write

Attenuation in Neper

Attenuation in Neper ……………………(117)

The Decibel:

Attenuation in Decibel ….………………(118)

 If the filter section is terminated on a characteristic impedance basis,

Attenuation in Decibels

…………………..(119)

A Fundamental Filter Equation

An equation which defines the propagation constant of a filter section wholly in terms of an
arbitrarily selected series arm (Z1) and an arbitrarily selected shunt arm (Z2) is necessary in the design of
filter sections.

Reference to Eq. 109 and Eq. 110 and to Eq. 111 shows that

Putting the value of Z0T and Z0π from Eq. 105 and Eq. 106 respectively we get the following equations.
 …………………..(120)

…………..…………………..(121)

Hence, for like values of Z1and Z2;

………………………………(122)

Since , it follows that

……………….(123)

Taking natural log in the both sides of Eq. 123 we have the following relationship.

……………………………(124)
 The above relation (Eq. 124) is one form of fundamental filter equation, since the attenuation
constant and the phase-shift constant are defined wholly in terms of the full series arm impedance (Z1) and
the full shunt arm impedance (Z2).

The analysis of any symmetrical T- or π-section composed of series and shunt arms of Z1 and Z2 ,
respectively, may be carried through with the aid of Eq. 124.

Filter Section Analysis Assuming Zero Resistance

It is quite customary to neglect the resistive components of Z1 and Z2 in filter section analysis because
the attenuation produced by these resistive components is incidental to the predominant filtering action that
takes place.

The discrepancy between theoretical results based on zero resistance and actual results will not be
great if the resistances are relatively small compared with the reactances.

Also the algebraic manipulations involved in filter design are greatly simplified by neglecting the
resistive components of Z1 and Z2.

If the above resistances are neglected and if the filter sections are properly terminated, the pass bands
are transmitted with zero attenuation while the stop bands experience certain varying degrees of
attenuation. It will also he shown that the phase shift is 180° throughout the stop band under the
conditions stated above.

Example 8. (book by Corcoran, fourth edition) Consider a symmetrical T-section in which Z1 = jωL1
and in which Z2 = ‒1/jωC2 Let it be required to predict the behavior of the filter section wholly in terms of
the relationship stated in Eq. 124.

Solution:

Here

Since Z1/4Z2 possesses the general form given above, it will be convenient to reckon ω in
 units, thereby giving Z1/4Z2 definite numerical values for various different
frequency units.

The evaluation of the right-hand member of Eq. 124 for various frequencies is shown in tabular form
in Table I below.

The variations of attenuation and phase shift can readily be determined from an examination of
columns (8) and (9) of the table. It will be observed that the filter section which is under discussion has

theoretical zero attenuation between the limits of ω = 0 and ω = radians per

second.

The section obviously operates as a low-pass filter. The arrangement of the series and shunt arms of
this low-pass filter together with the general trends in the variations of attenuation and phase shift are shown
in Fig. 12.
 Fig. 12. Variations of phase shift and attenuation in a prototype low-pass filter section in the example
8.

The fact that the cut-off point occurs at

radians per second will be given more attention in a later article. The present

example concerns itself primarily with the development of Eq. 124 in a particular case.

Example 9. (book by Corcoran, fourth edition) Consider a symmetrical T-section in which Z1 =

‒1/jωC1 and in which Z2 = jωL2 Let it be required to predict the behavior of the filter section wholly in
terms of the relationship stated in Eq. 124.

Solution:

In the present case:

 In the similar way as in Example 8, The calculations are indicated in Table II, and results are shown
graphically in Fig. 13. T

Fig. 13. Variations of phase shift and attenuation in a prototype low-pass filter section in the example
9.
 A study of Tables I and II will show that the pass hands are limited to those regions where Z1/4Z2
possesses values between 0 and - 1. These results might have been anticipated mathematically by
investigating the possible values of α and β when Z1 and Z2 are reactances of opposite types.

Let

It is plain that A = |A|<± π since

are complex numbers which have associated angles of + π or - π radians, respectively.

If

Hence α = 0 and β = when A = Z1 /4Z2 lies between 0 and —1.

When Z1/4Z2 lies between —1 and —∞ similar analysis will show that for Z1/4Z2 = A‘<±π, A‘ being
greater in magnitude than unity.

Hence α = and β = ± π when A = Z1 /4Z2 lies

between —1 and —∞.
 The above analysis shows that the pass bands are limited to those regions as shown below.

………………………(125)

From Eq. 125 it is clear that the boundaries of a pass band in a particular case may be obtained by
setting:

and

or by setting

and ..…………(126)

Cut-Off Frequencies of Elementary Low- and High-Pass Sections

The frequency limits of the pass band for an elementary low-pass filter without resistance may be
obtained from Eq. 123. For low-pass filter

and .

Then from Eq. 123 we get

 ……………………(127)

For no attenuation α = 0, and

……………………(128)

Since the last term of equation (44) is the only one that may become imaginary, it follows that the real
part must be cosβ. Therefore,

.......……………….(129)

Since cosβ can vary from 1 to —1, the limits for ω may be obtained.

Hence

and
 or

…........………………….(130)

For a high-pass filter,

and

If these values are substituted in Eq. 123, a similar method of analysis as used in obtaining Eq. 129 gives

…………..…………….(131)

Substituting the limits of ±1 for cosβ, the upper limit of frequency is found to be ∞ while the lower
limit or cut-off frequency is:

........………………….(132)

Eq. 130 and Eq. 132 gives the cut-off frequency for an elementary low- and high-pass filter
respectively. This means that any frequency within the pass-band (below the cut-off frequency for low-pass
case and above the cut-off frequency for high-pass) is passed with no attenuation if the filter section is
terminated in the characteristic impedance for the particular frequency.

Constant-k Low-Pass Filter

Filter sections in which the series and shunt arms are inverse impedance functions possess a peculiar

property. The product of Z1 and Z2 is independent of frequency. Reference to either the T- or π

section of Fig. 14 will show that
 is an important characteristic of the filter section, and inasmuch as

Fig. 14. Prototype or constant-k low-pass filter sections.

 filter sections of this type are called constant-k sections. There are many other types of filter
sections, several of which are derived in one way or another from constant-k sections.

The general theory of the constant-k low-pass filter has already been presented. It remains only to
develop the design equations for this type of filter.

…………………..(133)

The boundaries of the pass hand are determined by setting Z1k /Z2k equal to —4 and equal to zero.

……………(134)

……………(135)

The cut–off frequency of a low-pass, constant-k-type filter is:

……………(135)
 It will be observed that fC is governed wholly by the magnitude of the L1kC2k product. The lower the
cut-off frequency, the higher is the L1kC2k product, and vice versa.

Another important consideration in either the theory or design of a filter section is the matter of
correct terminating impedances.

A single section can be properly matched to its sending and receiving ends if terminated on an
image basis, as explained previously.

For a constant-k, low-pass T-section:

……………(136)

(see Eq. 135)

Therefore, for a constant-k, low-pass T-section:

……………(137)
 For a constant-k, low-pass π-section:

……………(138)

The variations of ZOTk and ZOπk from f = 0 to f = fC are illustrated in Fig. 15.

The fact that the correct terminating impedance of a constant-k section varies over such wide limits is
a very serious limitation in certain communication circuits.

For a fixed receiving impedance it is plain that either the T- or π-section is correctly terminated at only
one frequency.

The opposite trends in Z0Tk and ZOπk are combined in one form of filter section to obtain a
characteristic impedance which is reasonably constant over the frequency range of the pass band.

The zero-frequency value of either ZOTk or Z0πk is:

(see Eq. 137, 138) …………………(139)

L1k and C2k can be related to one another through the value of RK as seen from Eq. 139 as follows.

……………………………(140)
 ……………………………(141)

The design values of L1k and C2k are usually specified in terms of cut-off frequency, fC and the zero-
frequency value of the characteristic impedance, RK.

In similar way, the equation of cut-off frequency and characteristic impedance of Constant-k
High-Pass Filter are as follows.
 Fig. 15. Variations of the characteristic impedances of low-pass and high-pass constant-k; filter
sections.