0% found this document useful (0 votes)
74 views

Cavity Resonators Lecture2

1) A cylindrical cavity resonator is formed by closing the ends of a cylindrical waveguide with conducting material. Maxwell's equations are used to derive the wave equations for the electric and magnetic fields. 2) The wave equations are solved using variable separation, resulting in Bessel functions for the radial dependence and trigonometric functions for the angular and axial dependencies. 3) The resonant frequencies are determined by the boundary conditions that the fields must be zero at the ends of the cavity, resulting in quantized values of the axial wave number.

Uploaded by

Kamran Razi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
74 views

Cavity Resonators Lecture2

1) A cylindrical cavity resonator is formed by closing the ends of a cylindrical waveguide with conducting material. Maxwell's equations are used to derive the wave equations for the electric and magnetic fields. 2) The wave equations are solved using variable separation, resulting in Bessel functions for the radial dependence and trigonometric functions for the angular and axial dependencies. 3) The resonant frequencies are determined by the boundary conditions that the fields must be zero at the ends of the cavity, resulting in quantized values of the axial wave number.

Uploaded by

Kamran Razi
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Cavity (Waveguide) Resonators

Cylindrical Cavity Resonator


A Cylindrical cavity Resonator is formed by closing the ends of a cylindrical waveguide by
perfect conducting material. Consider a cylindrical cavity of radius ‘a’ and length‘d’ filled with a
lossless dielectric material having permittivity 𝜀 and permeability 𝜇 as shown.

o d
Y
a

Maxwell equations for harmonically varying fields in a source free lossless dielectric medium
are expressed as
  H  j  E (7)
  E   j  H (8)
. D  0 (9)
. B  0 (10)
Also D  E (11)
B  H (12)
JE (13)
Taking curl of (8) and substituting (12) we get
    E   j  (  H )
(  . E )   2 E   j  ( j   E )
 . D 
    E    E
2 2

  
 E   2   E
2
(14)
Similarly one can derive
2 H    2   H (15)
Eqs. (14) and (15) represent three dimensional electromagnetic wave equations.
Assuming that the electromagnetic wave is propagating along +z-direction, but in fact, there will
be only standing waves. The eq. (14) for z-component of electric field can be expressed as
 2 Ez    2   Ez
In cylindrical co-ordinate system this can be expressed as
Page | 1
Cavity (Waveguide) Resonators

2 1  1 2 2
E  E  E  EZ    2   E z (16)
     z
2 z z 2 2 z 2

This is a second order partial differential equation which can be solved by using variable
separable method. Ez is a function of ρ, ϕ, z and t i.e.
E z  F (  ,  , z, t )
Let Ez  F (  ) F (  ) F ( z ) e j  t (17)
Using eq. (16) the eq. (17) can be expressed as
2 1  1 2 2
Ez  Ez  2 E z  2 EZ    2   E z
 2
    2
z
1 2 1  1 1 2 1 2
F (  )  F (  )  F ( )  F ( z)    2  
F ( )   2  F ( )    2 F ( )   2 F ( z)  z 2
1 2
Let F ( )   n 2
F ( )   2

1 2
And F ( z )   k z2
F ( z)  z 2

1 2 1  n2
Therefore F (  )  F (  )    2
   k 2

F ( )   2  F ( )   2
z

2
Or F ( )  n 2 F ( )  0 (18)
 2
2
F ( z )  k z2 F ( z )  0 (19)
z 2

2 1   n2 
F ( )  F ( )   k 2  2  F (  )  0
 2
    
Replacing ρ by kρρ we get
 2
1   n 2 
Or F (k   )  
F (k   )  1  F (k   )  0 (20)
 (k   ) 2 (k   )  (k   )  (k  ) 2 
  
Where k2   2    kz2
The general solutions for (18), (19) and (20) are
F ( )  An Cos n  Bn Sin n
F ( z )  M Cos k z z  N Sin k z z
F (k   )  J n (k   )  N n (k   )
The complete general solution is given by
Ez  J n (k   )  N n (k   ) ( An Cos n  Bn Sin n )E Cos k z z  F Sin k z z  e j  t (21)
Since Nn(kρ ρ) is infinite at hρ ρ = 0, hence it cannot be the part of general solution for a source
free conditions.

Page | 2
Cavity (Waveguide) Resonators

Again from the boundary conditions the component Ez should be continuous at z=0 and z=d.
This implies that Ez = 0 at z = 0 and z = d,
Using the boundary condition Ez = 0 at z=0 in eq. (21) leads to constant M = 0
Using condition Ez =0 at z=d in eq. (21) gives kzd = pπ (where p= 1, 2, 3,4 …. )
p
or kz  (22)
d
Thus the complete solution of eq. (16) is
Ez  J n (h  )( An Cos n  Bn Sin n )N Sin k z z  e j  t (23)
Similarly one can arrive at a solution for Hz as
Ez  J n (h  )( An Cos n  Bn Sin n )N Sin k z z  e j  t (24)
The other field components like Eρ, Eϕ, Hρ and Hϕ can be derived in terms of Ez and Hz as
follows.
Maxwell curl equations in cylindrical co-ordinates can be expressed as
1 1 
  u u u
 z
 
   
 H    j   E  u   E u  E z u z 
   z 
 H   H H z 
 
 
1 1 
  u u u
 z
 
   
 E    j   H  u   H  u  H z u z 
   z 
 E  E H z 
 
 
1  
Or H  H  j   E (25A)
  z  z 
 
H  H z  j   E (25B)
z 
1  1 
( H  )  H  j   Ez (25C)
    
1  
Ez  E  j   H  (25D)
  z
 
E  Ez  j   H (25E)
z 
1  1 
( E )  E  j Hz (25F)
    
Multiplying eq. (25A) by -jωu we get

Page | 3
Cavity (Waveguide) Resonators

 j   
Hz  ( j  H  )   2   E
  z
Substituting (25E) we have
 j      
Hz   E  E z    2   E
   z  z  
 j   2  
Or H z  2 E  E z   2   E
  z  z
Since each field component is a function of (Sin kzz) we have
 2
E z  jk z E z And E    k z2 E 
z z 2

 j   
Hence we have H z  k z2 E  jk z E z   2   E
  
 j   
Or
 
H z  jk z

 
E z   2    k z2 E  k 2 E

 j   jk 
Thus E  H z  2z Ez (26A)
 k  
2
k  
j   jk 
Similarly E  H z  z2 E (26B)
k  
2
 k   z
jk z  j 
H  Hz  E (26C)
k  
2
k 2  z
jk z  j 
H  Hz  2 Ez (26D)
 k  
2
k  
Two important cases are
Case 1: Transverse Magnetic (TM) Mode
In this Case Ez≠0 , Hz = 0 ; In this case the field components are given by
Ez  J n (k   )( An Cos n  Bn Sin n ) N Sin k z z  e j  t (27A)
Hz 0 (27B)
jk z 
E  Ez (27C)
k 2 
jk z 
E  E (27D)
 k 2  z
j 
H   E (27E)
k 2  z
j 
H   Ez (27F)
k 2 
Applying boundary conditions which state Ez = 0 at ρ = a. This leads to

Page | 4
Cavity (Waveguide) Resonators

J n ( k  a)  0 (28)
If ξnm is the m root of eq. (28) then
th


k   nm  hnm (29)
a
Where n denotes the order of Bessel function and m is the mth root.
The first few roots of eq. (28) are given in table 1.
Table 1: Roots of Jn(kρa)
m n
0 1 2 3 4 5
1 2.405 3.823 5.136 6.380 7.588 8.771
2 5.520 7.016 8.417 9.761 11.065 12.339
3 8.654 10.173 11.620 13.015 14.732
4 11.792 13.324 14.796

The roots of eq. (28) and values of integer constant ‘p’ will describe the different TMnmp modes
in a cylindrical resonant cavity. Remember ‘p’ cannot be zero.
Resonant frequencies for TMnmp mode
If fr is the resonant frequency of TMnmp mode, then
Since hnm2   2    k z2
Or r2    h 2  k z2
nm

  nm  2  p  
2
1
Or fr     
2    a   d 
The lowest resonant frequency in T011

Case 2: Transverse Electric (TE) Mode


For TE modes we have Ez= 0 , Hz ≠ 0;
In this case the other field components are given by
Ez  0 (30A)
H z  J n (k   )(Cn Cos n  Dn Sin n ) N Sin k z z e j  t (30B)
 j  
E  H (30C)
 k 2   z
j  
E  Hz (30D)
k 2  
jk z 
H  Hz (30E)
k 2  
jk z 
H  H (30F)
 k 2   z
Applying boundary conditions which state that Eρ = 0 at ϕ = 0 and Eρ = 0 at ρ = a.
Thus from (30C) we have

Page | 5
Cavity (Waveguide) Resonators

 j  
E  H 0 at φ=0
 k 2   z
 j  
E  J (k  ) n C n Sin n  Dn Cos n  ( N Sin k z z ) e j t
 k 2   n 
 j 
E  J n (k   ) n C n Sin n  Dn Cos n  ( N Sin k z z ) e j t
 k 2

Thus substituting ϕ = 0 we get


 j 
E  J (k  ) n  Dn ( N Sin k z z ) e j t  0
 k 2 n 
This implies Dn= 0
Hence H z  J n (k   )Cn Cos n ( N Sin k z z)e j  t (31)
From (30D) we have
j  
E  J n (k   )Cn Cos n ( N Sin k z z )e j  t
k  
2

j  /
E  2
J n (k   )Cn Cos n ( N Sin k z z )e j  t
k
Since at the boundary Eϕ = 0 at ρ = a
j  /
Therefore 2
J n (k  a)Cn Cos n ( N Sin k z z )e j  t  0
k
This implies J n/ (k  a)  0 (32)
The first few roots of eq. (32) are given in following table 2.
Table 2: Roots of derivative of Bessel Function
m n
0 1 2 3 4 5
1 3.832 1.841 3.054 4.201 5.317 6.416
2 7.016 5.331 6.706 7.015 9.282 10.520
3 10.173 8.536 9.969 11.346 12.682 13.987
4 13.324 11.706 13.170
If τnm represent the root of (32), then

k   nm  hnm
a
Where n denotes the order of derivative of Bessel function and m is the mth root.
The roots of eq. (32) and values of integer ‘p’ will define the various TEnmp modes of cylindrical
cavity. Here again p cannot be zero.

Resonant Frequency for TEnmp modes


If fr is the resonant frequency of TEnmp mode then
1   nm p  
f r TE    
2    a d 

Page | 6
Cavity (Waveguide) Resonators

The lowest root is J1/(ρh) which is 1.841. This value corresponds n = 1 and m = 1. The dominant
mode for cylindrical resonant cavity is TE111 mode will result in the lowest resonant frequency.

Page | 7

You might also like