w. geyi- time-domain theory of metal cavity resonator

8/3/2019 W. Geyi- Time-Domain Theory of Metal Cavity Resonator

1/35

Progress In Electromagnetics Research, PIER 78, 219253, 2008

TIME-DOMAIN THEORY OF METAL CAVITYRESONATOR

W. Geyi

Research In Motion295 Phillip Street, Waterloo, Ontario, Canada N2L 3W8

AbstractThis paper presents a thorough study of the time-domaintheory of metal cavity resonators. The completeness of the vectormodal functions of a perfectly conducting metal cavity is first proved

by symmetric operator theory, and analytic solution for the fielddistribution inside the cavity excited by an arbitrary source is thenobtained in terms of the vector modal functions. The main focus ofthe present paper is the time-domain theory of a waveguide cavity,for which the excitation problem may be reduced to the solution of anumber of modified Klein-Gordon equations. These modified Klein-Gordon equation are then solved by the method of retarded Greensfunction in order that the causality condition is satisfied. Numericalexamples are also presented to demonstrate the time-domain theory.The analysis indicates that the time-domain theory is capable ofproviding an exact picture for the physical process inside a closed cavityand can overcome some serious problems that may arise in traditionaltime-harmonic theory due to the lack of causality.

1. INTRODUCTION

The rapid progress in ultra-wideband technologies has prompted thestudy of time-domain electromagnetics. Compared to the voluminousliterature on time-harmonic theory of electromagnetics, the time-domain electromagnetics is still a virgin land to be cultivated. Inthe time-domain theory, the fields are assumed to start at a finiteinstant of time and Maxwell equations are solved subject to initialconditions, boundary conditions, excitation conditions and causality.A metal cavity resonator constitutes a typical eigenvalue problem inelectromagnetic theory and has been investigated by a number ofauthors [e.g., 15], and the study of the transient process in a metal


2/35

220 Geyi

cavity may be carried out by the field expansions in terms of the modalvector functions. When these expansions are introduced into the time-domain Maxwell equations one may find that the expansion coefficientssatisfy the ordinary differential equations of second order [3], which

can be easily solved once the initial conditions and the excitationsare known. Recently this approach has been used to investigate theresponses of the metal cavity to digital signals [4].

Although the study of metal cavity resonator has a long history,there are still some open questions which need to be investigated. Thispaper attempts to answer these questions, and presents a thoroughdiscussion on the time-domain theory of metal cavities. The paperis organized as follows. Section 2 studies the eigenvalue theoryof a perfectly conducting metal cavity filled with lossy medium.A fundamental problem in metal cavity theory is to prove thecompleteness of its vector modal functions, which has been tried byKurokawa [1]. But there is a loophole in Kurokawas approach inwhich he fails to show the existence of the vector modal functions.

The main purpose of Section 2 is to provide a rigorous proof of thecompleteness of the vector modal functions on the basis of the theoryof symmetric operators. Section 3 summarizes the field expansionsinside an arbitrary metal cavity filled with lossy medium in termsof the vector modal functions, and discusses the limitations of thetime-harmonic theory when it is applied to a closed metal cavity.Section 4 is dedicated to a metal cavity formed by a section of uniformwaveguide, i.e., the waveguide cavity. The time-domain theory of thewaveguide developed previously [6] is applicable to this case, and thefields inside the waveguide cavity can be expanded in terms of thetransverse vector modal functions of the corresponding waveguide. Theexpansion coefficients of the fields are shown to satisfy the modifiedKlein-Gordon equation subject to homogeneous boundary conditions,

which are then solved by the method of retarded Greens function tosatisfy the causality requirement. In order to validate the theory, someexamples are expounded in Section 5, and the field responses to typicalexcitation waveforms are demonstrated.

An important observation in this paper is that the causality playsan important role in determining the field response of a closed cavity.In other words, one must take the initial conditions into considerationin order to obtain a correct field response (transient or steady-state).The time-domain theory shows that a sinusoidal response can be builtup if and only if the cavity is excited by sinusoidal source whosefrequency coincides with one of the resonant frequencies of the cavity.The traditional time-harmonic theory, however, predicts that the fieldresponse is always sinusoidal if the excitation source is sinusoidal.


3/35

Progress In Electromagnetics Research, PIER 78, 2008 221

This discrepancy comes from the lack of causality in traditional time-harmonic theory, which assumes that the source is turned on at t = instead of a finite instant of time and has ignored the initial conditions.

Another interesting observation is that the field responses in a

lossless cavity predicted by the time-harmonic theory are singulareverywhere inside the cavity if the frequency of the sinusoidalexcitation source coincides with one of the resonant frequencies ofthe cavity, while the time-domain theory always gives finite fieldresponses. The singularities in time-harmonic theory can be removedby introducing losses inside the cavity, which is essentially required bythe uniqueness theorem for a time-harmonic field in a bounded region.

2. EIGENVALUE THEORY FOR METAL CAVITYRESONATOR

The metal cavity resonator constitutes a typical eigenvalue problem,where the eigenvalues correspond to resonant frequencies of thecavity and eigenfunctions correspond to the natural field distributions.Given the eigenvalue problem, one must show the existence andcompleteness of the eigenfunctions, and the latter implies that the setof eigenfunctions can be used to expand an arbitrary function.

2.1. Eigenvalue Problems for a Metal Cavity

Let us consider a metal cavity with a perfectly conducting wall. Itwill be assumed that the medium in the cavity is homogeneous andisotropic with medium parameters , and . The volume occupiedby the cavity is denoted by V and the boundary of V by S (Figure 1).Since the metallic wall is a perfect conductor, the transient fields in

S

, ,

V

Figure 1. An arbitrary metal cavity.


4/35

222 Geyi

the cavity satisfy Maxwell equations

E(r, t) = tH(r, t)

H(r, t) = tE(r, t) + E, r V

E(r, t) = 0 H(r, t) = 0

(1)

with boundary conditions un E= 0 and un H= 0, where un is theunit outward normal to the boundary S and all other notations havetheir usual meaning. From (1), one may obtain

E(r, t) +

2E(r, t)

t2+

E(r, t)

t= 0, r V

un E(r, t) = 0, r S(2)

H(r, t) +

2

H(r, t)t2

+ H(r, t)t

= 0, r Vun H(r, t) = 0, un H(r, t) = 0, r S

(3)

If the solutions of (2) and (3) can be expressed as a separable functionof space and time

E(r, t) = e(r)u(t), H(r, t) = h(r)v(t)

it follows from (2) and (3) that e k2ee = 0, e = 0, r Vun

e = 0, r

S

(4)

h k2hh = 0, h = 0, r Vun h = 0, un h = 0, r S (5)

The functions u(t) and v(t) satisfy

1

c22u

t2+

c

u

t+ k2eu = 0 (6)

1

c22v

t2+

c

v

t+ k2hv = 0 (7)

where k2e and k2h are separation constants, =

/ and c =

1/

. Both (4) and (5) form an eigenvalue problem. However their


5/35


eigenfunctions do not form a complete set. To overcome this difficulty,(4) and (5) can be modified as [1]

e e k2ee = 0, r V

un e = 0, e = 0, r S(8)

h h k2hh = 0, r Vun h = 0, un h = 0, r S (9)

These are the eigenvalue equations for the metal cavity system.

2.2. Completeness of the Eigenfunctions

The eigenvalue problems (8) and (9) have been discussed by Kurokawa[1]. Kurokawas approach suffers a drawback that has overlooked theproof of the existence of the eigenfunctions and eigenvalues. In thefollowing, a rigorous approach will be presented, which is similar to

what is used in studying the waveguide eigenvalue problem [6]. Let usrewrite the eigenvalue problem (8) as

B(e) = e e = k2ee, r Vun e = 0, e = 0, r (10)

where B = . The domain of definition of operator B isdefined by

D(B) =ee (C())2 , un e = 0, e = 0 on

where C(V) stands for the set of functions that have continuous

partial derivatives of any order. Let L

2

(V) stand for the spaceof square-integrable functions defined in V and H = (L2(V))3 =L2(V) L2(V) L2(V). For two vector fields e1 and e2 in (L2(V))3,the inner product is defined by (e1,e2) =

Ve1 e2dV (a bar is used

to designate the complex conjugate) and the corresponding norm isdenoted by = (, )1/2. (10) can be modified as an equivalent formby adding a term e on both sides

A(e) = e e+ e = k2e + e, r Vun e = e = 0, r (11)

where is an arbitrary positive constant. It is easy to show that

the operator A is symmetric, strongly monotone. In fact, for all


6/35

224 Geyi

e1,e2 D(A) = D(B), the energy product is given by(e1,e2)A = (A(e1),e2) =

V

( e1 e1 + e1) e2d

=V

[ e1 e2 + ( e1)( e2) + e1 e2] d (12)

Therefore the new operator A is symmetric. Thus e can be assumedto be real. A is also strongly monotone since (A(e),e) e2.Therefore one can introduce the energy inner product of the operator Adefined by (u, v)A = (Au, v). The energy space HA is the completion

of D(A) with respect to the norm A = (, )1/2A [7]. Let e HA,and by definition, there exists admissible sequence {en D(A)} for esuch that en e 0 as n and {en} is a Cauchy sequence inHA. It follows from (12) that

en

em

2A

=

en

em2+

en

em

2+

en

em

2

Consequently { en} and { en} are Cauchy sequences in H. Asa result, there exist h H, and L2(V) such that en h and en as n . From integration by parts, one may write

V

en dV =V

en dV, (C0 (V))3

V

( en)dV = V

en dV, C0 (V)

Letting n yields

V

h

dV =

V

e

dV,

(C0 (V))

3

V

dV = V

e dV, C0 (V).

In the above C0 (V) is the set of all functions in C(V) that vanish

outside a compact subset of V. Therefore e = h and e = hold in the generalized sense. For arbitrary e1,e2 HA, there are twoadmissible functions {e1n} and {e2n} such that e1n e1 0 ande2n e2 0 as n . Define

(e1,e2)A = limn

(e1n,e2n)A

= V

[

e1

e2 + (

e1)(

e2) + e1e2] dV


7/35


where the derivatives must be understood in the generalized sense.Now one can show that the embedding HA H is compact. LetJ(e) = e, e HA. Then the linear operator J : HA H iscontinuous since

J(e)2 = e2 1 e2 + e2 + e2 = 1e2HA.A bounded sequence {en} HA implies

en2HA = en2 + en2 + en2

=

V

(enx)

2 + (eny)2 + (enx)2 + (eny)2

d c

where c is a constant. So the compactness of the operator J followsfrom the above inequality and the Rellichs theorem [1, 7]. Thus thefollowing theorem may apply to the operator A [7, pp.284]

Theorem: Let H be a real separable Hilbert space with dim H = and A : D(A) H H be a linear, symmetric operator. Assumethat A is strongly monotone, i.e., there exists a constant c1 such that(Au, u) > c1u2 for all u D(A). Let AF be the Friedrichs extensionofA and HA be the energy space of operator A. If the embeddingHA H is compact, then the following eigenvalue problem

AFu = u, u D(AF)has a countable eigenfunctions {un}, which form a completeorthonormal system in the Hilbert space H, with un HA. Eacheigenvalue corresponding to un has finite multiplicity. Furthermore1 2 , and n n.

It follows from this theorem that there exists a complete set ofreal eigenfunctions {en}, called electric field modal functions, andthe corresponding eigenvalues, denoted by k2e,n, which approach toinfinity as n . The set of modal functions will be assumed tobe orthonormal, i.e.,

Vem endv = mn. It can be shown that each

modal function can be chosen from one of the following three categories[1]:

I. en = 0, en = 0II. en = 0, en = 0

III. en = 0, en = 0.Similarly one can show that the eigenvalues k2h,n of (9) are real and

positive, and the corresponding magnetic modal functions hn are


8/35

226 Geyi

real and constitute a complete set, which will be assumed to beorthonormal, i.e.,

Vhm hndv = mn. Also each modal function can

be chosen from one of the following three categories:

I. hn = 0, hn = 0II. hn = 0, hn = 0

III. hn = 0, hn = 0.The modal functions belonging to category II in the two sets of modalfunctions {en} and {hn} are related to each other. Actually let enbelong to category II. Then ke,n = 0 and one can define a function hnthrough

en = ke,nhn (13)Therefore hn belongs to category II. Furthermore

h

k2e,nh = k

1e,n

e

k2e,ne = 0, r

V

and

un hn = k1e,nun en = k1e,nun k2e,nen = 0, r SConsider the integration ofun hn over an arbitrary part ofS, denotedS

S

un hnds = k1e,nS

un ends = k1e,n

en ud

where is the closed contour around S and u is the unit tangentvector along the contour. The right-hand side is zero. Thus un hn = 0since S is arbitrary. Therefore hn satisfies (9) and the corresponding

eigenvalue is k2e,n. Ifhm is another eigenfunction corresponding to embelonging to category II, thenV

hm hndv = (ke,mke,n)1V

em endv

= (ke,mke,n)1S

unem ends+(ke,n/ke,m)V

em endv

= mn

Therefore the eigenfunctions hn in category II can be derived from theeigenfunction en in category II. Conversely ifhn is in category II, onecan define en by

hn = kh,nen (14)


9/35


and a similar discussion shows that en is an eigenfunction of (4) withkh,n being the eigenvalue. So the completeness of the two sets are stillguaranteed if the eigenfunctions belonging to category II in {en} and{hn} are related through either (13) or (14). From now on, (13) and(14) will be assumed to hold, and ke,n = kh,n will be denoted by kn.Note that the complete set {en} is most appropriate for the expansionof electric field, and {hn} for the expansion of the magnetic field.

3. TRANSIENT FIELDS IN A METAL CAVITY FILLEDWITH LOSSY MEDIUM

If the cavity contains an impressed electric current source J and amagnetic current source Jm, the fields excited by these sources satisfy

H(r, t) = E(r, t)t

+ E+ J(r, t), r V

E(r, t) =

H(r, t)

t J

m(r, t), r

V

(15)

and can be expanded in terms of the vector modal functions as

E(r, t) =n

en(r)

V

E(r, t) en(r)dv +v

ev(r)

V

E(r, t) ev(r)dv

=n

Vn(t)en(r) +v

Vv(t)ev(r)

H(r, t) =n

hn(r)

V

H(r, t)hn(r)dv+

h(r)

V

H(r, t) h(r)dv

=n

In(t)hn(r) +

I(t)h(r)

(16) E(r, t) =n

hn(r)

V

E(r, t) hn(r)dv

+

h(r)

V

E(r, t) h(r)dv

H(r, t) =n

en(r)

V

H(r, t) en(r)dv

+v

ev(r)

V

H(r, t) ev(r)dv

(17)

where the subscript n denotes the modes belonging to category II, andthe Greek subscript v and for the modes belonging to category I or


10/35

228 Geyi

III, and

Vn(v)(t) =

V

E(r, t) en(v)(r)dv

In()(t) =V

H(r, t) hn()(r)dv(18)

Making use of the following calculationsV

E hndv =V

E hndv +S

(E hn) unds = knVnV

E hdv =V

E hdv +S

(E h) unds = 0V

H ends =V

H endv +S

(H en) unds = knIn

V

H evds = V

H evdv + S

(H ev) unds = 0

(17) can be written as

E=n

knVnhn, H=n

knInen

Substituting the above expansions into (15) leads to

n

knInen = n

enVnt

+ v

evVvt

+ n

enVn + v

evVv + J

n knVnhn = n hn

In

t hI

t JmThus

Vnt

+

Vn kn

In = 1

V

J endv

Vvt

+

Vv = 1

V

J evdv

Int

+kn

Vn = 1

V

Jm hndv

I

t

=

1

V

Jm

hdv

(19)


11/35


From the above equations one may obtain

2Int2

+ 2Int

+ 2nIn = nSIn

2

Vnt2

+ 2Vnt

+ 2nVn = nSVn

(20)

where n = knc, = /2 and

SIn = c

V

J endv 1kn

t

V

Jm hndv ckn

V

Jm hndv

SVn =

kn

t

V

J endv cV

Jm hndv

To find In and Vn, one may use the retarded Greens function definedby

2Gn(t, t

)

t2+ 2

Gn(t, t)

t+ 2nGn(t, t

) = (t t)Gn(t, t

)t


12/35

230 Geyi

=n

2n 2t

0

e(tt) sin

2n 2(t t)

c V

J endv 1kn

t

V

Jm hndv dt (24)Similarly

Vn(t) =

Gn(t, t)nS

Vn (t

)dt

=n

2n 2t

0

e(tt) sin n(t t)

kn

tV

J endv c V

Jm hndv dt (25)Note that

Vv(t) = 1

ett

0

dtet

V

J evdv

I(t) = 1

t0

dtV

Jm hdv(26)

In order to validate the theory, let us consider a cavity excited by aninfinitesimal electric and magnetic dipole at r0 [3, pp. 538]

P = P0f(t)(r r0)M = M0f(t)(r r0)

The fields in the cavity satisfy

H(r, t) = 0E(r, t)t

+ P0f(t)(r r0), r V

E(r, t) = 0H(r, t)t

M00f(t)(r r0), r V(27)

Comparing (27) with (15), the following identifications may be made

J(r, t) = P0f(t)(r

r0), r

V

Jm(r, t) = M00f(t)(r r0), r V (28)


13/35


Introducing these into (24) gives

In(t) = cP0 en(r0)t

0

sin n(t t)f(t)dt

c0M0 hn(r0)

t0

cos n(t t)f(t)dt

= cnP0 en(r0)t

0cos n(t t)f(t)dt

M0 hn(r0)f(t) n

t0

sin n(t t)f(t)dt

(29)

If the excitation waveform is sinusoidal which is turned on at t = 0,i.e., f(t) = H(t)sin t, then (29) may be written as

In(t) =

2 2n[cnP0 en(r0)cos t + M0 hn(r0)sin t]

+

22n[cnP0 en(r0)cos t+M0 hn(r0)sin nt] (30)

Similarly one can obtain the expressions of Vn(t).From the time-harmonic theory in which f(t) = sin t, one would

obtain [3, 7.130b]

In(t) =

2 2n[cnP0 en(r0)cos t + M0 hn(r0)sin t] (31)

for a lossless cavity. Thus the second term of the right-hand side of(30) does not occur in (31). Also note that (31) is sinusoidal but (30)is not. Furthermore (30) does not approach to (31) as t , whichcontradicts our usual understanding. In fact, the response (30) basedon the time-domain analysis is not sinusoidal if the frequency of theexcitation sinusoidal waveform does not coincide with any resonantfrequencies, whereas the response (31) based on the time-harmonictheory is always sinusoidal. Therefore the time-harmonic analysis fora metal cavity suffers a theoretical drawback that its solution does notcorrespond to any practical situation in which a source is always turnedon at a finite instant of time. Apparently this theoretical drawback isdue to the lack of causality in the time-harmonic theory.

It can also be seen from (31) that In(t) becomes singular when approaches n, which implies that the fields are infinite everywhere


14/35

232 Geyi

inside the cavity. This phenomenon is discussed in Bladels book [5]and is compared to a lossless resonant LC network. In Collins book[3], these singularities do not occur because of the introduction of lossesin the metal cavity. However, the time-domain solution (30) has no

singularities even for a lossless cavity. In fact (30) may be rewritten as

In(t) =

nt

+ ncP0 en(r0)sin + n

2t

2t

+ nM0 hn(r0)cos + n

2t

sin

n2

t

n2

t

As approaches n, the above becomes

In(t) =nt

2[cP0 en(r0)sin nt M0 hn(r0)cos nt] (32)

for a finite time t, and no singularities appear in (32). Theabove phenomenon can be explained by the uniqueness theorem ofelectromagnetic field. For a bounded region (such as a cavity), thetime-harmonic Maxwell equations have a unique solution if and only ifthe region is filled with lossy medium, while the time-domain Maxwellequations always have a unique solution even if the medium is lossless[9, 10]. Therefore introducing losses in the metal cavity is required bythe time-harmonic electromagnetic theory, which guarantees that thesolution is unique and has no singularities.

Therefore the time-domain solution gives a more reasonablepicture for the physical process inside a metal cavity. More examplesin Section 5 will demonstrate this point.

4. TRANSIENT FIELDS IN A WAVEGUIDE CAVITYFILLED WITH LOSSY MEDIUM

The evaluation of modal functions in an arbitrary metal cavity isnot an easy task. When the metal cavity consists of a section of auniform metal waveguide, the analysis of the transient process in themetal cavity can be carried out by means of the time-domain theoryof waveguide [6].

4.1. Field Expansions in a Waveguide Filled with LossyMedium

Consider a waveguide cavity with a perfect electric wall of length L,as shown in Figure 2. The transient electromagnetic fields inside the


15/35


L 2z 1z

nu

z

J mJ

Figure 2. A metal cavity formed by a waveguide of cross-section .

waveguide cavity with current source J and Jm can be expressed as[6]

E(r, t) =n=1

vn(z, t)etn() + uz

n=1

etn()kcn

ezn

H(r, t) =

n=1

in(z, t)uzetn() + uz

1

uz H

+n=1

etn()kcn

hzn

(33)

where = (x, y) is the position vector in the waveguide cross-section; etn are the transverse vector modal functions, and

vn(z, t) =

E etnd, in(z, t) =

H uz etnd

hzn(z, t) =

H etn

kcn

d, ezn(z, t) =

uz E etn

kcn

d

Similar to the time-domain theory of waveguide filled with losslessmedium [6], the modal voltage and current for TEM mode satisfy theone-dimensional wave equation

2vTEMnz2

1c2

2vTEMnt2

c

vTEMnt

=

c

t

J etnd z

Jm uz etnd

2iTEMnz2

1c2

2iTEMnt2

c

iTEMnt

=

Jmuz

etnd

z

J

etnd+

1

c

t

Jmuz

etnd (34)


16/35

234 Geyi

Once vTEMv (or iTEMn ) is determined, i

TEMn (or v

TEMn ) can be

determined by time integration ofvTEMn (or iTEMn ). The modal voltage

vTEn satisfies the following hyperbolic equation

2vTEnz2

1c2

2vTEnt2

c

vTEnt

k2cnvTEn

=

c

t

J etnd z

Jm uz etnd

+kcn

(uz Jm)uz etn

kcn

d (35)

When = 0, the above equation reduces to Klein-Gordon equation.(35) will be called modified Klein-Gordon equation. The modal currentiTEn can be determined by a time integration of v

TEn /z

iTEn (z, t) = ct

vTEn (z, t)z

dt

c

t

Jm(r, t) [uz etn()]d()

dt (36)

The modal current iTMn also satisfies the modified Klein-Gordonequation.

2iTMnz2

1c2

2iTMnt2

c

iTMnt

k2cniTMn

=

Jm uz etnd z

J etnd

+1

c

t

Jm uz etnd kcn

uz J etn

kcn

d (37)

The modal voltage vTMn can then be determined by a time integrationof iTMn /z

vTMn (z, t) = ct

iTMn (z, t)

zdt c

t

J(r, t) etn()d()

dt

(38)


17/35


4.2. Retarded Greens Function of Modified Klein-GordonEquation

Since the tangential electric field on the electric conductor must bezero, the time-domain voltage satisfies the homogeneous Dirichletboundary conditions

vn(z, t)|z=z1 = vn(z, t)|z=z2 = 0 (39)Making use of the following relation [6]

in(z, t)z

+ kcnhzn(z, t) =1

c

vn(z, t)

t+ vn(z, t)

+

J(, z , t) etn()d()

and considering the boundary condition that the normal component of

the magnetic field on an electric conductor must be zero, the time-domain current must satisfy the homogeneous Neumann boundaryconditions

in(z, t)

z

z=z1

=in(z, t)

z

z=z2

= 0 (40)

In order to solve (34), (35) and (37) subject to the boundary conditions(39) and (40), one may introduce the following retarded Greensfunctions for the modified Klein-Gordon equation

2

z2 1

c22

t2

c

t k2cn

Gvn(z, t; z

, t) = (z z)(t t)Gvn(z, t; z

, t)

|t


18/35

236 Geyi

Fourier transform with respect to time

Gv,in (z, ; z, t) =

Gv,in (z, t; z, t)ejtdt

gives 2

z2+ 2n

Gv,in (z, ; z

, t) = ejt(z z) (43)

where 2n = (/c)2 k2cn j/c. The above equations can be solved

by the method of eigenfunctions, i.e.,

Gvn(z, ; z, t) =

m=1

gvm

2

Lsin

m

L(z z1)

Gin(z, ; z, t) =

m=1

gimmL

cosm

L(z z1)

where L = z2 z1, and m = 1(m = 0), m = 2(m = 0). Substitutingthese into (43) leads to

gvm =1

2n (m/L)2

2

Lsin

m

L(z z1)ejt

gim =1

2n (m/L)2

mL

cosm

L(z z1)ejt

Thus

Gvn(z, ; z, t) =

m=1

12n(m/L)

2

2

L

sinm

L

(z

z1)sin

m

L

(z

z1)e

jt

Gin(z, ; z, t) =

m=0

12n(m/L)2

mL

cosm

L(zz1)cos m

L(zz1)ejt

Taking the inverse Fourier transform

Gv,in (z, t; z, t) =

1

2

Gv,in (z, ; z, t)ejtd

one may obtain

G

v

n(z, t; z

, t

) =

m=1

c2

L sin

m

L (z z1)sinm

L (z

z1)


19/35


ej(tt)d

2 (ckcn)2 (mc/L)2 j/

Gin(z, t; z

, t

) =

m=0

mc2

2L cos

m

L (z z1)cosm

L (z

z1)

ej(tt)d

2 (ckcn)2 (mc/L)2 j/

The integral in the summation can be evaluated by the residue theorem[6]. The results are

Gvn(z, t; z, t) =

m=1

2c

Lsin

m

L(z z1)sin m

L(z z1)

e(tt)

sin c(t t)k2cn + (m/L)2

2k2cn + (m/L)

2 2 H(t t

)

(44)

Gin(z, t; z, t) =

m=0

mc

Lcos

m

L(z z1)cos m

L(z z1)

e(tt)sin

c(t t)k2cn + (m/L)2 2k2cn + (m/L)

2 2 H(t t)

(45)

where = /2. If one of the ends of the waveguide cavity extends toinfinity, say, z2

, the discrete values m/L become a continuum.

In this case, (44) and (45) can be rewritten as

Gvn(z, t; z, t)

z2

= c

e(tt)

0

[cos k(z + z 2z1) cos k(z z)]sin

c(t t)k2cn + k2 2k2cn + k

2 2 dk

Gin(z, t; z, t)

z2

=c

e(tt

)

0

[cos k(z + z 2z1) + cos k(z z)]sin

c(t t)k2cn + k2 2

k2cn + k

2 2 dk


20/35

238 Geyi

where k = /c. These integrations may be carried out by using

0

sin q

x2 + a2

x2 + a2

cos bxdx =

2J0 aq

2 b2H(q b)a > 0, q > 0, b > 0

and the results are

e(tt)Gvn(z, t; z

, t)|z2 = c

2J0

k2cn 2

1/2c2(t t)2 |z + z 2z1|2

H[c(t t) |z + z 2z1|]+

c

2J0

k2cn 2

1/2c2(t t)2 |z z|2

H[c(t t) |z z|]

(46)

e(tt)Gin(z, t; z

, t)|z2 =c

2J0

k2cn 2

1/2c2(t t)2 |z + z 2z1|2

H[c(t t) |z + z 2z1|]+

c

2J0

k2cn 2

1/2c2(t t)2 |z z|2

H[c(t t) |z z|]

(47)

4.3. Solution of Inhomogeneous Klein-Gordon Equation

The retarded Greens functions can be used to solve the modified

Klein-Gordon equation. Consider the inhomogeneous Klein-Gordonequations2

z2 1

c22

t2

c

t k2cn

vn(z, t) = f(z, t), z1 < z < z2

2

z2 1

c22

t2

c

t k2cn

in(z, t) = g(z, t), z1 < z < z2

with the boundary conditions (39) and (40). It is easy to show thatthe solutions of the above equations are


21/35


vn(z, t) = z2

z1

dz

f(z, t)Gvn(z, t; z, t)dt, z (z1, z2)

in(z, t) = z2

z1

dz

g(z, t)Gin(z, t; z, t)dt, z (z1, z2)

(48)

Thus the solutions of (34), (35) and (37) can be obtained from (48) as

vTEn (z, t) =

c

z2z1

dz

t

J(, z, t) etn()d()

Gvn(z, t; z, t)dt

z2

z1

dz

Jm(

, z

, t

) uz etn(

)d(

) Gvn(z, t; z, t)

z dt

kcnz2

z1

dz

uz Jm(, z, t)uz etn()

kcn

d()

Gvn(z, t; z, t)dt (49)

iTMn (z, t) = z2

z1

dz

J(, z, t)etn()d()Gin(z, t; z, t)

z dt

1c

z2

z1

dz

t

Jm(, z, t) uzetn()d()

Gin(z, t; z, t)dt

+kcn

z2z1

dz

uz J(, z, t) etn()

kcn

d()

Gin(z, t; z, t)dt

(50)

In deriving the above expressions it has been assumed that all sourcesare confined in (z1, z2). It should be notified that if the magneticcurrent Jm approaches to z1 or z2 so that it is tightly pressed on theelectric wall z = z1 or z = z2, the time-domain voltage and currentwill not satisfy the homogeneous boundary conditions (39) and (40) atz = z1 or z = z2.


22/35

240 Geyi

5. APPLICATIONS

The time-domain theory developed above may be used to study thetransient process inside a cavity resonator. It will be shown that the

time-domain response of the cavity to an arbitrary excitation waveformturned on at a finite instant of time will be severely distorted. Forexample, the response of a cavity to a sinusoidal excitation turned onat a finite instant of time is not sinusoidal in general, which is totallydifferent from the prediction of time-harmonic theory. To obtain asinusoidal oscillation in the cavity, the frequency of the excitationsinusoidal wave must coincide with one of the resonant frequencies.

5.1. A Shorted Rectangular Waveguide

Let us investigate the transient process in a shorted rectangularwaveguide excited by a line current extending across the waveguidecentered at x = x0 = a/2, z = z0

J(r, t) = uy(x x0)(z z0)f(t) (51)as shown in Figure 3. By the symmetry of the structure and excitation,only TEn0 mode will be excited, which are

etn(x, y) = eTEn0 (x, y) = uy

2

ab

1/2sin

nx

a, n = 1, 2, 3 (52)

with kcn = n/a. It follows from (46) and (49) that

vTEn (z, t) =b

2

2

ab

1/2sin

n

ax0

t|zz0|/c

df(t)

dtJ0

kcnc

(t t)2 |z z0|2/c2

dt

o

TE

nv

TE

ni

o

y

z

0z

y

a

b

xJ

Figure 3. A shorted rectangular waveguide excited by a centeredcurrent source.


23/35


t|z+z0|/c

df(t)

dtJ0

kcnc

(t t)2 |z + z0|2/c2

dt

Assuming that f(t) = H(t)sin t, the time-domain voltage may bewritten as

vTEn (z, t) =b

2

2

ab

1/2ka sin

n

ax0

ct/a|zz0|/a

cos ka(ct/a u)J0

kcna

u2 |z z0|2/a2

du

ct/a

|z+z0|/a

cos ka(ct/a u)J0

kcna

u2 |z + z0|2/a2

du

The time-domain response vTEn (z, t) may be divided into the sum of asteady-state part and a transient part

vTEn (z, t) = vTEn (z, t)

steady

+ vTEn (z, t)transient

where

vTEn (z, t)steady

=b

2

2

ab

1/2ka sin

n

ax0

|zz0|/a

cos ka(ct/a u)J0

kcna

u2 |z z0|2/a2

du

|z+z0|/a

cos ka(ct/a u)J0

kcna

u2 |z + z0|2/a2

du

vTEn (z, t)transient

= b2

2

ab

1/2ka sin

n

ax0

ct/a

cos ka(ct/a u)J0

kcna

u2 |z z0|2/a2

du

ct/a

cos ka(ct/a

u)J0 kcnau2 |

z + z0

|2/a2 du


24/35

242 Geyi

The transient part approaches to zero as t . The integrals in thesteady part can be carried out by use of the following calculations [11]

a

J0

b

x2

a2

sin cxdx =0, 0 < c < b

cos

ac2b2 /c2b2, 0 < b < ca

J0

b

x2a2

cos cxdx =

expab2c2

/

b2c2, 0 < c < b sin

a

c2b2

/

c2b2, 0 < b < c

Thus

vTEn (z, t)steady

=b

2

2

ab

1/2ka sin

n

a

1|(ka)2 (kcna)2|

sin

ka

ct

a |z z0|

a

(ka)2 (kcna)2

sinka cta

|z + z0|a

(ka)2 (kcna)2

, k > kcn

cos

ka

ct

a

exp

|z z0|

a

(kcna)2 (ka)2

cos

kact

a

exp

|z + z0|

a

(kcna)2 (ka)2

k < kcn

In the region 0 < z < z0, the above equation may be rewritten as

vTEn (z, t)steady

=1

2

b

a

1/2sin

n

2

k

n

2sin(nz)cos(t

nz0), k > kcn

cos(t) exp[n(z z0)]cos(t) exp[n(z + z0)], k < kcn

where n = (|k2 k2cn|)1/2. Therefore the time-domain voltage for theTEn0 mode in the shorted waveguide is a standing wave if the operatingfrequency is higher than cut-off frequency of the TEn0 mode, which isa well-known result in time-harmonic theory.

The time-domain current can be determined by (36)

iTEn (z, t) =

2b

a

1/2sin

n

2

12 sin (t |z + z0|/c) 12 sin (t |z z0|/c)


25/35


+

2b

a

1/2sin

n

2

kcn(z+z0)

2

t|z+z0|/c

0

J1 kcnc(t t)2 |z + z0|2/c2

(t t)2 |z + z0|2/c2 sin t

dt

kcn(zz0)2

t|zz0|/c0

J1

kcnc

(t t)2 |z z0|2/c2

(t t)2 |z z0|2/c2

sin tdt

The steady state part of iTEn (z, t) is given by

iTEn (z, t)steady

=

2b

a

1/2sin

n

2

1

2sin (t |z + z0|/c) 1

2sin (t |z z0|/c)

+

2b

a

1/2sin

n

2

kcn(z +z0)

2

|z+z0|/c

J1

kcnc

u2 |z + z0|2/c2

u2 |z + z0|2/c2

sin (t u)du

kcn(zz0)2

|zz0|/c

J1

kcnc

u2 |z z0|2/c2

u2 |z z0|2/c2

sin (t u)du

Assuming that k > kcn and making use of the following calculations

a

sin cxx2 a2Jv

b

x2 a2

dx =

2Jv/2

a

2

c

c2 b2

Jv/2

a

2

c +

c2 b2

a

cos cxx2 a2Jv

b

x2 a2

dx =

2

Jv/2

a

2

c

c2 b2

Nv/2

a

2

c +

c2 b2

(a > 0, 0 < b < c)


26/35

244 Geyi

one may obtain

iTEn (z, t)steady

= 12

b

a

1/22sin

n

2cos(nz)sin(t nz0)

Let VTEn (z) and ITEn (z) be the phasors of vTEn (z, t)|steady andiTEn (z, t)|steady respectively, then

VTEn (z) =1

2

b

a

1/2 kn

2sinn

2sin(nz)e

jnz0 , k > kcn

ITEn (z) = j1

2

b

a

1/22sin

n

2cos(nz)e

jnz0 , k > kcn

Since the current is assumed to be in positive z-direction, theimpedance at z (0, z0) is thus given by

Zn(z) =VTEn (z)

ITEn (z)= j

k

ntan(nz), k > kcn

which is a well-known result and validates the time-domain theory.

5.2. A Rectangular Waveguide Cavity

Let the shorted waveguide shown in Figure 3 be closed by a perfectconducting wall at z = L with L > z0 (Figure 4) and the excitationsource be given by (51). By symmetry, only TEn0 mode will be excited.It follows from (44) and (49) that

vTEn (z, t) =2

L

2b

a

1/2sin

nx0a

m=1sin

m

Lz sin

m

Lz0

(n/a)2 + (m/L)2

df(t)

dtsin

c(t t)

(n/a)2+(m/L)2

H(t t)dt

(53)

Lo

y

z

0z

y

a

b

xJ

Figure 4. A rectangular waveguide cavity excited by a current source.


27/35


Again if it is assumed that f(t) = H(t)sin t, the above expressionmay be written as

vTE

n (z, t) =

2

L2b

a1/2

sin

nx0

a

m=1

sinm

Lz sin

m

Lz0

(n/a)2 + (m/L)2

t

sin

c(t t)

(n/a)2+(m/L)2

cos tdt

= 2kL

2b

a

1/2sin

nx0a

m=1

sinm

Lz sin

m

Lz0

cos kct cos

ct

(n/a)2+(m/L)2

k2 (n/a)2+(m/L)2 (54)

where k = /c. The electromagnetic field is given by (33)

E(r, t) = uyEy =n=1

vTEn (z, t)eTEn0 (x, y)

= uy4k

La

n=1

m=1

sinnx

asin

nx0a

sinm

Lz sin

m

Lz0

cos kct cos

ct

(n/a)2+(m/L)2

k2 (n/a)2+(m/L)2 (55)

0 10 20 30 40 50 600.5

0

0.5

ctoa

yaE

0 0/ ( 0.5 , 0.75 , 0.5 )ct a x a z a x z a= = = =

Figure 5. Electric field excited by a sinusoidal wave f(t) = H(t)sin t

when k = /c = (n/a)2 + (m/L)2 (ka = ).


28/35

246 Geyi

where eTEn0 (x, y) are given by (52). As can be seen from (54) and(55), one cannot separate the response of a closed cavity into atransient part and a steady-state part. Figure 5 shows the normalizedelectric field aE(r, t)/ at x = 0.5a, z = 0.75a, excited by a source

defined by (51) with sinusoidal waveform f(t) = H(t)sin t witha = b = L. It is assumed that k is below any resonant wavenumbers(n/a)2 + (m/L)2 (m, n 1) with k = . It can be seen from

the plot that the electric field does not approach to a pure sinusoidalwave as t . It should be noted that (54), (55) are finite as kapproaches to any of the resonant wavenumbers

(n/a)2 + (m/L)2.

For example, when k approaches to

(/a)2 + (/L)2, the singularterm for (n, m) = (1, 1) in (55) becomes

cos kct cos

ct

(/a)2 + (/L)2

k2 (/a)2 + (/L)2 = ct sin

ct

(/a)2 + (/L)2

2

(/a)2 + (/L)2

(56)

which is a finite number for a finite t.

0 10 20 30 40 50 6040

20

0

20

40

E1 ctoa( )

ctoa

yaE

0 0/ ( 0.5 , 0.75 , 0.5 )ct a x a z a x z a= = = =

Figure 6. Electric field exited by a sinusoidal wave f(t) = H(t)sin twhen k = /c =

(/a)2 + (/L)2.

Figure 6 shows the normalized electric field aE(r, t)/ at x =0.5a, z = 0.75a, excited by a sinusoidal wave f(t) = H(t)sin t whenk =

(/a)2 + (/L)2 with a = b = L. In this case a sinusoidal wave

will be gradually built up as t . Therefore the response of a metalcavity is a sinusoidal wave if and only if the frequency of the excitingsinusoidal wave coincides with one of the resonant frequencies of the

metal cavity.


29/35


If the excitation waveform is a unit step function, i.e., f(t) = H(t),(53) becomes

vTE

n(z, t) =

2

L2b

a1/2

sinnx0

a

m=1

sinm

Lz sin

m

Lz0

sin

ct

(n/a)2+(m/L)2

(n/a)2+(m/L)2

(57)

and the electric field is

E(r, t) = uyEy = uy 4La

n=1

m=1

sinnx

asin

nx0a

sinm

Lz sin

m

Lz0

sin ct(n/a)2+(m/L)2

(n/a)2+(m/L)2 (58)

Figure 7 shows the normalized electric field aE(r, t)/ at x = 0.5a, z =0.75a, excited by the unit step waveform. Note that the response ofthe metal cavity is not a unit step function.

0 1 2 3 4 55

0

5

E2 ctoa( )

ctoa

yaE

0 0/ ( 0.5 , 0.75 , 0.5 )ct a x a z a x z a= = = =

Figure 7. Electric field excited by a unit step waveform f(t) = H(t).

5.3. A Coaxial Waveguide Cavity

A coaxial waveguide cavity of length L consisting of an inner conductorof radius aand an outer conductor of radius b is shown in Figure 8. Let

the coaxial waveguide be excited by a magnetic ring current located at


30/35

248 Geyi

z = z0

Jm(r, t) = uf(t)(z z0)( 0), a < 0 < b, 0 < z0 < Lwhere (,,z) are the polar coordinates and u

is the unit vector

in direction. According to the symmetry, only TEM mode andthose TM0q modes that are independent of will be excited. Theorthonormal vector mode functions for these modes are given by [12]

2a2b

o

a

y

x

b

0L zz

Figure 8. Cross-section of a coaxial waveguide.

kc1 = 0, et1(, ) = ue1(), e1() =1

2 ln c1

kcn =na

, etn(, ) = uen()

en() =

2

na

J1(n/a)N0(n) N1(n/a)J0(n)J20 (n)/J

20 (c1n) 1

, n 2

where c1 = b/a, u is the unit vector in direction, and nis the nth nonvanishing root of the eqauation J0(nc1)N0(n)

N0(nc1)J0(n) = 0. It follows from (45) and (50) that

iTMn (z, t) = 2

cen(0)0

df(t)

dtGin(z, t; z0, t

)dt

= 2L

0en(0)

m=0

mcos

m

Lz cos

m

Lz0

(n/a)2 + (m/L)2

t

df(t)

dtsin

c(t t)

(n/a)2 + (m/L)2

dt (59)


31/35


Let f(t) = H(t)sin t, the above expression may be written as

iTMn (z, t) =

2kc

L

0en(0)

m=0

mcos

m

Lz cos

m

Lz0

(n/a)

2

+ (m/L)2

t0

sin

c(t t)

(n/a)2 + (m/L)2

cos tdt

=2k

L0en(0)

m=0

m cosm

Lz cos

m

Lz0

cos kct cos

c(t t)(n/a)2 + (m/L)2

k2 (n/a)2 + (m/L)2 (60)

0 10 20 30 40 50 602

1

0

1

2

Hc ctoa(

ctoa

0 0/ ( 2 , 1.5 , 3 , 1.5 , 2 )ct a b a a z a a z a = = = = =

aH

Figure 9. Magnetic field exited by a sinusoidal wave f(t) = H(t)sin twhen ka = 3 (k = (n/a)2 + (m/L)2).

The magnetic field H(r, t) = uH in the coaxial waveguidecan be determined from (33). Figure 9 shows the magnetic field at = (a + b)/2 and z = 3a when ka = 3 with the assumption thatb = 2a, L = 2b, 0 = (a + b)/2, z0 = L/2. In this case k is not equal

to any resonant wavenumbers

(/a)2 + (m/L)2 (m, n 1) and thefield response is not a sinusoidal wave. When the frequency of theexcitation waveform approaches to one of the resonant frequencies, sayka = 2, a sinusoidal wave will gradually build up inside the coaxial

waveguide cavity as shown in Figure 10.


32/35

250 Geyi

0 10 20 30 40 50 6010

5

0

5

10

Hc ctoa(

ctoa

0 0/ ( 2 , 1.5 , 3 , 1.5 , 2 )ct a b a a z a a z a = = = = =

aH

Figure 10. Magnetic field exited by a sinusoidal wave f(t) =H(t)sin t when ka = 2 = 3.123.

0 1 2 3 4 53

2

1

0

1

Hc ctoa(

ctoa

0 0/ ( 2 , 1.5 , 3 , 1.5 , 2 )ct a b a a z a a z a = = = = =

aH

Figure 11. Magnetic field exited by a sinusoidal wave f(t) = H(t).

If the coaxial waveguide cavity is excited by unit step waveformf(t) = H(t), (59) may be written as

iTMn (z, t) = 2

L0en(0)

m=0

mcos

m

Lz cos

m

Lz0

(n/a)2 + (m/L)2

sin

ct

(n/a)2 + (m/L)2

The field response is shown in Figure 11, and is no longer a unit stepwaveform.


33/35


6. CONCLUSION

This paper provides a thorough discussion on the time-domain theoryof metal cavities. A rigorous proof of the completeness of the vector

modal functions inside an arbitrary metal cavity has been presented onthe basis of the theory of symmetric operators. The major topic of thispaper is the waveguide cavity resonator, which has been examined byusing the time-domain theory of the waveguide developed previously.The fields inside the waveguide cavity have been expanded in terms ofthe vector modal functions of the corresponding waveguide, and thefield expansion coefficients are shown to satisfy the modified Klein-Gordon equations subject to the homogeneous Dirichlet or Neumannboundary condition, which can then be solved by the method ofretarded Greens function. Such an approach guarantees that thesolutions satisfy the causality condition. Some numerical exampleshave been presented to demonstrate the time-domain theory. It hasbeen shown that, for a closed cavity, the time-domain response to a

sinusoidal waveform turned on at a finite instant of time is generally notsinusoidal even when the time tends to infinity. A sinusoidal responsecan develop inside the cavity only when the frequency of the excitationsinusoidal wave coincides with one of the resonant frequencies of themetal cavity.

The traditional time-harmonic theory, however, always yields asinusoidal response if the excitation waveform is sinusoidal. Thereforethe time-harmonic theory fails to give a correct theoretical predictionfor a closed cavity whenever the source is turned on at a finite instantof time which corresponds to all practical situations. In addition thesingularities occur in the time-harmonic theory for a lossless metalcavity if the frequency of the excitation source coincides with one ofthe resonant frequencies. In this case, the field distributions are infinite

everywhere inside the cavity, leading to an awkward situation. On theother hand, the time-domain theory always gives a finite solution andis thus more appropriate to describe what happens in a closed metalcavity, which is the foundation for a number of important applicationsin microwave engineering and is also helpful in studying various cavity-related problems [1320].

REFERENCES

1. Kurokawa, K., An Introduction to the Theory of MicrowaveCircuits, Academic Press, New York, 1969.

2. Goubau, G. (ed.), Electromagnetic Waveguides and Cavities,

Pergmon, London, 1961.


34/35

252 Geyi

3. Collin, R. E., Foundations for Microwave Engineering, 2ndedition, IEEE Press, 2001.

4. Aksoy, S. and O. A. Tretyakov, The evolution equations in studyof the cavity oscillations excited by a digital signal, IEEE Trans.

Antennas and Propagation, Vol. 52, No. 1, 263270, 2004.5. Bladel, J. V., Electromagnetic Fields, 1st edition, Hemisphere

Publishing Corporation, 1985.

6. Geyi, W., A time-domain theory of waveguide, Progress InElectromagnetics Research, PIER 59, 267297, 2006.

7. Zeidler, E., Applied Functional Analysis-Applications to Mathe-matical Physics, Springer-Verlag, 1995.

8. Byron, F. W. and R. W. Fuller, Mathematics of Classical andQuantum Physics, Addison-Wesley, 1969.

9. Colton, D. and R. Kress, Integral Equation Methods in ScatteringTheory, John Wiley, 1983.

10. Harrington, R. F., Time-Harmonic Electromagnetic Fields,

McGraw-Hill Book Company, Inc., 1961.11. Grandshteyn, I. S. and I. M. Ryzhik, Table of Integrals, Series,

and Products, Academic Press, New York, 1994.

12. Marcuvitz, N., Waveguide Handbook, McGraw-Hill, 1951.

13. Bopp, III, C. L. and C. M. Butler, Analysis of transmissionof a signal through a complex cylindrical/coaxial cavityby transmission line methods, Progress In ElectromagneticsResearch, PIER 56, 3351, 2006.

14. Mohsen, A. A., E. M. Elkaramany, and F. G. A. El-Hadeed,Analysis of microwave cavities using LTL-FD method, J. ofElectromagn. Waves and Appl., Vol. 19, No. 2, 145162, 2005.

15. Yang, D., C. Liao, and W. Chen, Numerical solution on coupling

of UWB pulse into a rectangular cavity through slots, J. ofElectromagn. Waves and Appl., Vol. 19, No. 12, 16291638, 2005.

16. Kim, J. H. and H. J. Eom, Radiation from multiple annular slotson a circular cylindrical cavity, J. of Electromagn. Waves andAppl., Vol. 21, No. 1, 4756, 2007.

17. Xiao, J.-K., W.-S. Ji, S. Zhang, and Y. Li, A field theoreticalmethod for analyzing microwave cavity with arbitrary cross-section, J. of Electromagn. Waves and Appl., Vol. 20, No. 4,435446, 2006.

18. Zhang, J.-P. and K.-M. Chen, Mode matching analysis ofthe induced electric field in a material sample placed withinan energized cylindrical cavity, Progress In Electromagnetics

Research, PIER 28, 295311, 2000.


35/35


19. Van Rienen, U., Frequency domain analysis of waveguide andresonator with fit on non-orthogonal triangular grids, ProgressIn Electromagnetics Research, PIER 32, 357381, 2001.

20. Wang, C.-F., Y. Xu, and Y.-B. Gan, 3-dimnensional implemen-

tation of the field iterative method for cavity modeling, ProgressIn Electromagnetics Research, PIER 47, 2747, 2004.

w. geyi- time-domain theory of metal cavity resonator

Documents