Upload
nag-challa
View
223
Download
0
Embed Size (px)
Citation preview
8/9/2019 Samar as 2004
1/3
A Note on the Impedance Variation with
Feed Position of a Rectangular
Microstrip Patch Antenna
T. Samaras
A. Koulog lou
and
J . N
Sahalos
Radio comm unication s Laboratory, Departme nt of Phys ics. Aristotle University
of
Thessaloniki
GR-54124 Thessa lonik i , Gre ece
Tel: t30 231 0 998161; Fax: t3023 10 9980 69; E-mail : saha los@auth .gr
Abstract
The variation with fee d position of th e input impedance of a rectangular patch
antenna
is investigated theoretically. Two differ-
ent f e e d types are examined: an
inset
microstrip
line,
a n d a coaxial probe. Th e Finite-Difference Time-Domain (FDTD) tech-
nique
is
u s e d for
the
calculations.
Numerical
results are com pared with published
measurements
a nd o the r theoretical mod-
e l s .
Keywords: Impe dance ; microstrip antennas; antenna f e e ds ; coaxial
coupler
1. In t roduc t ion
icrostnp-patch antennas have been widely used in applica-
tions where lo w-profile antennas are needed. The input
impedance of a microship antenna depends
on
its geometrical
shape and dim ensions, the physical pro perties of the materials
involved, and the feed type. The transmiss ion-line model [ l ] pre-
dicts that the input resistance of a pro he-fed rectangular patch
antenna is proportional to the cosine-square d
(
cos2) of the normal-
ized feed-point distance from the patch edge. This depen dence can
also be seen in [Z], where measured values of resistance were
compare d with calculations based on modal analysis for different
positions of the probe feed. Using the Method of Moments (M UM ),
similar res ults were obtaine d in
[3]
for three width laspect ratios of
the rectan gular patch. However, recent work [4] has shown that
when the patch is fed with an inset microstrip line, the resistance
90
n/I.
dependen ce becomes proportional to the fourth power of the cosine
cos4 ), although no theoretical justification was given there for
this result. This short note presents numerical calculations that
confirm the different behaviors of the
two
feed types, as shown in
[41.
2. Method
The results were obtained with
SEMCAD
(Schmid and
Partner Engineering
AG,
Zeughausstrasse
43 8004
Zurich, Swit-
zerland), which implements the Finite-D ifference Time-Domain
(FDTD) technique
[ 5 ]
The antenn a modeled was the one measured
in
[4] ;
the dimensions are shown in Figure 1. The distance s was
chosen to he equal to the width of th e microstrip feed 3.8 mm). In
both configuration s, the impedance was calculated as function of
IEEEAnfennas and Propagation Magazine
Vol.
46
No
Apri l 2004
.
8/9/2019 Samar as 2004
2/3
14cm
W=5.94cm
Figure l a. The dimensions of the antennas examined: micro-
stri p inset feed.
w = 5 . ~ c m I
~
. . . . . . . . . . . . . . . . . . . . . .
I
. . .
1
. .
. . . . .
..
....
. . . . . .
. .
Figure lh. The dimensioiis of the antennas examined: probe
feed.
y o , where yo varied
fiom
zero (at the edge
of
the patch) to
L 2
= 20.2 mm
(at the center of the patch). The substrate was
1.27
mm thick, and had dielectric properties of ta n6
=
0.0019 and
r = 2.42 .
The antenna was d esigned to work at 2.3 GHz.
A non-uniform grid was used in the model. The maximum
cell size (near the boundaries of the computational domain) was
2 mm Af6 .5) . The Gaussian pulse that excited the antenna was
cehtered at 2.3 CHz, hut had harmonics up to 4
GHz.
Therefore,
iEEEAnfenn as and Propagation Magazine
Vol.
46 No. 2 April 2004
the maximum ccil size corresponded to about
1/26
of thc minimum
wavelength. The computational domain was truncated with a six-
cells-thick perfectly matched layer
(PML).
3. Results
and
Discuss ion
The variation of the input impedance with the feed position
is
shown in Figure 2 for the two feed types. The values of the red s-
tancc for the microstrip inset feed followed the
cos4
variation, and
those for the probe fecd followed the cos2 variation
[I],
normal-
ized at 2y0/L
= 0.25.
The results for the resistance compared well
with the measurements of
[4]
specially for feed positions close to
the center of the patch. The agreement of the reactance values was
also good.
One would expect that symmetrical excitation of the patch
( y o = L I Z ) should lead to zero resistance, since
11
mode can he
excited. However, experimental [4] and numerical rcsults (this
letter) showed that the input still resistance takes
some
value in
this casc, since the symmetty of the patch is destroydd by the inset.
Thus, a mode
is
produced that makes the patch radiate. The trans-
. .
........
-
.
0 0 2 0.4
0.6
0.8 1
normalised d istance, 2y /L
Figur e 2a. The variation of th e inpu t impedance with feed posi-
tlon for the antenn a with a microstrip inset feed:
0 FDTD,
X C2) ; A
Measured
141 R C2) ; A
Measured
141
a);
Transmission-line model, [ 6 ]
R
C l ) .
- c o s ~ ( A Y ~ / L ) ; - - -
C O S * ~ ~ ~ / L ) ;FDTD,
R R);
Figure 2b. The variation of the inp ut impedaiicc with feed posi-
tion for th e antenn a with a pro be feed (legend as in Figure 2a).
91
8/9/2019 Samar as 2004
3/3
Table 1 Results for the microstrip-fed antenna.
0.50
0.65
Ta b le 2. Results fo r the probe-fed antenna.
mission-line model [6] fails to predict this phenomenon (Table
I .
Mor eover, the values it gives follow the cos2 law (F igure 2).
In
the c, ae of the probe feed at the center of the patch, the sym metry
was respected and the input impedan ce was zero (Table 2).
The results confirmed that the variation of the input imped-
in ce with the feed position is different, depending on the feed type.
4.
References
1. C.
A.
Balanis, Advanced Engineering Elecfromagnefics,New
York, John Wiley Sons, 1989.
2. W.
F:Richards, Y. T.
Lo,
and D. D. Harrison, “An Improved
Theory for Microship Antennas and Applications,”
IEEE Trans-
actions on Antennas and Propagation, AP-29, 1 1981, pp. 38-46.
3. J. R. M osig, R. C. Hall, and
F.
E. Gardiol, “Numerical Analysis
of M icrostrip Patch Anten nas,” in J. R. James and P. S Hall (eds.),
Handbook of Micro sfrip Antennas,
London, P eter Peregrinus Ltd.,
1989.
4. L.
I.
Basilio, M. A. K hayat, J. T. W illiams and
S.
A. Long, “The
Dependence of the Input Impedance on Feed Position of Probe and
Microstrip Line-Fed
Patch Antennas,” IEEE Transacfions on
Antennas and P ropagafion,AP-49, 1,200 1, pp, 45-47.
6. A. Taflove, CompufafionalElecfrodynamics: The Finife-Differ-
ence Time-Domain Method,
Norwood, MA, Artech House, 1995.
7. R. A. Sainati,
CAD OfMicrostrip Antennasfor Wireless Appli-
cations, Norwood, MA, Artech House Inc., 1996.
Ideas
for
Antenna Designer’s Notebook
Ideas are needed for future issues of
the
Antenna Designer’s
Notebook. Please send your suggestions to Tom Milligan and
they will be considered for publication
as
quickly as possible.
Topics c an include antenna design tips, equations, nomographs,
or shortcuts, as well as ideas to improve or facili tate
measurements. @
92
IEEE
Antennas and
Propagalion Magazine
Vol.
46 o. 2, April 2004