Microstrip Applications (1) - Depends Geometrical Practical Adjust Input Patch Function

Download as pdf or txt
Download as pdf or txt
You are on page 1of 4

A Study of the Input Impedance of the Inset-Fed Rectangular Microstrip

Antenna as a Function of Notch Depth and Width


Ying Hu*, Erica Lundgren, David R. Jackson, Jeffery T. Williams,
and Stuart A. Long
Department of Electrical and Computer Engineering
University of Houston, Houston, TX 77204-4005
Introduction
Microstrip antennas have been widely used in modem applications where size,
cost, performance, ease of installation and aerodynamic profile are constraints [1].
Among different types of microstrip antennas, rectangular patch antennas are
most commonly used. The input impedance of the rectangular patch antenna
depends on its geometrical size, dielectric substrate, and feed type.
A practical way to adjust the input impedance of a rectangular patch antenna is to
vary the feed position. The input impedance of a coaxial probe-fed patch antenna
varies as a cos2 function of the normalized feed distance from the patch edge. The
inset-fed patch provides a method of impedance control with a planar feed
configuration [2-4]. However, recent experimental and numerical results showed
that the input impedance of an inset-fed rectangular patch varied as a cos
function of the normalized inset depth [2]. The goal of this paper is to further
study the inset-fed patch in order to examine the dependence of the input
impedance on the inset notch geometry, as well as to realize design tables that
show the notch depth to obtain the best input match (minimum IS, II) for the inset
patch when using a 50 Q feed.
Patch Geometries
The two models shown in Fig. 1 were investigated. Figure l(a) shows the inset-
fed patch, where the notch is located symmetrically along the width of the patch.
In the C-shaped patch shown in Figure 1 (b), the microstrip line feed is replaced
by a coaxial-probe feed. The purpose of this geometry is to aid in the
investigation of input impedance by establishing (approximately) what the direct
input impedance at the feed point is, eliminating any de-embedding issues
introduced from the inset-fed patch, where interaction between the feed line and
radiating patch disturbs the transmission line behavior in the notch region. The
normalized feed distance is defined as X,K, = xO /(L /2). The substrate of the patch
has a relative permittivity £r of 2.42 and a thickness h of 1.27 mm. To provide a
50 Q characteristic impedance, the width of the microstrip line Wf was kept at
0.38 cm, and the inner and outer radius of the 50 Ql coaxial probe for the "C-
patch" was fixed at 0.635 mm and 2.324 mm, respectively. For the C-patch
investigation, the dimension W was kept at 5.94 cm and L at 4.04 cm (WIL 1.5).
The simulation process was done by the finite-element code HFSS.

0-7803-8883-6/05/$20.00 ©2005 IEEE

330
L L

Ww coaox al
feed probe

Figure l(a). Geometry Figure l(b). Geometry of


of the inset-fed patch. the probe-fed C-patch.

Probe-Fed C-patch
The C-patch investigation studied the behavior of the resonant input resistance of
the probe-fed C-patch shown in Fig. 1(b) with changing notch depth and width.
The resonant frequency is defined by the frequency that corresponds to a
maximum input resistance. A special case of the probe-fed C-patch is the probe-
fed rectangular patch, where S approaches zero. A cos2 function characterizes the
resonant resistance of a probe-fed C-patch versus notch depth in this limiting
case.

Figure 2 plots the simulation results for input resistance Rin versus normalized
distance xnOn for C-patches with different notch widths S. In contrast to the
prediction from the pure cos2 function, Figure 2 indicates that Ri, touches zero
before xn.n approaches one. To fit these curves, a shifted sin2 function, as shown
in formula (1), was introduced, where the parameters A and x(°O,, are dependent
on notch width S. As S goes to zero, x(°) approaches 1.0 and formula (1) becomes
the pure cos2 function that governs the probe-fed rectangular patch.

R Asin2[-(xn-x(o))] (1)

Figure 3 illustrates the curve-fit results using formula (1) with A = 162.93 and
x-()= 0.90 for the S/Wf = I case. Very good agreement can be observed from
the plot. Further investigations showed that formula (1) also worked well for the
S/Wf= 2, 3 and 4 cases by using different combinations of A and xn(_) values.

331
200
r"150
U)
-0-S=Wf
-*- S = 2Wf
E 100 -*- S = 3Wf
0 E+S = 4Wf
.' 50
0
0 0.5 1
Xnorm
Figure 2. Resonant input resistance Ri, vs normalized feed
position xn,, for probe-fed C-patch with S / Wf = 1, 2, 3, and 4.
The next investigation kept the notch depth xo fixed at 1.01 cm and varied the
notch width S. As indicated by Fig. 4, the resonant resistance Ri, decreases as the
notch becomes wider. This indicates that for wider notches, the feed point needs
to be inset less into the patch in order to match to a particular impedance level.

200 X HFSS_Simuiation I-I au


E 10S0ifted sinA2 fit E
0 100 n40
v

0 0.5 1 0 2 S/V, 6 8
Xnorn
Figure 3. Comparision between Figure 4. R,, vs. S/Wf for the probe-
simulated Rin vs. x,,r and a shifted sin2 fed C-patch with xo = 1.01 cm.
curve for the probe-fed C-patch with
S/Wf = 1.

Design Table for the Inset-Fed Rectangular Patch


An inset-fed patch was chosen with the same dimensions as before
(corresponding to W/L = 1.5) and a 50 Q feed line, on the same substrate as in the
previous section. Table I shows the inset depth that gives the minimum IS, for
the feeding microstrip line, the value of the minimum IS,I5, and the frequency at
which the minimum IS,, occurs. The phase of 5S1 and the input impedance,

332
calculated using the closest edge of the corresponding rectangular patch as a
reference plane, is also shown in the table. Once again, the simulation package
HFSS was used for the calculation. The inset patch with a length of feeding
microstrip line was simulated, and the input impedance was de-embedded to the
edge of the patch assuming an ideal 50 Q line. The optimum inset depth decreases
as the notch width increases, as expected from the C-patch results.
Table 1. Design table showing the inset depth x0 to give minimum IS,II.
| x0 SI I arg(Si 1) Rin Xin f
S/ Wf [cm] dB] [Deg] [Ohms] [Ohms] [GHz
2.00 1.12 -33.49 150.68 48.59 1.02 2.26
3.00 0.94 -34.72 116.58 49.65 1.65 2.25
5.00 0.72 -34.68 -140.82 49.12 -1.13 2.31

Conclusions
For the probe-fed C-patch, the dependence of the resonant input resistance on the
notch depth can be characterized by a shifted sine-squared function. For a fixed
the notch depth, the resonant input resistance decreases as the notch becomes
wider. For the inset-fed patch, a design table was given for an inset-fed
rectangular patch with one particular aspect ratio of 1.5, which shows the
optimum notch depth to obtain the best possible match (minimum ISIII) to a
feeding 50 Q microstrip line.

Acknowledgements
This work was supported in part by the National Science Foundation through
Grants EEC-0240889 and ECS-0218629 and by the State of Texas Advanced
Technology Program.

References
[1] Constantine A. Balanis, Antenna Theory: Analysis and Design. New York: John
Wiley & Sons Inc., 1997.
[2] L. I. Basilio, M. A. Khayat, J. T. Williams and S. A. Long, "The Dependence of the
Input Impedance on Feed Position of Probe and Microstrip Line-fed Patch
Antennas," IEEE Trans. Antennas and Propagation, Vol. AP-49, pp. 45-47, Jan.
2001.
[3] M. Ramesh and K. B. Yip, "Design Inset-Fed Microstrip Patch Antennas,"
Microwaves & RF," pp. 64-72, Dec. 2003.
[4] T. Samaras, A. Kouloglou, and J. N. Sahalos, "A note on the impedance variation
with feed position of a rectangular microstrip antenna," IEEE Antennas and
Propagation Magazine, Vol. 46, pp. 90-92, April 2004.

333

You might also like