Content uploaded by Kaagita Venkatramana
Author content
All content in this area was uploaded by Kaagita Venkatramana on Dec 26, 2018
Content may be subject to copyright.
Indian Journal of Radio & Space Physics
Vol 43, February 2014, pp 124-129
Design and fabrication of crossed Yagi antennae for dual frequency satellite signal
reception at ground
N Dashora1,$,*, K Venkatramana2 & S V B Rao2
1National Atmospheric Research Laboratory, Gadanki 517 112, India
2Department of Physics, S V University, Tirupati 517 502, India
$E-mail: ndashora@narl.gov.in
Received 5 July 2013; revised 14 August 2013; accepted 23 August 2013
Design and fabrication of two crossed dipole Yagi antennae have been accomplished with an aim of estimation of total
electron content (TEC) through Faraday rotation technique from satellite signal in VHF/UHF band received at ground. The
design simulations for 3 and 5 elements crossed Yagi were performed by multiple iterations and optimization through a
latest version of an established numerical electromagnetic code (NEC). First, design parameters were calculated and then
optimization schemes were tailored to achieve the requisite performance from the crossed dipoles. Some of the most
important aspects for Yagi antenna, like radiation characteristics for forward and backward gain, input impedance,
bandwidth, front to back ratio and voltage standing wave ratio (VSWR) of a typical 3 and 5-element antenna have been
analyzed. Following the successful software design, a step-wise precise hardware fabrication was taken up. Mechanical
drawings were made for achieving mm level accuracy in fabrication. Also, thin aluminum tubes were used that matched
precisely with the designed parameters and lengths of elements. The impedance matching in hardware fabrication was
achieved by connecting Baluns of exact calculated lengths and of particular cable type. Finally, RF characterization was
performed. Both the crossed Yagi antenna passed the basic tests that could be crossed checked with software design. The
results from each stage of this process have been given with analysis and elemental details. This paper will be a practical
guide for in-house fabrication and also a ready reference for interested researchers and students for making a satellite signal
receptor in VHF/UHF band at ground.
Keywords: Crossed Yagi antenna, Total electron content
PACS No.: 84.40.Ba
1 Introduction
Measurement of total electron content (TEC) of the
ionosphere using beacon satellite signals has been in
vogue for almost three-four decades now. Globally
and regionally, TEC studies gained a new impetus
with advent of Global Navigation Satellite Systems
(GNSS) that has modernized the TEC measurements,
not to mention its accuracy and simultaneous global
observations. Due to much higher altitudes of GNSS
satellites, the column integrated TEC comes as a sum
of ionospheric and plasmaspheric electron densities.
The variations in plasmaspheric TEC are drastically
different than the ionospheric TEC (highly variable).
The plasmaspheric TEC remains mostly stable than
during geomagnetic storms. Therefore, during space
weather events and storms, there remains a need of
another source of measuring ionospheric TEC to
accurately ascertain and remove the plasmaspheric
TEC from GNSS-TEC. Polar orbiting satellites with
an altitude more than 550-600 km are appropriate
tools and have been found suitable for ionospheric
tomography based experiments1. A user needs to have
an antenna and back-end electronics to receive signals
from such satellites.
Many beacon satellites transmit signal in
VHF/UHF band with two frequencies around 130-140
and 400-450 MHz for various purposes (like beacon
for ships, weather information satellites etc). A Yagi-
Uda (hereafter, termed as Yagi) antenna is best suited
for satellite signal reception at ground when compared
to axial mode helical and parabolic antennae. A
quadri-filler-helix (QFH) antenna2 also serves as an
alternate to Yagi. But, due to suppressed
hemispherical radiation pattern, low gain at lower
elevation angles and presence of nulls due to
practically imperfect fabrication of QFH antenna,
Yagi is preferred choice for a ground station.
Moreover, Yagi provides larger directive gain which
is best suitable for weak signals. For Yagi, a rotor
based framework is only additional requirement for
tracking a satellite. Yagi is also preferred with due
consideration of comparable factors, like mutual
DASHORA et al.: CROSSED YAGI ANTENNA: DESIGN & FABRICATION
125
independence of the crossed elements of the Yagi
while measuring the signal strength in two
perpendicular directions (for Faraday rotation
measurement), the beam width, the cross and co-
polarization levels, portability and reproducibility3.
Moreover, the design and fabrication of Yagi antenna
is mechanically simple and it can be accomplished in
laboratory environment (without industrial support).
One of the objectives of the present work is to
contribute towards a ground station for receiving
signals in VHF/UHF band from PRATHAM satellite4
at National Atmospheric Research Laboratory
(NARL), Gadanki. Nevertheless, on a broader
perspective, this work provides a step-wise procedure
for design and fabrication of crossed Yagi antenna for
an interested researcher. Although, designing and
fabrication of Yagi antenna is an established
techniques but the classical approach of such designs
follow standard tables and graphs5 for estimation and
optimization of lengths and spacing of various
elements. Chen & Cheng6 introduced optimization
schemes for Yagi and later Burke & Poggio7
developed now famous, generalized numerical
method and computer codes for any target antenna
design. The accessibility of these very complex
computer codes in public domain and handling
various optimization schemes has hindered a
beginner’s interest in building an antenna. Moreover,
variety of applications required flexible and
accessible tools for multi-element, multi-frequency
optimized Yagi design.
For the purpose of measurement of Faraday
rotation and thence, TEC using satellite signals at 145
and 437 MHz, the authors could not find a guiding
paper. Therefore, a procedure for making Yagi antenna
for aforementioned purpose has been developed and
provided. The rotation of the planes of polarization of
the signals, which have passed through the
ionosphere, can be measured by crossed Yagi
antennae at the ground station8. Since the initial phase
(rotation) of the polarization is unknown, two
frequencies are used. Then, the dispersive medium
like ionosphere would induce different rotations on
these two frequencies. Finally, the differential
Faraday rotation between ∆φf1 and ∆φf2 provides a
direct measure of the TEC, where the mean magnetic
field becomes a constant quantity for a given altitude
of ionospheric layer (~300 km). Measurement of
Faraday rotation of the radio signal has been a topic
of intense research since 1970’s, wherein beacon
signals from BE and ATS series9 of satellites were
used to obtain ionospheric TEC. The differential
Faraday rotation will be measured using two crossed
Yagi antennae designed and developed for this
purpose.
2 Design and Optimization of antenna
The fundamental antenna design parameters that
determine the characteristics of antenna are gain, band
width, front-to-back ratio, beam width, return loss,
radiation pattern, input impedance and directivity10.
Each of these parameters was simulated for a Yagi
antenna.
2.1 First estimate of elements of Yagi
The first estimate of lengths and spacing of
reflector, driven and director elements are obtained
from the standard formulas as given by Prasad3. But
in practical sense, these lengths may not meet all the
constraints imposed by an application. Hence,
optimization of these lengths is of prime importance
as a follow up. Table 1 gives the first length estimates
for 3-element and 5-element antennae for respective
crossed Yagi at 145 MHz (λ1=206.896 cm) and 437
MHz (λ2=68.649 cm).
2.2 Software design and optimization of antenna parameters
The numerical methods and computer codes
developed by Burke & Poggio7 were based on
numerical solution of electromagnetic field integrals for
thin, perfectly conducting wire segments using the
method of moments. It was called numerical
electromagnetic code (NEC). Such segments can be
freely arranged in three-dimensional space and excited
in different ways. Different analysis functions were
built-in to calculate the electromagnetic properties of
antennas (e.g. input impedance, current distribution or
Table 1 — First length estimates for 145 MHz
(3-element) and 437 MHz (5-element) antennae
Element/spacing Multiplier
factor for
respective λ
Lengths for
145 MHz, cm
Lengths for
437 MHz, cm
A. Reflector 0.48 99.310 32.951
B. Driven /Dipole 0.46 95.172 31.578
C. Director 1 0.44 91.034 30.205
D. Director 2 0.44 NA 30.205
E. Director 3 0.44 NA 30.205
F. Reflector spacing 0.22 45.517 15.102
G. Director spacing 0.16 33.103 10.983
H. Dipole arm spacing
0.02 04.137 01.372
INDIAN J RADIO & SPACE PHYS, FEBRUARY 2014
126
radiation pattern). The computer codes written in
FORTRAN were later released in public domain in late
1990s. Now the same code is available in many
different versions for PCs and UNIX platforms
augmented with visualization. The Windows based
software called 4NEC2 (version 5.8.7) developed by a
Dutch radio amateur is available and is obtained from
http://www.qsl.net/4nec2/. The 4NEC2 provides a very
general simulation environment for designing,
optimization and 3-D visualization of various types of
antennae and radiation patterns. A user shall be
knowledgeable enough (in theory of NEC and thence
skill of using it) for a very specific design of antenna,
e.g. crossed Yagi in present case.
No step-wise procedure is available in literature for
design and optimization of Yagi using NEC. The
authors devised step-wise procedure as an outcome of
multiple efforts and test trials. This procedure has
been shared for any interested user. It is first
demonstrated for 437 MHz and the same is
reproduced for 145 MHz Yagi antenna.
2.2.1 Design of 437 MHz Yagi antenna (5 elements)
2.2.1a Basic element design input (unoptimized)
(i) Selection of frquency (437 MHz), wire and
metal (aluminium in this case).
(ii) Take lengths of elements from Table 1 and
creat variables for respective elements
symbols. Assuming a cartesian coordinate
system at the centre of dipole on boom, creat
geometry for all the defined elements.
(iii) Add voltage source to dipole and free space as
ground.
(iv) Generate output of the design. Then select far
field pattern for a full 3-D at resolution of 5
degrees.
(v) Figure 1 shows a basic radiation pattern as the
first product from formula based calculation.
It may be noted that this pattern is not
optimized for desired performance.
2.2.1b Optimization
(i) Open optimization window to provide desired
performance parameters.
(ii) For present case, the following values of design
parameters were set. SWR=100 with target
value 4, Gain=100 with target value 10 dB, F/B
ratio=50 with target to maximize, F/R=0 with
target to maximize, R-in=100 with target=200,
X-in=100 with target to minimize, Rad=0 with
target to minimize, Theta value=90 and 90 and
Phi value=360 and 180 and finally Resolution=5
degrees.
(iii) Start optimization process and after few seconds
(not more than 100 sec) the optimized output is
provided. This output may not match with
required values set in step (ii) above.
(iv) Repeat the optimization till the desired output is
obtained. A user can check this process through
the calculated results, variable sensitivity and
variable values for reference.
Fig. 1 — 3-D (left) and 2-D (right) radiation patterns from un-optimized design of 437 MHz antenna
DASHORA et al.: CROSSED YAGI ANTENNA: DESIGN & FABRICATION
127
(v) Once desired results are obtained, upload the
NEC file. This would update all the variables
and save the final design in new file.
2.2.1c Results of Optimization
The process of optimization modifies the lengths and
spacing of the elements. A comparison is given in Table 2.
The final value of the design parameters as obtained
from 4NEC2 software are SWR=1; GAIN=9.94 dB;
resistance=200 ohm and X-in (impedance)=0.23 ohm.
The final radiation patterns created from optimized
output are shown in Fig. 2. A direct comparison of Fig. 1
and Fig. 2 is possible now, wherein one could observe a
substantial enhancement in forward radiation pattern and
total gain when the design is optimized.
A very important plot given in Fig. 3 is obtained
from the 4NEC2 software that shows variation of
reflection coefficient with respect to change in
frequency. It shall be noted that reflection coefficient
(or VSWR) is tending to be minimum around the
central frequency of 437 MHz. This signifies that the
design of antenna is almost perfect with regard to
losses due to reflection within desired frequency band.
2.2.2 Design of 145 MHz Yagi antenna (3 elements)
The step-wise procedure as given above was
repeated except changing central frequency to 145
MHz and creating variables accordingly. For brevity,
the comparison of unoptimized and final lengths of
elements is given in Table 3. The plot of reflection
coefficient vs frequency is given in Fig. 4 to show the
performance of design. The final values of design
parameters are SWR=1; Gain=7.3 dB; resistance=200
ohm and X-in (impedance)=0. 3 ohm.
3 Hardware fabrications of antennae
3.1 Mechanical drawing
The driven element is given as a wire in the 4NEC2
simulations. However, according to a typical Yagi
design, total length of the driven element has to be
folded into a dipole with finite spacing between arms
and spacing between feed points at the ends. The curved
folding on both sides of dipole has to be symmetric and
calculated (not shown here). The spacing between feed
points is taken equal to 2.0 cm. Figure 5 shows a general
mechanical drawing annotated by length of respective
elements given in Tables 2 and 3. All elements are
placed on a boom of length equal to 50.0 cm (for 437
MHz) and 90.0 cm (for 145 MHz) and diameter equal to
2.0 cm. The diameter of all the antenna elements is
equal to 0.6 cm. A nylon cube couplers of dimensions
Table 2 — Optimized performance comparison for 437 MHz
Element/spacing Lengths for
437 MHz, cm
before optimization
Lengths for
437 MHz, cm
after optimization
A. Reflector 32.951 32.872
B. Driven /Dipole 31.578 30.577
C. Director 1 30.205 28.802
D. Director 2 30.205 28.802
E. Director 3 30.205 28.802
F. Reflector spacing 15.102 15.144
G. Director spacing 10.983 10.852
H. Dipole arm spacing 01.372 02.000
Fig. 2 — 3-D (left) and 2-D (right) radiation patterns from optimized design of 437 MHz antenna
INDIAN J RADIO & SPACE PHYS, FEBRUARY 2014
128
40 × 40 × 25 mm has been selected considering its
insulation properties.
3.2 Calculation for balun
The balun length depends upon wavelength (λ) of
the signal and velocity factor of the cable:
* velocity factor
Balun length
2
λ
=
The velocity factor depends upon the dielectric
constant of the cable wire used for balun. The
dielectric constant is the ratio of the permittivity of a
substance to the permittivity of free space.
1
velocity factor
dielectric constant
=
The velocity factors for cable RG-223 used for 437
MHz and cable LMR-400 used for 145 MHz antennas
are 0.66 and 0.85, respectively. Thus, balun length for
437 MHz antenna is 27.63 cm and for 145 MHz
antenna is 87.89 cm.
3.3 Fabrication of crossed Yagi and connection of baluns
Solid aluminum bars of specific diameter (0.6 cm)
have been used to construct antenna elements. As
noted earlier, the Faraday rotation measurement
requires crossed Yagi antenna to estimate differential
rotation of plane of polarization of received signal. So,
the crossed Yagi for both the frequencies has been
designed. The baluns connection are so made that each
crossed Yagi has a common ground connection on
boom. An aluminum boom for both the antennas has
been used. One end of feed line is soldered on one of the
feed point on respective dipoles for each crossed Yagi
and other end is fixed into a TNC connector. The final
output of the present work is a fully fabricated crossed
Yagi antenna for respective central frequency. Figure 6
shows photograph of the fabricated antennae at NARL.
Fig. 3 —
Reflection coefficient (dB) with respect to frequencies
around 437 MHz
Table 3 — Optimized performance comparison for 145 MHz
Element/spacing Lengths for
145 MHz, cm before
optimization
Lengths for
145 MHz, cm
after optimization
A. Reflector 99.310 104.720
B. Driven /Dipole 95.172 92.191
C. Director 1 91.034 85.862
D. NA NA NA
E. NA NA NA
F. Reflector spacing 45.517 45.282
G. Director spacing 33.103 37.801
H. Dipole arm spacing
04.137 04.200
Fig. 4 —
Reflection coefficient (dB) with respect to frequencies
around 145 MHz
Fig. 5 — Mechanical drawing for the optimized lengths as
given
in Tables 2 and 3 [A-H correspond
to lengths of the respective
elements given in tables]
Fig. 6 —
Photograph of fabricated crossed Yagi for 437 MHz
(left) and 145 MHz (right)
DASHORA et al.: CROSSED YAGI ANTENNA: DESIGN & FABRICATION
129
4 RF characterizations and testing of antenna
The designed parameters for both the antenna are
given earlier. The input impedances of both antenna
designs are 50 ohm and designed VSWRs are given in
Figs 3 and 4, respectively for 437 MHz and 145 MHz.
RF testing facility is available in Radar and
Application Development Group at NARL. Two of
the first basic test parameters, i.e. input impedance
and return loss11 are measured using Agilent
Technology’s ENA-L RF Network Analyzer
(E5061A) and Anritsu’s S311d Antenna Site Master,
respectively. The advantage of a network analyzer is
its ability to measure both the magnitude and the
phase of the power received. It provides the Smith
chart so that one can easily measure impedance at
desired band width. In another test, the return loss is
measured across the frequency range that provides a
proxy to VSWR12. Figure 7 gives the Smith charts for
437 MHz and 145 MHz, while Figure 8 gives the plot
of return loss with respect to frequency in both the
cases, respectively.
It is obvious from Fig. 8 that the return loss of the
fabricated antennae shows a clear dip at the desired
central frequency, respectively. As a matter of fact,
Figs 3 and 4 exhibit reflection coefficients of software
design and Fig. 8 confirms that the same return loss
has been obtained by a precise hardware fabrication.
Thus, both the crossed Yagi antenna successfully
passed the fundamental RF test for designed antennae.
5 Conclusions
Two crossed Yagi antenna at the central
frequencies of 437 MHz and 145 MHz have been
indigenously designed, fabricated and tested. First, the
calculated parameters were optimized, then precise
fabrication was followed, that showed basic radiation
patterns and VSWR plots. Finally, these VSWR plots
were compared with measured plots and comparison
Fig. 7 — Input impedance of antenna at 437 MHz (left panel) and 145 MHz (right panel) taken from display of
Agilent Technology’s
(E5061A) network analyzer
Fig. 8 — Observed return loss curves for 437 MHz (left panel) and 145 MHz (right panel) [panels show the photographs of front dis
play
of the Anritsu’s site master S311d]
INDIAN J RADIO & SPACE PHYS, FEBRUARY 2014
130
comes very satisfactorily. At present, a software
defined radio receiver is being designed using USRP
boards that would receive the signals from these
antennae. The scientific utilization of these antennae
would be to measure Faraday rotation and this would
lead to estimation of ionospheric TEC.
Acknowledgment
The authors acknowledge the comments,
suggestions and support during lab testing by
P Srinivasulu, M Durga Rao and P Kamraj of NARL
at various levels of this work. One of the authors
(KV) is thankful to ISRO for M Tech fellowship.
They gratefully acknowledge Director, NARL for his
encouragement.
References
1 Austen J R, Franke S J & Liu C H, Ionospheric imaging using
computerized tomography, Radio Sci (USA), 23 (1988)
pp 299-307.
2 Kraus J D, Marhefka R J & Khan S A, Antenna and wave
propagation, 4th edition (Tata McGraw Hill Publication, New
Delhi), 2010.
3 Prasad K D, Antenna and wave Propagation, 3rd edition (Satya
Prakash Publications, New Delhi), 2007, pp 777-781.
4 Bandyopadhyay S, Jha J, Haripriya, Damle A, Thakur D, Mulay
S, Sachdeva P, Joshi J, Unhelkar V, Chati Y, Chaturvedi M,
Parab N, Rachh M, Tamaskar S, Bommanahal M, Goel A,
Neema K, Das S, Sresht V, Pai R & Chiplunkar A, Introduction
to Pratham, IIT Bombay’s Student Satellite Project, Paper
presented at Indian Small Satellite Systems Conference (ISSSC),
Bangalore, India, April 2010.
5 Viezbicke P P, Yagi antenna design: NBS technical note 688
(Time and Frequency Division, National Bureau of Standards, U
S Dept of Commerce), December 1976.
6 Chen C A & Cheng D K, Optimum element lengths for Yagi-
Uda arrays, IEEE Trans Antennas Propag (USA), 23 (1) (1975)
pp 8–15.
7 Burke G J & Poggio A J, Numerical electromagnetics code
(NEC) — method of moments (Lawrence Livermore National
Laboratory, Livermore, California, USA), 1981.
8 Klobuchar J & Mendillo M, Total electron content studies of the
ionosphere, report on air force survey in geophysics (Air Force
Cambridge Research Laboratories, Massachusetts, USA), 1973.
9 Davies K, Recent progress in satellite radio beacon studies with
particular emphasis on the ATS-6 radio beacon experiment,
Space Sci Rev (Netherlands), 25 (1980) pp 357-430.
10 Balanis C A, Antenna theory, analysis and design, 3rd edition,
(John Wiley, New York), 2005.
11 Raju G S N, Antennas and wave propagation (Dorling
Kindersley India Pvt Ltd, Pearson Education, New Delhi), 2004.
12 Bird Trevor S, Definition and misuse of return loss, IEEE
Antennas Propag Mag (USA), 51 (2) (April 2009) pp 166–167.
