RE: NEC-LIST: Quadrifilar Helix Modeling on NEC

Joe:

The QFH antenna can be modeled quite successfully in NEC with results
that more or less match measured data of pattern and impedance.

One of the critical issues that I have noticed with NEC was differing
results depending on how you short-circuit the helix windings at the
bottom of the antenna (assuming you use a short circuit design). I
have found that it is best to construct the QFH antenna in NEC as two
separate bifilar helix antennas.

As for how to design a QFH, the best references have been written by
C. Kilgus: 1) "Multielement, Fractional Turn Helices" - IEEE
transactions July 1968; 2) "Resonant Quadrifilar Helix Design" - The
Microwave Journal December 1970; 3) "Shaped Conical Radiation Pattern
Performance of the Backfire Quadrifilar Helix" - IEEE transactions May
1975. Kilgus gives very detailed design graphs relating pattern shape
to the number of turns, helix radius and helix axial length. Kilgus
also presents the formulas describing the geometry of the antenna.

To facilitate modeling of the qfh antenna in NEC, I have written a
FORTRAN program that builds a NEC wire model of the QFH antenna where
the input data is the number of turns, the helix radius and the helix
axial length. I will e-mail the source code and executable program
directly to you if you are interested.

Given your specific interest in weather satellite QFH antennas you
have two basic options. 1) A qfh antenna for almost full hemispheric
pattern coverage will require a 1/2 turn design. 2) A qfh antenna for
shaped coverage with the beam pattern at about 30 degrees will require
a 1 turn design.

There are numerous ways to construct a QFH antenna. For ease of
construction, it is best to wind the antenna on some form of
dielectric support tube. The windings on the tube can be made using
copper tape. A tube about 5 - 6 inches in diameter works quite well
at 137 MHz. The 1/2 turn antenna will be about 36 inches in length or
shorter (depends on the dielectric material used as the support tube).
The 1 turn antenna will be about 72 inches in length or shorter.

To feed the antenna, a split balun feed can be used at the top of the
antenna. Since the antenna will have a feed point impedance of about
5 ohms (1/2 turn design), a coaxial transformer will be required to
bring the impedance to 50 ohms. The 90 degree relative phase shift
between the elements can be achieved by shortening one set of windings
and lengthening the other. This is somewhat of an iterative process
but a length difference of about +/- 1 to 2 inches should work at 137
MHz (I need to check my files to verify this).

Should you require more detailed information, please do not hesitate
to contact me directly.

Regards,

Steven R. Best
Received on Thu Jun 29 2000 - 17:01:37 EDT