NEC-List: MININEC Shows Standing Waves at Resonance on Twinline

Hello,
 
My name is Jerry Ehman and I have a problem that I hope one of you can
help me with.
 
We are using MININEC Broadcast Professional by EM Scientific, Inc. to
model several antennas. We are getting strange results from the
twinline and dipole model. Can someone help us determine what is
wrong? The bottom line is that at resonance (as defined by zero
reactance), there are standing waves of current on the twinline; this
is not supposed to occur.

Let me now give the details of the model. The twinline and dipole are
located in free space in the x-y plane so that z = 0 for all
coordinates.

I started out with just a dipole with a gap at the center (between the
two arms) and with each arm (nominally a quarter wave) being 7.5 cm
long. The intent was for the wavelength to be about 4 * 7.5 cm = 30 cm
(in free space), corresponding to a resonant frequency of 1000 MHz. Of
course, propagation of current along a conductor would make it too
long to resonate at 1000 MHz. We would expect it to resonate someplace
above 900 MHz but below 1000 MHz.
 
Four points were defined in (x,y) coordinates, measured in cm (note
that z = 0 for all points and wires):

Point # x y
  1 0 -0.18
  2 0 +0.18
  3 0 -7.68
  4 0 +7.68
 
Three wires were connected among these points as follows:

Wire # 1st Point 2nd Point Wire Radius Number of Segments
  1 1 3 0.15 cm 9
  2 2 4 0.15 cm 9
  3 1 2 0.15 cm 2

A current source was mathematically inserted at (x,y) = (0,0) at the
center of wire 3. The total dipole has a length of 15.36 cm
corresponding to a free-space frequency of 976.56 MHz. Due to the fact
that the propagation of information (i.e., group velocity) in a
conductor is a few percent slower than the free-space speed of light,
we expect the resonant frequency to be lower than 976.56 MHz. Indeed,
MININEC showed that the reactance component of the complex impedance
went to zero (and thus, resonance occurred) at about 921.82 MHz (or
about 5.6% below the free-space resonant frequency, consistent with
tables and graphs in the literature).

I then modelled the same dipole with a 30-cm long twinline
inserted. The new tables of points and wires are as follows:

Point # x y
  1 0 -0.18
  2 0 +0.18
  3 30 -0.18
  4 30 +0.18
  5 30 -7.68
  6 30 +7.68

Wire # 1st Point 2nd Point Wire Radius Number of Segments
  1 1 3 0.15 cm 36
  2 2 4 0.15 cm 36
  3 3 5 0.15 cm 9
  4 4 6 0.15 cm 9
  5 1 2 0.15 cm 2

The current source is mathematically inserted at (x,y) = (0,0) at the
center of wire 5 (note that the only purpose of wire 5 is to provide a
place for the current source to be inserted). Wires 1 and 2 make up
the twinline. Wires 3 and 4 are the dipole. The diameter of each wire
(d) is 0.3 cm and the center-to-center spacing between the wires (D)
is 0.36 cm. The formula for the characteristic impedance (Z0) of a
twinline in free space is given by:

   Z0 = 120 * inverse hyperbolic cosine (D/d)

This formula yields Z0 = 74.68 ohms. Since a dipole in free space has
a characteristic impedance (resistance) at resonance of about 72 ohms,
the dimensions of this twinline should make a good match for both the
dipole at one end of the twinline and a 75-ohm cable as a feedline at
the other end.

We would expect a resonant frequency of 921.82 MHz as well as
negligible standing waves at resonance, but this is not what we
got. MININEC gave us a resonant frequency of 968.62 MHz (up 46.80 MHz
or about 5% higher than that for the dipole alone). We got noticeable
standing waves as based on the printout and graphs of peak current
vs. location (i.e., x coordinate) along the twinline. For example, at
921.82 MHz, the ratio of the maximum peak current (Imax, located at
the point of largest current amplitude) to the minimum peak current
(Imin, located at the point of smallest current amplitude) was 1.69;
at a frequency of 968.62 MHz, the corresponding ratio of Imax/Imin was
1.43 . Taking a frequency range from 900 MHz to 1000 MHz, at no
frequency did this ratio get below 1.40; in other words, standing
waves (as measured by the peak amplitude of the currents) always
occurred regardless of frequency. We expected a constant current
amplitude and hence a ratio of Imax/Imin = 1 at resonance but we
didn't get that. What's up?

Thanks for any ideas you have.

 
Received on Wed Jun 10 1998 - 09:26:14 EDT