n previous issues of antenneX magazine, we have discussed methods of making broadband antennas. These include the Prismatic Polyhedral antennas and their variants, the cage dipoles.

Alan Boswell suggested that the broadband property of the Pn antennas derives principally from the fact that they are very similar to fat dipoles.

Dave Cuthbert suggested that the Pn antennas were really a variant of the cage dipole structure. For a discussion of the wideband properties of cage dipoles,  please see antenneX archive V, number 98.

It is our belief that for all these antennas, details of the feed at the antenna gap are indeed important, and it is for this reason that the Pn antennas are found to out-perform the other variants.

Accordingly, we present (in this short paper) a simulation of Alan Boswell's suggested "fat dipole" structure. As we shall see, these are also significantly wideband antennas, well behaved, with benign feeding characteristics; but they also are somewhat sensitive to details of the feeding arrangements.

All these related studies and papers indicate that broadband antenna structures may reliably be implemented by taking a resonant dipole and greatly increasing its diameter/length ratio. By this method we can extend the bandwidth over which the VSWR is less than 2:1 beyond an octave, for reasonable gain values and acceptable squint. Broadband antenna structures are discussed also in antenneX archive V, article numbers 1, 18, 32, 49, and 56.

Figure 1 shows a schematic NEC diagram of the construction of a skeleton fat dipole. What the diagram does not show is the relationship between the overall dimensions and the diameter of the radiating (vertical) elements, and the diameters of the elements at the feed planes (in the centre) and the ends.

The data for a particular simulated structure is presented in Figures 2, 3, 4 and 5. This structure has been chosen to have optimum gain at the bottom end of the bandpass. The dimensions of this implementation are as follows:

overall height, 0.681 metres,
width, 0.094 metres,
gap between the feed planes, 0.05 metres,
wire diameter in the feed planes, 0.01 metres, and
wire diameters for the rest of the structure, 0.007 metres.

If the feed plane wires were made thinner, and the feed plane separation lessened to maintain the characteristic impedances Zo of the radial transmission lines, then the gain dropped off due to resistive losses in the copper.

Figure 2

Figure 2 shows the VSWR plot for this antenna over the range 140-360 MHz. If we take the height of the antenna as a scale, this would be a free-space half wavelength at a frequency of 220MHz. The VSWR < 2.0 bandwidth is shown as 140MHz at the bottom to 357MHz at the top, with an upwards excursion touching 2.0 at the half-wave frequency of 220MHz. The points on the figures are plotted at 10MHz intervals. The reference impedance for the plots is taken to be 50 ohms, rather than the 72 ohms characteristic impedance of a thin dipole. 50-ohm feeder is readily available, and this subterfuge extends the defined bandwidth.

Figure 3

Figure 3 shows the reactance plot for the antenna structure; it appears to be resonant (real impedance, reactance zero) at frequencies 159MHz, 192MHz, and 352MHz. At these frequencies, the driving point resistance (from Figure 4) appears to be 47, 82, and 27 ohms respectively.   The antenna seems to be well behaved over its wide bandwidth, which is a factor of about 2.5 to 1 (top frequency to bottom frequency).

Figure 4

The behaviour of this antenna is summarised neatly in the SMITH chart plot shown in Figure 5, which is in some ways the optimum format for displaying this kind of data.  Again, following the data points around we remember that they are spaced by 10MHz increments, and that the SMITH chart is normalised to 50 ohms.

Figure 5

The gain of this structure, in free space, is predicted to be 1.5dB at 140MHz and 2.65dB at 360MHz. No amount of playing around with the overall dimensions, the wire diameters, or the reference impedance (in the range 35-100 ohms) produced a VSWR < 2 bandwidth greater than 140 to 360 MHz.  However, an alternative modelling package produced gain predictions more in line with those we believe pertain to the prismatic polyhedron antennas. This is still more evidence that bare NEC models should be treated with circumspection. As the feed arrangement is manifestly impracticable for our current models, we should take these numerical predictions as being indicative of the situation, and not necessarily achievable in a real construction of the antenna.

Comparing the results with a properly designed P6 antenna, which can be produced with a bandwidth ratio of 4:1 and is smaller, we can see that making a "fat dipole" structure provides a significant broadbanding effect, but does not of itself achieve the best possible result. Moreover, the structure of a "fat dipole" is not as simple as that of a Pn antenna, where the skeleton is reduced to its minimal form for a given n.

The very broadband Pn antennas have a propensity to squint at the top of their frequency range, because of the reversal of current on the radiating verticals. At the top end of the frequency range the total radiator length exceeds 1.25 wavelengths at that particular frequency, and as is well known for ordinary dipoles this produces additional squint lobes.

This disadvantage can be worked around by making the Pn structures physically smaller. The squint does not then occur within the two-octave passband. Pn antennas have optimum performance at greater width to height ratios, for the same total perimeter of radiator.

For the fat dipoles considered here, the bandwidth is not intrinsically large enough for this squint effect to be a problem.

For Amateur radio purposes, there is clearly an advantage to be had (in bandwidth) by constructing skeleton antennas which mimic very thick conductors.

It is found that vertically stacked arrays work exceedingly well, with gain enhancement and wide bandwidths, as the coupling between the elements is minimal for this direction of spacing. –30-

Dr. David J. Jefferies School of Electronics and Physical Sciences University of Surrey Guildford GU2 7XH Surrey, England D.Jefferies email Click Here for the Authors' Biography

BRIEF BIOGRAPHY OF AUTHOR Dan Handelsman, N2DT~ Email Dan Handelsman, N2DT was first licensed as WA2BCG in 1957at age 13. He became interested in antennas at that time when he had to figure out a way to operate from the 6th floor of his apartment house. This resulted in a mobile whip being stuck out from a window without a counterpoise. At that point he became an "expert" in TVI. He was licensed as N2DT in 1977 and is a DX'er and contester. He is now playing with experimental antennas and low power. Professionally, he is a Pediatric Endocrinologist and holds M.D. and J.D. degrees and is Clinical Professor of Pediatrics at the New York Medical College. As far as his antenna work he is an "amateur" in the truest sense of the word (Dan's words!).

Send mail to webmaster@antennex.com with questions or comments.
Copyright © 1988-2003 All rights reserved worldwide - antenneX©
Last modified: December 26, 2002