The RASER ANTENNA
by W2OZH
Programed by K2HQ
The length of a high-gain antenna for the 75 meter band is often limited by the
available space. For example, my lot measures somewhat over 200 feet in the north-
south direction and I would like to improve signal strength (gain) in the east-west
direction. Conventional wisdom would dictate that I'm stuck with a half-wave dipole
(length -120 feet) because there isn't sufficient space available for a collinear
two half-waves in phase (-240 feet). I would like to add length in the center of the
dipole (where the radiation is greatest) in increments much less than 120 feet and
still have the currents remain in-phase so as to increase the gain in the east-west
direction. Design details will be shown for two such enhanced dipoles. Both are fed
with coaxial cable without the need for a separate tuner. One is end-fed and its
development is described in some detail. The other is center-fed and it is covered
at the end of this article. In each of the antennas the power gain relative to a
dipole is a factor of two, with a length of less than 210 feet.
The Franklin/CCD Antenna Concept:
Those who are familiar with the history of radio may know, of the Franklin
antenna, named after its inventor. This concept involves the modification of current
distribution in the elements of an antenna by the introduction of series capacitors.
General descriptions of early applications of the concept may bc found in H. Jasick's
Antenna Engineers Handbook. First Edition, pp. 4-35 and 4-36. McGraw-Hill publisher:
or F.Tennan's Radio Engineers' Handbook. First Edition. pp. 773 and 774. McGraw-Hill.
Harry, Mills W4FD and others have adapted the concept to the ham bands in the form of
resonant dipoles or loops fed with high impedance line. Mills developed a resonant
radiating system which. for 80 meters, was made up of 48 self-resonant sections, each
70 inches Iong, a total length of 280 feet.See H. Mills & G. Brizendine.
"Antenna Design: Something New!," Oct.1978, pp. 282-289. Kaplan & Bauer developed
calculations for "stretched" resonant radiators made up of multiple tuned sections
where the series tuning capacitance is half that needed to resonate the wire in each
section. the other half being used to resonate thesystem. In this case, care must be exercised
to avoid compromising the phase and, therefore, the coherence of the radiation from the
separate sections (see S. Kaplan & E. Bauer,"The Controlled Current Distribution
Antenna," ARRL Antenna Compendium, Vol. 2,pp. 132-135).
This project uses a different approach. Here we insert series self-resonant sections
into a resonant dipole antenna. This results in a coherent (in-phase) radiator for 75 me-
ters. having extended length with a corresponding increase of gain and aperture. A
simple empirical method is given to accomplish this without complicated computations.
In the past such an arrangement has been referred to by the acronym CCD: Controlled
Current Distribution. However, that acronym is now almost universally accepted to mean
Charge-Coupled Device. Thus, I prefer to use the less confusing term "DCR":
Divided Coherent Radiator.
The Divided Coherent Radiator Concept:
If we consider a short length of wire carrying RF current, it has an inductance which can
be readily calculated. If the current is to be essentially constant along the wire, its length must be a small fraction of a wavelength--for example, a fiftieth of a wavelength. For a chosen frequency the value of series capacitancerequired for resonance can then be calculated. At this frequency the tuned circuit is, of course, non-reactive: that is, essentially, it acts like an element of pure radiation resistance. If we place several of these tuned sections in series, as in Figure 1 , their currents will be in-phase and the resulting radiation will be coherent, i.e. mutually reinforcing. Note that we are placing the DCR elements of pure radiation resistance in the center of a dipole which is then trimmed for resonance, rather than demanding that the entire multi-tuned structure be self resonant.
The RASER Configurations--- End-Fed and Center-Fed
This antenna is called a RASER because of its broad functional commonality with the
LASER--both utilize multiple coherent radiating elements to achieve gain. Two RASER
configurations were developed in response to needs generated by different site restrictions.
The first, for end-feed, is derived from the RFD design ("RFD-1 and RFD-2: Resonant
Feed-Line Dipoles," by J. Taylor. QST. August 1991, pp. 24-27).
A second configuration, for center-feed, is reviewed briefly. Both use coaxial feedline.
Neither design requires an antenna tuner and each provides an excellent impedance match
with adequate bandwidth for normal amateur use. Figure 2 shows the final dimensions of
the end-fed RASER and Figure 3 shows the center-fed arrangement. Of course. the heights
above ground may vary for other locations.
Increasing the Aperture
Figure 4 is a diagram of the basic RFD-1 antenna system (shown in QST, August
1991, pp. 24-27, ref. above). Here I have labeled the input branch of the dipole radiator
the "Injector" and the output branch the "Terminator." To develop this antenna, I first
resonated the RFD-1 in the normal fashion, then introduced as many DCR sections as
desired between the injector and the terminator. Since the RFD-I is a resonant dipole
antenna it continues to function as such even after the essentially non-reactive DCR sec-
tions are inserted, but with increased aperture and gain. (Ed. Note: Due to the sinusoidal
distribution of current in the dipole,the principal radiation will be from near its center.
For example, the distance between the 6 dB power points (current 1/2 the maximum value)
will be Wavelength/3 for 75 meters, about 80 feet. This is a measure of the aperture over
which the radiated wavefront is approximately plane. Thus, if we can add a DCR
effectively equal to this length we will have doubled the aperture of the antenna.)
Residual mutual inductive and capacitive effects within the radiating system are com-
pensated for by shortening both the DCR sections and the terminator. Simple coupler
units see Figure 5 and Figure 10 assures accurate impedance matching at the desired resonant
frequency.
Determination of DCR Design Parameters:
The optimum lengths of the tuned sections of the DCR were determined by first
calculating the inductance of a X/50 length, then calculating the capacitance required for
resonance (see F. Terman's, Radio Engineer's Handbook, First Edition, p. 48ff.).
These simple calculations do not take into account mutual inductance and capacitance
among the adjacent sections of the DCR. These effects were conveniently compensat-
ed for experimentally. The resulting parameters are shown in Table I. Values for other
frequencies can be scaled from these values. These values are of key importance in the
scaling of future RASER antennas for other frequencies. During the development of the
design, I used, successively, DCRs having segments of several different numbers of
sections which were mechanically separable by coaxial connectors. This was to derive
and confirm the parameters of coupling and the optimum lengths of the sections and of
the terminator as described above. However, now that the parameters have been deter-
mined, these tests need not be repeated in the future. I decided on a 20-section RASER
because my lot is only about 220 feet from front to back. However, you can use more or
fewer sections, depending on site dimensions--only the terminator dimensions and
the coupler constants need be appropriately readjusted (see Table 1 ) to compensate.
Alternatively, the RASER can be bent around the site, but with a less predictable pattern.
Nevertheless, the increased aperture will still be beneficial, as will the other advantages
cited by Kaplan and Bauer, including: improved directivity, reduced end effects and
attendant losses, improved flexibility of scaling its length, broadband characteristics,3
and better operation close to the earth.
Coupling to the RASER:
The addition of the DCR to the RFD-I antenna increases the input impedance from '
the 50 ohm resistlye value. This is because ' of the added radiation resistance and also
because of any residual mutual reactance introduced. Several approaches were consid-
ered but the simplest and most satisfactory involved the use of a powdered iron toroidal
autotransformer with a selected, fixed, series capacitor at its input, as shown in Figure 5 "
The bifilar transformer serves, primarily, to match the impedance to that of the line; the
capacitor tunes out residual series inductance. This simple, compact coupler circuit,
housed in a convenient plastic housing, (see Parts List) enables a precise 1:1 SWR.
Referring to the Parts List, the recommended feedline is RG-8/M (Minifoam),.
having a total length of 143': (59' Injector + 22' T-Choke + -62' 1/4 Wave Lead-in). The
minifoam is chosen for its light weight and the 62' (1/4 wave) lead-in provides a measure
of added isolation. For convenience, I have used coaxial connectors at strategic spots.
Two SO-239s are mounted in the two ends of the coupler box which contains the auto-
transformer and the fixed capacitor(s), C, as shown in Figures 5 and Figure 6 . Final adjustments
are described below. The T-choke is wound on a plastic spool. The spools which I used
were the red plastic items which wire suppliers have. The winding channel is 6" diame-
ter x 2-1/4" wide. The T-choke comprises seven turns close-wound with six turns wound back in a second layer so that turn #13 is adjacent to turn #1. This coil is adjusted using a noise bridge or an SWR bridge. A supporting rope is tied around the spool and the T-choke is raised for final adjustment of the RFD-1.
Fabrication of the DCR Sections:
The DCR sections were designed for strength, lightness of weight and low wind
resistance. Cut 20 lengths of the antenna wire to 57" each. The 750 pF capacitors are
each contained in 1/2" long rings cut from the PVC tubular conduit. The rings can be
neatly cut using a rotary copper tubing cutter I did it best by supporting the conduit in-
ternally using a plastic fitting (available at the plumbing distributor) slipped inside. Thecutter was clamped in a bench vice and the tubing was rotated to produce a clean cut.
(The DCRs should be tested in the system before the potting compound is applied.)
The capacitor assemblies are shown in Figure 7 and Figure 8 , The leads of the capacitor
and the stranded wire are bent for stress relief. An excess of solder is used here to
reinforce these joints as there can be considerable torsion during high winds. The ter-
minator was initially measured to be 59' 2- I/4" long after allowance for the end insula-
tor which is, conveniently, one of the 1/2" long rings of conduit. This terminator length
was used for initial adjustments of the T- choke in the RFD-I. After these adjustments, the terminator length was reduced to the final value of 43-1/2' (for the 20-section RASER).
Construction of the Supporting Mast:
As with any low frequency antenna system. all parts of the RASER should be mounted as high as possible above ground. At W2OZH I have, for a numbleT of years, used a 2-element phased array (see J. Taylor,"An 80m Phased Array," 73 Magazine, March 1975, pp. 52-54, 56) with the manifest advantages of switchable directivity. I now use two 20-section RASERs in such an array. The arrangement shown in Figure 9 has proved to be quite practical for the two masts supporting the RASER elements. The 4" diameter aluminum pipe is light, easily
erected, and it is sufficiently rigid to permit minimal guying. The pivoted cord-reel at the
top acts like a huge pulley so that rope, knots, clamps, insulators, etc. can be easily
pulled over the top with no trouble--a great advantage both for installation and experi-
mentation.
Resonating the RFD-1:
The RFD-1 was assembled as in Figure 4 and adjusted for 50 ohms input resistance as
mentioned above (without the DCR and coupler). After the T-choke has been adjusted,
the coil is taped in place so that the windings won't shift. We are now ready to adjust the
complete RASER.
Adjusting the Coupler:
The capacitance and the position of the tap in the coupler are determined after the
length of the terminator has been reduced, from Table 1, as appropriate for the number
of sections chosen for the DCR. First, the tap and the silver-mica capacitors can be
clipped in place and the antenna raised to its normal height before measuring the input
impedance. for example, using a noise bridge. From my experience, it should not
be necessary to compromise on these values--a precise 1:1 SWR should be attainable.
For the two 20-section end-fed RASERS constructed the taps turned out to
be at 24 turns and 28 turns and the capacitance 465 pF and 487 pF respectively. Thus,
the mean values of 26 turns and about 476 pF should be a good starting point for a
specific installation of a 20-section RASER. For another number of sections in the DCR
the terminator will be changed appropriately from the initial value from Table 1. The
proper tap and capacitor are then determined experimentally. I found a variable capacitor
or decade box to be quite valuable for such preliminary measurements, which were
made with the coupler box at stepladder height.
Results:
One question which occurs for any antenna is: What is the SWR as a function of
Frequency? I measured the SWR for two 20 element RASER systems using a Heath SWR
bridge at the input to the two-wavelength long feedline used. For each, the measur
value was 1:1 from 3.900 to 4.000 Mhz . The value was less than 1 to 1 from 3.850 to
4.050 and under 1.5:1 from 3.750 to MHz. Thus, the system has a relative broad passband.
One other experiment done to confirm the proper operation of DCR. A 10 section RASER
was erected at stepladder height and the RF current in DCR sections and in the adjacent
injector and the terminator were checked, using a MFJ-Field probe. The meter measured
essentially the same reading throughout, indicating the desired constant-current operation
of the DCR sections.
The RASER system produces a readily discernible gain over a non-enhanced dipole.
Two RASERs used in a switchable phased array (see J- Taylor, "An 80m phased
Array," referenced above) gave front-to-back ratios as high as 35 decibels. The
directivity of each RASER is quite pronounced Reported signal strengths are outstanding
with output power of 100 watts or less. Stations in the preferred E-W direction worked
with uniform superiority, as those in the N-S direction are seldom worked, there is no free
lunch! The RASER phased array has been in operation at W2OZH for a year now and it
has shown clear improvement over the dipole-based array previously used.
Addendum: The Center-Fed RASER
(A Double-Edged RASER)
The RASER described, shown in Figure 2 , is most suitable for sites which favor an
end-fed antenna. For sites which favor center-feed, the version shown in Figure 3 was
constructed using the principles already developed. Briefly, it was only necessary
to split the 20-section DCR into two equal parts and connect a coaxial feedline into
a coupler unit at this point. The injector of the RASER was replaced by a second
terminator. Each of the two terminators is 51' 4"- The diagrams, Figures 3 and Figure 10 ,
indicate the changes in geometry and coupler constants. The center-fed RASER is currently
in daily use in the phased-array at W2OZH, with results which are essentially the same as
those experienced for the endfed RASER.
Conclusion:
This development project was initiated to satisfy the need for an end-fed dipole
antenna system having enhanced gain while retaining the efficiency and simplicity of
endfeed which is characteristic of the RFD approach. The RASER system described
above provides the desired enhancement. It offers an excellent match to the transceiver
without a seperate antenna tuner, and without a dangling feedline to contend with. The
concept is applicable to other bands and because of its broadband characteristics
all-band operation, using an external turner should be possible. For locations where
center feed is desired, suitable changes in design values were developed an tested.
Acknowledgments:
I wish to acknowledge encouraging telephone conversations with Harry W4FD, and
with Gene W4ATP whose joint paper originally triggered my interest. Thanks are also
due the numerous 75 meter hams who showed interest and who gave me comparative
signal strength reports.