From March 2007 QST © ARRL
It has now been more than three years since my article on the
dual band J-pole (DBJ-1) appeared in the February 2003
issue of QST.1
I have had over 500 inquires
regarding that antenna. Users have reported
good results, and a few individuals even
built the antenna and confirmed the reported
measurements. Several major cities are using
this antenna for their schools, churches and
emergency operations center. When asked
why they choose the DBJ-1, the most common
answer was value. When budgets are
tight and you want a good performance-toprice
ratio, the DBJ-1 (Dual Band J-pole–1)
is an excellent choice.
In quantity, the materials cost about $5 per
antenna and what you get is a VHF/UHF base
station antenna with λ/2 vertical performance
on both VHF and UHF bands. If a small city
builds a dozen of these antennas for schools,
public buildings, etc it would cost about $60.
Not for one, but the entire dozen!
Since it is constructed using PVC pipe, it
is UV protected and it is waterproof. To date
I have personally constructed over 400 of
these antennas for various groups and individuals
and have had excellent results. One
has withstood harsh winter conditions in the
mountains of McCall, Idaho for four years.
The most common request from users
is for a portable “roll-up” version of this
antenna for backpacking or emergency use.
To address this request, I will describe how
the principles of the DBJ-1 can be extended
to a portable roll-up antenna. Since it is the
second version of this antenna, I call it the
DBJ-2.
Principles of the DBJ-1
The earlier DBJ-1 is based on the J-pole,2
shown in Figure 1. Unlike the popular
ground plane antenna, it doesn’t need ground
The DBJ-2: A Portable VHF-UHF Roll-Up
J-pole Antenna for Public Service
Edison Fong, WB6IQN
WB6IQN reviews the theory of the dual band 2 meter / 70 cm J-pole
antenna and then makes detailed measurements of a practical, easy to
replicate, “roll-up” portable antenna.
radials. The DBJ-1 is easy to construct using
inexpensive materials from your local hardware
store. For its simplicity and small size,
the DBJ-1 offers excellent performance and
consistently outperforms a ground plane
antenna.
Its radiation pattern is close to that of an
ideal vertical dipole because it is end-fed,
with virtually no distortion of the radiation
pattern due to the feed line. A vertically
polarized, center-fed dipole will always have
some distortion of its pattern because the
feed line comes out at its center, even when a
balun is used. A vertically polarized, centerfed
antenna is also physically more difficult
to construct because of that feed line coming
out horizontally from the center.
The basic J-pole antenna is a half-wave
vertical configuration. Unlike a vertical
dipole, which because of its center feed is
usually mounted alongside a tower or some
kind of metal supporting structure, the radiation
pattern of an end-fed J-pole mounted at
the top of a tower is not distorted.
The J-pole works by matching a low
impedance (50 Ω) feed line to the high
impedance at the end of a λ/2 vertical dipole.
This is accomplished with a λ/4 matching
stub shorted at one end and open at the other.
The impedance repeats every λ/2, or every
360° around the Smith Chart. Between the
shorted end and the high impedance end of
the λ/4 shorted stub, there is a point that is
close to 50 Ω and this is where the 50 Ω coax
is connected.
By experimenting, this point is found to
be about 11⁄4 inches from the shorted end on
2 meters. This makes intuitive sense since
50 Ω is closer to a short than to an open circuit.
Although the Smith Chart shows that
this point is slightly inductive, it is still an
excellent match to 50 Ω coax. At resonance
the SWR is below 1.2:1. Figure 1 shows
the dimensions for a 2-meter J-pole. The
151⁄4 inch λ/4 section serves as the quarter
wave matching transformer.
A commonly asked question is, “Why
151⁄4 inches?” Isn’t a λ/4 at 2 meters about
181⁄2 inches? Yes, but twinlead has a reduced
velocity factor (about 0.8) compared to air
and must thus be shortened by about 20%.
A conventional J-pole configuration
works well because there is decoupling of
the feed line from the λ/2 radiator element
since the feed line is in line with the radiating
λ/2 element. Thus, pattern distortion is
minimized. But this only describes a single
band VHF J-pole. How do we make this into
a dual band J-pole?
Adding a Second Band to the
J-pole
To incorporate UHF coverage into a VHF
J-pole requires some explanation. (A more
detailed explanation is given in my February
2003 QST article.) First, a 2 meter antenna
does resonate at UHF. The 1Notes appear on page 40. key word here is
Figure 1 — The original 2 meter ribbon
J-pole antenna.
From March 2007 QST © ARRL
Figure 5 — The λ/4 UHF decoupling stub made of RG-174A, covered with heat shrink
tubing. This is shown next to the BNC connector that goes to the transceiver.
Figure 2 — Elevation plane pattern
comparing 2 meter J-pole on fundamental
and on third harmonic frequency (70 cm),
with the antenna mounted 8 feet above
ground. Most of the energy at the third
harmonic is launched at 44º.
Figure 3 — The original DBJ-1 dual-band
J-pole. The dimensions given assume that
the antenna is inserted into a 3⁄4 inch Class
200 PVC pipe.
Figure 4 — The dualband
J-pole modified
for portable operation
— thus becoming
the DBJ-2. Note that
the dimensions are
slightly longer than
those in Figure 3
because it is not
enclosed in a PVC
dielectric tube.
Please remember that
the exact dimensions
vary with the manufacturer
of the 300 Ω
line, especially the
exact tap point where
the RG-174A feed
coax for the radio is
connected.
resonate. For example, any LC circuit can
be resonant, but that does not imply that it
works well as an antenna. Resonating is one
thing; working well as an antenna is another.
You should understand that a λ/4 146 MHz
matching stub works as a 3λ/4 matching
stub at 450 MHz, except for the small
amount of extra transmission line losses of
the extra λ/2 at UHF. The UHF signal is
simply taking one more revolution around
the Smith Chart.
The uniqueness of the DBJ-1 concept
is that it not only resonates on both bands
but also actually performs as a λ/2 radiator
on both bands. An interesting fact to note
is that almost all antennas will resonate at
their third harmonic (it will resonate on any
odd harmonic 3, 5, 7, etc). This is why a
40 meter dipole can be used on 15 meters.
The difference is that the performance at the
third harmonic is poor when the antenna is
used in a vertical configuration, as in the
J pole shown in Figure 1. This can be best
explained by a 19 inch 2 meter vertical over
an ideal ground plane. At 2 meters, it is a λ/4
length vertical (approximately 18 inches).
At UHF (450 MHz) it is a 3λ/4 vertical.
Unfortunately, the additional λ/2 at UHF is
out of phase with the bottom λ/4. This means
cancellation occurs in the radiation pattern
and the majority of the energy is launched at
a takeoff angle of 45°. This results in about
a 4 to 6 dB loss in the horizontal plane compared
to a conventional λ/4 vertical placed
over a ground plane. A horizontal radiation
pattern obtained from EZNEC is shown in
Figure 2. Notice that the 3λ/4 radiator has
most of its energy at 45°.
Thus, although an antenna can be made
to work at its third harmonic, its performance
is poor. What we need is a simple,
reliable method to decouple the remaining
λ/2 at UHF of a 2 meter radiator, but have
it remain electrically unaffected at VHF. We
want independent λ/2 radiators at both VHF
and UHF frequencies. The original DBJ-1
used a combination of coaxial stubs and
300 Ω twinlead cable, as shown in Figure 3.
Refer to Figure 3, and start from the
left hand bottom. Proceed vertically to the
RG-174A lead in cable. To connect to the
antenna, about 5 feet of RG-174A was used
with a BNC connector on the other end. The
λ/4 VHF impedance transformer is made
from 300 Ω twin lead. Its approximate
length is 15 inches due to the velocity factor
of the 300 Ω material. The λ/4 piece is
shorted at the bottom and thus is an open
circuit (high impedance) at the end of the λ/4
section. This matches well to the λ/2 radiator
for VHF. The 50 Ω tap is about 11⁄4 inches
from the short, as mentioned before.
For UHF operation, the λ/4 matching
stub at VHF is now a 3λ/4 matching stub.
This is electrically a λ/4 stub with an additional
λ/2 in series. Since the purpose of the
matching stub is for impedance matching
and not for radiation, it does not directly
affect the radiation efficiency of the antenna.
It does, however, suffer some transmission
loss from the additional λ/2, which would
not be needed if it were not for the dual
band operation. I estimate this loss at about
0.1 dB. Next comes the λ/2 radiating element
for UHF, which is about 12 inches. To
From March 2007 QST © ARRL
make it electrically terminate at 12 inches, a
λ/4 shorted stub at UHF is constructed using
RG-174A. The open end is then connected
to the end of the 12 inches of 300 Ω twinlead.
The open circuit of this λ/4 coax is only
valid at UHF. Also, notice that it is 41⁄2 inches
and not 6 inches due to the velocity factor of
RG-174A, which is about 0.6.
At the shorted end of the 41⁄2 inch
RG-174A is the final 18 inches of 300 Ω
twinlead. Thus the 12 inches for the UHF
λ/2, the 41⁄2 inches of RG-174A for the
decoupling stub at UHF, and the 18 inches
of twinlead provide for the λ/2 at 2 meters.
The total does not add up to a full 36 inches
that you might think. This is because the
λ/4 UHF RG-174A shorted stub is inductive
at 2 meters, thus slightly shortening the
antenna.
Making it Portable
The single most common question that
people asked regarding the DBJ-1 is how it
could be made portable. The original DBJ-1
had the antenna inserted into Class 200 PVC
pipe that was 6 feet long. This was fine for
fixed operation but would hardly be suitable
for portable use. Basically the new antenna
had to have the ability to be rolled up when
not in use and had to be durable enough for
use in emergency communications.
The challenge was to transfer the concepts
developed for the DBJ-1 and apply them to
a durable roll-up portable antenna. After
much thought and experimenting, I adopted
the configuration shown in Figure 4.
The major challenge was keeping the
electrical characteristics the same as the
original DBJ-1 but physically constructing
it from a continuous piece of 300 Ω twinlead.
Any full splices on the twinlead would
compromise the durability, so to electrically
disconnect sections of the twinlead, I cut
small 1⁄4 inch notches to achieve the proper
resonances. I left the insulating backbone
of the 300 Ω twinlead fully intact. I determined
the two notches close to the λ/4 UHF
decoupling stub by experiment to give the
best SWR and bandwidth.
Because this antenna does not sit inside
a dielectric PVC tube, the dimensions are
about 5% longer than the original DBJ-1.
is significant. I have confidence in these
measurements since the flexible antenna is
about −6 dB from that of the λ/4 ground
plane antenna, which agrees well with the
literature.
Also notice that at UHF, the loss for the
flex antenna is only 2.0 dB, compared to the
ground plane. This is because the flexible
antenna at UHF is already 6 inches long,
which is a quarter wave. So the major difference
for the flexible antenna at UHF is the
lack of ground radials.
Summary
I presented how to construct a portable,
roll-up dual-band J-pole. I’ve discussed its
basic theory of operation, and have presented
experimental results comparing the DBJ-2
to a standard ground plane, a traditional
2 meter J-pole and a flexible antenna. The
DBJ-2 antenna is easy to construct, is low
cost and is very compact. It should be
an asset for ARES applications. It offers
significant improvement in both the VHF
and UHF bands compared to the stock flexible
antenna antenna included with a handheld
transceiver.
If you do not have the equipment to
construct or tune this antenna at both VHF
and UHF, the antenna is available from the
author tuned to your desired frequency. Cost
is $20. E-mail him for details.
Notes
1E. Fong, “The DBJ-1: A VHF-UHF Dual-Band
J-Pole,” QST, Feb 2003, pp 38-40.
2J. Reynante, “An Easy Dual-Band VHF/UHF
Antenna,” QST, Sep 1994, pp 61-62.
Table 1
Measured Relative Performance of the Dual-band
Antenna at 146 MHz
VHF Flexible Standard Dual-Band
VHF λ/4 GP Antenna VHF J-Pole J-Pole
4 radials
0 dB −5.9 dB +1.2 dB +1.2 dB
reference Table 2
Measured Relative Performance of the Dual-band
Antenna at 445 MHz UHF Fexible Standard Dual-Band
UHF λ/4 GP Antenna VHF J-Pole J-Pole
4 radials 0 dB −2.0 dB −5.5 dB 0.5 dB
reference I used heat shrink tubing to cover and protect
the UHF λ/4 decoupling stub and the
four 1⁄4 inch notches. Similarly, I protected
with heat shrink tubing the RG-174A coax
interface to the 300 Ω twinlead. I also
attached a small Teflon tie strap to the top
of the antenna so that it may be conveniently
attached to a nonconductive support string.
Figure 5 shows a picture of the λ/4 UHF
matching stub inside the heat shrink tubing.
The DBJ-2 can easily fit inside a pouch or a
large pocket. It is far less complex than what
would be needed for a single band ground
plane, yet this antenna will consistently outperform
a ground plane using 3 or 4 radials.
Setup time is less than a minute.
I’ve constructed more than a hundred
of these antennas. The top of the DBJ-2 is
a high impedance point, so objects (even if
they are nonmetallic) must be as far away
as possible for best performance. The other
sensitive points are the open end of the λ/4
VHF matching section and the open end of
the λ/4 UHF decoupling stub.
As with any antenna, it works best as
high as possible and in the clear. To hoist the
antenna, use non-conducting string. Fishing
line also works well.
Measured Results
I measured the DBJ-2 in an open field
using an Advantest R3361 Spectrum
Analyzer. The results are shown in Table 1.
The antenna gives a 7 dB improvement over
a flexible antenna at VHF. In actual practice,
since the antenna can be mounted higher
than the flexible antenna at the end of your
handheld, results of +10 dB are not uncommon.
This is the electrical equivalent of giving
a 4 W handheld a boost to 40 W.
The DBJ-2 performs as predicted on
2 meters. It basically has the same performance
as a single band J-pole, which gives
about a 1 dB improvement over a λ/4 ground
plane antenna. There is no measurable
degradation in performance by incorporating
the UHF capability into a conventional
J-pole.
The DBJ-2’s improved performance
is apparent at UHF, where it outperforms
the single band 2 meter J-pole operating
at UHF by about 6 dB. See Table 2.
This Ed Fong was first licensed in 1968 as WN6IQN.
He later upgraded to Amateur Extra class
with his present call of WB6IQN. He obtained
BSEE and MSEE degrees from the University
of California at Berkeley and his PhD from the
University of San Francisco. A Senior Member
of the IEEE, he has 8 patents, 24 published
papers and a book in the area of communications
and integrated circuit design. Presently,
he is employed by the University of California
at Berkeley teaching graduate classes in RF
design and is a Principal Engineer at National
Semiconductor, Santa Clara, California working
with CMOS analog circuits. You can reach the
author at edison_fong@hotmail.com.
