W6SDO.COM                                                    SAN DIEGO, CALIFORNIA USA
        NVIS Antenna

An NVIS antenna (near vertical incident
skywave) for 75, 60 and 40 meters is a
part of the antenna plan because of the
performance improvement that this type
of antenna offers, especially in the zone
between the point where ground waves
leaves off and where good sky wave
performance begins. It should be noted
that the improvements that are realized
from a specialized NVIS antenna are
sometimes not as good as the builder
expects. This happens because the
“conventional” antenna configuration,
against which the NVIS antenna is being  
compared, is often sufficiently low to the
ground that it also exhibit substantial NVIS
properties. Therefore, this NVIS antenna
has been designed so that it will clearly
differentiate itself from my off center fed
(OCF) inverted V antenna with its apex at
45 feet. This antenna is practical,
uncomplicated, inexpensive and easy to
install. It provides versatility and
performance that is well beyond a typical
field day antenna concept for which NVIS
antennas are well known. Since I have put
more effort into the research and
optimization of this antenna than I have for
any of the other antennas that I have
installed, I am very gratified that the extra
effort has paid off. The result is a very
competitive antenna that is rewarding to
use on a daily basis whenever I am on the
75, 60 or 40 meter bands.

The design parameters necessary for a
successful NVIS antenna are:

1. The design must provide coverage for
each of the ham bands that respond
favorably to the principals of NVIS
operation, i.e. 75, 60 and 40 meters.

2. The antenna must maximize the
transmitted power upward at 90 degrees
and, if possible, provide some gain in the
vertical direction.

3.  The response to both man made and
natural atmospheric noise must be
minimized in order to achieve the best
possible receiving signal to noise ratio.

First, three band coverage is achieved for
this NVIS antenna design by employing
three separate dipoles, one for each band.
The three dipoles are suspended parallel
to each other and connected to a common
feed point.

Next, a literature search revealed that the
typical radiation pattern for a dipole that is
placed close to the ground (less than 0.1
wave length) resembles a round balloon
that has been squashed partially flat on the
bottom. As the dipole is lowered more, to
within a few feet of the ground (especially
if it is located directly above a metallic
reflector), the radiation pattern transitions
toward a more oval balloon shape, with
more of the radiation at 90 degrees. This
shape is ideal for an NVIS antenna as it
helps to discriminate against noise
arriving at low angles yet maximizes the
radiation at 90 degrees when transmitting.
The analysis that was performed on the
final NVIS antenna that is deployed here
showed that the low angle gain horizontally
toward the sides is significantly reduced
while the vertical gain is only reduced

The ground directly below the antenna is
actually a 12 inch wide by 12 inch deep
galvanized steel drain channel. This metal
strip represents something of a skyward
reflector for the dipoles, or at least,
provides a really low local ground
resistively under the antenna.

Somewhat contrary to a number of NVIS
antenna simulations that I have read, an
analysis of this particular antenna
configuration predicts the horizontal
radiation to be about -6dB compared to a
dipole and the vertical radiation to be in
the area of +2db. Most other simulations
that I have seen calculate that the vertical
radiation would be closer to -2db (maybe
more testing will eventually sort this out).
Whether the modeling is correct of not, this
antenna is an outstanding performer, is
exceptionally quiet and has a very good
signal to noise ratio.  The lack of vertical
gain when transmitting can be made up by
adding power (thanks to Alpha for 1500
watts PEP whenever it is needed). When
on 60 meters, where the power is limited
to 100 watts, you just live with what you
get. However, I must say that the signal
strength reports that I get are usually equal
to or better than those given to most other
stations at similar distances and direction.

Even though this NVIS antenna is
horizontally polarized and is oriented North
and South on my property, the transmitted
signal gets thoroughly stirred up by the
ionosphere before it is reflected back
down to earth, thereby delivering a more
or less omni-directional signal with a mix
of polarities. There is always something
sent back from the ionosphere for
everyone to hear!

In my urban environment, the most
noticeable noise that is received is man-
made noise from close by sources such
as power lines, street lighting, household
appliances, fluorescent lighting, lighting
and fan controllers, computers, plasma
screen television sets and the like. There
is also a steady din of assorted noises
that originate from the nearby city center
(fifteen miles to the south) as well as from
several  industrial facilities to the east and

In addition, there is the noise that is
caused by thunder storm discharges both
nearby or globally.

While some man-made noise will arrive at
the receiving antenna by skimming along
the ground, most will arrive at angles of
up to 15 degrees or so. The typical mode
of travel for man made noise is by ground
wave propagation. Further, most man
made noise will be vertically polarized
since the e field component of horizontally
polarized noise that is in contact with the
earth  is greatly attenuated by being
“shorted out” by the ground along the path
between the source and the receiving
antenna. Therefore, in order to reject as
much man made noise as possible, a
good NIVS antenna will have minimum
sensitivity to signals arriving from low
angles and will be horizontally polarized.

The noise from thunder storm type events
will probably arrive at an angle between
20 and 60 degrees above the horizon
depending on where they originate and
the condition of the ionosphere. Noise
from some nearby thunderstorm events
can arrive at very high angles as well as
by ground wave - take your pick.

The very low height of this NVIS antenna
is intended to reduce the gain in the low
angle range without excessively
attenuating signals at 90 degrees
(whether for receiving or for transmitting).
Typically, the low gain for the received
signals can be compensated for by kicking
in the receiver preamplifier. The 75 meter
antenna has been intentionally mounted
at the lowest position of the three, i.e. 5
feet above the ground. This is so that it
can be expected to have the lowest
response to both man made and natural
noise and therefore the highest signal to
noise level. This choice was made
because 75 meters is generally the
noisiest of the three bands at my

The construction of this antenna is very
easy, especially since all of the pieces
are within easy reach of the ground. The
antenna is basically made up of a 75
meter dipole, a 60 meter dipole and a 40
meter dipole (in parallel) that are fed in
the center with a single 1:1 balun. This
balun is connected to a 50 ohm RG-8
coaxial cable that leads to the shack.
Vinyl insulated 12 gauge stranded
copper wire is used for the dipoles in
order to get a little wider bandwidth, to
minimize wire stretching after the tuning
is completed and to accommodate the
legal limit power level.  The balun is rated
at 5KW so that full power can be used if
necessary. As with my other antennas,
the coaxial cable that feeds this antenna
is connected to a ground stake and
lightening surge arrestor on the way to
the shack.


This is a great antenna for use on
the local 75 meter
morning nets
and often works on the early
evening AM nets as well. I
t is very
quiet and tends to eliminate much
of the QRM from far away places.
Photo above: This is the center feed point
for the three band NVIS antenna. The 5 kw
1:1 balun feeds the three dipole wires. The
wires fan out to a 12 inch spacing which is
then maintained along the antenna length
by the PVC poles pictured below.


The 75 meter dipole is 5 feet above the
reflector, the 60 meter dipole is 6 feet above
the reflector (0.033 wavelength) and the 40
meter dipole is spaced 7 feet above the
reflector (0.052 wavelength). Twelve inches of
separation between each of the three dipoles
is sufficient to minimize any significant
negative interaction among the three

The uniform spacing above the reflector and
between the three dipoles in maintained by
PVC support posts that are located every ten
feet along the length of the dipoles.

As described previously, the antenna has
been placed directly above a metal drainage
ditch that was fortuitously located at the top of
the block wall that runs along the east edge of
my property. If the antenna was to be installed
in a different location, I would have placed a
number of copper wires on the ground under
the antenna in order to achieve a similar

Lightening and safety issues have been
considered in the design and installation of
this antenna. Low mounted NVIS antennas
can have a couple of safety issues that must
be addressed. Two of these issues are
exposure to electric shock and the lack of
headroom clearance. Because the 12 by 12
inch galvanized steel drainage ditch that is
used for the reflector for this antenna is
located at the top of a five foot high concrete
block retaining wall (running along one side
of our property), the lowest antenna wire is
typically ten feet above the side yard walking
surface. This insures that there will be no
accidental contact with the antenna while it is
in used for transmitting. For a short distance
at one end of the antenna, where the wall is
shorter than five feet, the antenna is protected
from accidental contact by placing it inside of
a length of PVC tubing.

To reduce the chance of lightening caused
surge damage to equipment in the shack, the
braid on the coaxial cable connecting the
antenna to shack is connected to a group of
three 10 foot long ground rods just outside of
the shack. The coaxial cable is connected to
an Alpha-Delta 2 kw transient surge
suppressor at this point before it proceeds
on to the shack.

Tuning was performed by starting with wire
lengths for the dipoles that were a few feet
longer than needed. In the case of the 75
meter dipole, the antenna was a few feet
longer that the length of my lot, from front to
back. Therefore, a short length of antenna
wire is allowed to hang down from the
insulator at one end. There is no casual
exposure at this end of the antenna as it is
separated from the back yard level by a five
foot high concrete block wall.  

A MFJ-269 was connected to the end of the
coaxial transmission line to monitor the SWR
during the tuning process. Measurements
were continuously made on each side of the
SWR minimum as the wire length was t
rimmed. This insures that it was not possible
to trim off too much wire. Be sure to remove
the same amount of wire from each end
(I save the cutoffs so that, if necessary, I can
reconstruct what I have done). Since there is
some interaction between the parallel dipoles,
tuning was rotated back and forth among the
three dipoles as the sweet spot was
approached. The 75 meter and 40 meter
antennas were tuned for minimum SWR in
the center of the phone band (general class
for now) and the 60 Meter wire was tuned for
minimum SWR on channel 3. Although the
SWR minimum point is not at the exact
resonant point for the antennas, this choice
makes them easier to load without the need
for a tuner. Although this compromise might
lead to a slight reduction in the receiving
capture area, the differences are negligible.
Maybe some day in the future I will try moving
the feed points slightly off center in order to
try to get the minimum Xs to align with the
optimum Rs – although at this point I don’t
believe that the benefits would warrant the
additional effort.

The resulting SWR for each of the three
tri-band dipoles at the frequency noted is as

Band               75 m     60 m     40m

Frequency      3.900      ch 3     7.237

Kenwood         1.8:1     1.4:1    1.2:1

Alpha                1.7:1     1.4:1    1.3:1

MFJ-269          1.9:1      1.5:1   1.4.1

The differences in the SWR that is measured
by each of the measuring devices is probably
the result of each device having a slightly
different way of performing the SWR
computation. Never the less, each shows the
SWR minimum at virtually the same

While the 40 meter performance of this NVIS
antenna is generally effective throughout the
entire day, the 75 meter performance begins
to drop off quickly quite early in the day –
generally by 9 am for example. However on
some days, it works much later into the day
(but not often). In addition, it is appropriate to
note that on a typical day, the farther south
that you are located in the continental US, the
higher the maximum frequency that can be
successfully used for NVIS operation. This
probably explains the exceptional success
that I have had with the NVIS antenna on 40
Meters here in San Diego. In the northern
states you’d be likely to wonder what all of the
fuss is about – at least in the winter months or
during periods when there is minimal sun spot

Overall, the results from using this NVIS
antenna for both receiving and transmitting
confirm the success of the design. The outer
effective perimeter seems to vary from 150
miles to as much as 500 miles, depending on
the band used and on band conditions.
Without question, the NVIS antenna is a very
useful tool to add to any antenna assortment.
It is often the only antenna that will pull in a
readable signal on 75, 60 or 40 meters and
is therefore very worth having. This antenna
is definitely a real workhorse and not just for
field day use!
Last Revised December 10, 2012