[AReU] ANTENNA TEST RESULTS

[w4tas]

SATURDAY FEBRUARY 16, 2002
DAVIS PARK AO 40 DOWNLINK
ANTENNA AND CONVERTER
TEST RESULTS.

First of all I want to thank al of those who braved the cold and
rainy weather to participate in the activities. Apparently Doug
(KD4LPP) partially finished a fine looking J-Pole antenna
for two meters. We'll be looking for a fine signal from Doug
very soon.

We had 6 types of antennas to test that day. Five did not have
converters attached and one did. I had my R-Myers dish there
which was the largest dish and the one we used for a reference.
When testing each antenna, we used the same converter so that
only the antennas were the unknown part. We later tested
converters with a very small (less that 2 in.) log periodic
dipole array which was good for 2.1 to 6.0 GHz.

The signal source was "home brew" and consisted of a 50
MHz "brick" oscillator which emitted a square wave at TTL
levels (zero to five volts). This signal was sent to a couple of
antiparallel 1N4148 diodes to generate a plus and minus 0.6
volt square wave at 50 MHz. This was fed into an MMIC
which was biased to generate harmonics. The 48 th. harmonic
was filtered out with a "pipe" filter to generate a clean and
relatively stable 2400 MHz signal. This signal was fed into
a 3.5 turn right hand circular polarized helix antenna with
a 4 in. diameter back plane and a quarter turn matching
section. The power was supplied by 8 AA batteries. All
internal voltages were regulated so as to increase both
frequency and amplitude stability.


The antennas tested were as follows:

1. R-Myers dish approximately 2 ft. by 3 ft. with dipole feed.
2. Small MMDS dish about 1 ft by 1 ft. with dipole feed.
3. Home brew 29 turn helix with 5 in. back plane.
4. VERY small LPDA array for 2.1 to 6.0 GHz.
5. A somewhat larger LPDA array for 0.9 to 2.6 GHz.
6. W7LGD's 2 ft. by 2 ft. dish with installed converter and
    dipole feed.

I had a variable attenuator between the 2 meter output of the
converter and the receiver, a Yaesu FRG 8800 with VHF
converter installed. The attenuation was adjustable in 1 db
steps up to a maximum of 60 db. With the signal source about
100 ft away the attenuator setting was 21 db for an S 5 meter
reading with the R-Myers and my Transsystem 3733 converter.
This is the base reading for the remainder of the antenna tests.
I assumed that all the other antennas would not perform as well
as the R-Myers dish and would require a smaller attenuation
factor for an S 5 reading.


Results:

R-Myers dish (no. 1)------21 db
Ed's small dish (no. 2)----7 db
29 turn helix (no. 3)-------12 db
Small LPDA (no. 4)-------3 db
Larger LPDA (no. 5)------3 db
Ed's Large dish (no. 6)----10 db ***

***Ed's large dish had a different downconverter
mounted on the dish. Even though it was the same
type as used for the other tests, I am not sure
of the validity of this measurement.

Also note that the higher to attenuator setting, the
higher the antenna gain.


Discussion:

I was surprised by the difference between the 29 turn helix
(no. 3) and Ed's small dish (no. 2). I expected the dish to
perform better than the helix. The fact that the helix was
circular polarized as opposed to the linear polarization
of the dipole feed on the dish would give a 3 db advantage.
I don't know were the other 2 db advantage of the helix
came from. Everything else was generally as planned.
Notice that the large and small LPDA's performed identically
even though one was about 5 times larger than the other.
With LPDA's only the elements that are resonant at the
particular frequency that is being applied work to generate
a signal and the relative gain of LPDA's, no matter what their
size, is the same.


Converter testing


The next phase of the tests was to determine which converter
of four types had the best noise figure. All of these converters
except the Transsystem 3731 were converted to have outputs
at 144 MHz for a 2400 MHz input from the antenna. The trusty
old Yaesu receiver was put into the FM mode and an AC
voltmeter was connected to the line output jack. The noise
voltage with no signal was measured as a reference. The small
LPDA was used as the reference antenna because the signal
source was relatively strong and we wanted the noise at full
signal level to be measurable. We measured the noise at
maximum signal from each converter with the small LPDA
attached. That meant that the converter under test was
establishing the quieting figure that we were measuring.
The amount of quieting was measured in db below the
 reference noise with no signal.


The following converters were tested:

1. Transsystem 3733 converted for 144 MHz if
    and Murata filter added---------------------------------34.3 db.quieting
2. California Amplifier 130001 converted for
    144 MHz if and IF amplifier modified--------------------8.2 db.quieting
3. California Amplifier 31737 converted for
    144 MHz if and image filter modified with teflon tape---24.7 db.quieting
4. Transsystem 3731 with 122 MHz if and no modifications----5.5 db.quieting

Note that the higher the quieting number, the better
the converter.


Discussion:

I knew that no. 1 would be the best but I had measured
no. 2 to be as good as no. 3 with my makeshift test
equipment on my work bench. It was a big surprise
that no. 3 performed so much better that no. 2. In
my searches on various web sites the no. 2 converter
is supposed to perform nearly as well as the no. 1
converter. I have another unmodified 130001 converter
and I will thoroughly check it out to determine the
problem. Ed and others bought the no. 4 converter
at the Orlando hamfest last Saturday. It was supposed
to work very well unmodified but that is obviously not
the case. I will work on my 3731 to get it up to speed
with the other types.

Again, I want to thank everyone for coming. We had a
great time. I am especially grateful to Ed (W7LGD)
for having a spare jacket. I was miserable until Ed
came through. I will admit it was stupid-stupid-stupid
to not bring a jacket.

If anyone has comments please send them to me and
I will distribute them to the same mailing lists
that I am using to send this message.

73

Tony