In a message dated 9/8/01 3:05:41 PM Pacific Daylight Time,
alsopb@gloryroad.net writes:
<<
Also had problems with adding 12m quad to a 10-15-17-20M 4 element GEM
quad. These four bands coexisted very well even with a single
feedline. Looked at it in AO and the patterns on all four bands were
reasonable. The model did show a degredation in F/B ratio relative to
a single band antenna. It didn't drop below 15db which was fine with
me. Front/side ratio was also less but still reasonable. The forward
gain seemed unaffected-- except on 10m where there was approximately
1db loss in gain. The azimuthal pattern was wider and appeared to be
smeared (two headed) due to contributions from the 20M elements.
These calculations on 10M were a bit suspect because of the limited
number of pulses allowed. The spreading of the pattern could have
been due to not having enough segments to numerically converge out a
numerical harmonic from the answer. Anyhow, direct A/B comparisons of
the 4 el quad on 10 (approx 20' boom on 10M) consistently outperforms
a 3el 12' boom antenna at the same height on another tower. That was
good enough for me.
When I tried adding just a 2 element 12M quad, things went completely
to pot on 10M. Measured impedance on 10 went wild. Attempting to
resonate the 12m quad was a failure. Reducing driven element length
significantly didn't alter the resonant point hardly at all! It
seemed clear that feeding the 10 and 12 driven elements at the same
time just was a bad idea. Was not able to model the five bands in AO
because of too few segments. Finally gave up and built a separate 2
el quad and put it on another tower with a TV rotor.
I'm not saying it can't be done. I did help a fellow put up a 2 el 5
band lightning bolt quad up. It used a single feedline and appeared
to work.
Perhaps there are some unique dimensions which make it look acceptable
from an SWR standpoint. It would be interesting to model the
lightning bolt quad to find out how the patterns look.
73 de Brian/K3KO
>>
Ok on your findings. I had a LB 2 element 5 band with all the DE's all tied
together and got exactly the same poor patterns in actual practice as I did
in Eznec. So if I find I can get great patterns in Eznec with them not tied
together on all 5 bands and even driving the off band DE, I have to assume I
can get these patterns in actual practice using reflectors. Now if you have
a 2 element with directors or a 3 element 5 band quad, this starts to get
sticky.
When I drove the off band DE and still got great patterns it proved that the
DE doesn't have to be resonant for the best pattern--by a long ways either
way. It's only convenient for a resistive load. When I drove say the 17M DE
with 15M RF the Rr values was much higher which is highly desirable but it
was reactive which open wire line and a tuner can handle. Who knows, it may
be a more efficient way to feed it. It certainly is on yagis and verticals
with ground systems.
No one has give me an answer why all the off bands high and or low don't
affect each other also--Why just 12&10M give a problem? 12M is 400 KHz
higher than the harmonic multiple of 3.5 MHz so it's a bit closer to 10M and
may be just enough dig into it? k7gco
I'll have to go back and look the reflector lengths I converted to a formula
for length for each band and see if they are all about the same or was my 12M
reflector retuned somehow for both bands. Since I didn't find a problem I
had no reason to retune the 12M reflector when on 10M or visa-versa.
List Sponsored by AN Wireless: AN Wireless handles Rohn tower systems,
Trylon Titan towers, coax, hardline and more. Also check out our self
supporting towers up to 96 feet for under $1500!! http://www.anwireless.com
-----
FAQ on WWW: http://www.contesting.com/FAQ/towertalk
Submissions: towertalk@contesting.com
Administrative requests: towertalk-REQUEST@contesting.com
Problems: owner-towertalk@contesting.com
|