Hello Clive and the list.
Yep, been there, done that. Got the same results. Now I model them
with
EZNEC and make adjustments in the computer FIRST then build.
Here are a few tips to keep in mind. They are the result of my learning
curve and are based entirely on my own methods of limiting my failure rate
with new antenna design or building.
Decide which type of wire you intend to build with and if possible stick
with it. If you are like me, I have used all types of wire in the past from
common single house hold wire to lamp cord, to 7-22 stranded bare copper
antenna wire, to copper weld. My favorite is from the "Wireman" and is a
copper clad steel twisted flexible #16 or #14. Very strong and flexible and
insulated to keep the corrosion off it with a tough coating that the birds
won't peck through.
Build a test dipole to get a handle on the effects of the insulation if
you
use insulated wire. Say for 10 meters where you can waste less material and
still be able to erect the antenna at say 1/2 wave using simple supports for
testing. IE: PVC pipe or two trees or ? If the formula and computer say
that a 1/2 wave #14 wire at 1/2 wave above ground fed with an electrical
wave length of cable (so that the readings at the end of the cable are the
same as the antenna is reflecting) is length X, and when you build it and
test it you are able to agree with the results, then you are ready to build
an identical dipole with the wire you intend to use and compare resonate
points to see how much (as a percentage of length) you will have to adjust
for the dielectric change the insulated wire makes. If you use the same
wire regularly for all of your finished products antennas then you now know
the constant for future projects.
Trust the computer predictions in EZNEC and build accordingly.
Find and duplicate consistently your own method of securing wires to
insulators so the result are predictable. If end support wires are bared for
soldering, then use the exact same technique for attachment. One turn thru
the insulator, loop diameter same, 6 wraps over the wire with the loose end,
(whatever your method is) be consistent. This end effect may not be
noticeable with say a long dipole on 80 or 160 meters but by the time you
get to say 12 or 10 meters is will have a pronounced effect in the correct
length/resonate frequency of your finished product. Buy the time you get to
6 meters you will be amazed at the effects so be consistent. 1/2 inch isn't
much at 80 meters but it sure is significant at 6 meters!
Last year I built a PVC supported 6 meter wire moxon. I used the
formula
based on a #18 wire. This assumes a BARE wire. When I built it I ended up
way off freq because although the wire used was approx #18, it was called
COMO wire and is the insulated, camouflaged wire used 20 years ago by the
Military. Copper Clad Steel Stranded with Teflon jacket and cloth braid. I
had to shorten the lengths approx 5 inches to get the correct resonate
frequency. Once it was adjusted by the same percentage on both elements, it
tuned perfect and performed just as EZNEC said it would.
Keep metallic away from the elements. They will seriously detune the
antenna especially the closer to the ends they are. Within say 10 to 15% of
the center of each element? Not much of a problem.
Extreme changes will take place if you screw with the end spacing of the
elements to one another. That places the High Voltage ends of each element
VERY close to each other and has a pronounced effect. Use a HIGH dielectric
insulator and limit sharp points. No PVC! I have had good luck using the
same small porcelain dog bone insulators for all my antennas and using a
small two hole set screw wire ground strip as the anchor at each end and
slipping a 3 inch piece of heat shrink tube over the loose end and the
antenna wire to keep the loose end secured and free from water migration.
Look at the "Wireman" pages to see a similar example. Same method,
different parts.
Test feed line should be a multiple of an "ELECTRICAL" half wave long at
the test frequency. I usually use one wave times the velocity factor of the
cable and "shoot" it with the MFJ-259B and confirm the freq with a receiver
in the shack.
Use these tools and constant methods of construction and you will
produce
predictable, consistent results.
If you have not invested in a good NEC antenna modeling program, make
the
investment and learn how to make it "sing". It will save you more "Grief
and aggravation" in building and designing your antenna systems both now and
in the future. Computer Aided Antenna Modeling is THE single best use of a
computer for all grades of licensee in my mind. YMMV. ;>)
BTW, I have some 13 ft long Crappie poles coming via UPS soon and a
spool
of RG58U with stranded center and poly insulation to arrive shortly and I am
going to build either a Moxon Stack using the same method as the
"Spider/Hex" beam or a HEX beam using the RG58 coax as the elements fed and
connected in Bazooka fashion to acheive linear loading and make the array
about .66 size of full size! I'll post my results when it is done IF I am
successful.
Best 73 and good luck with your project!
Mike Baker K7DD
k7ddmjb@qwest.net
-----Original Message-----
From: towertalk-bounces@contesting.com
[mailto:towertalk-bounces@contesting.com]On Behalf Of Clive Whelan
Sent: Friday, March 25, 2005 11:03 AM
To: towertalk@contesting.com
Subject: [TowerTalk] FW: Moxon rectangles
[ posted in error to antennaware, forwarded here for any other help]
-----Original Message-----
From: Clive Whelan [mailto:clive.whelan@btinternet.com]
Sent: 25 March 2005 11:32
To: antennaware@contesting.com
Subject: Moxon rectangles
Hi list
I am a newbie here. I have searched the archives for a clue to my problem
without success. So what is it?
I have made and erected a 15m Moxon using the dims. from W4RNL's site. The
construction ( a prototype) consists of a small (1x1 metre) X of 20x20mm Al
angle, supporting bamboo canes to about 9ft ( sorry about the mixed units!),
and then 1mm wire ( probably about AWG14G?) stretched around. The trial
height is only about 20ft. The performance is disappointing; that is to say,
if it were a rotary dipole at 20ft I'd be pleased, but there is no f/b (
fixed N/W for the trial), and certainly no fwd gain ( cf a simple vertical).
In fact, I suspect-if anything- it is firing off the back, as the 5Z4 beacon
is a cracking signal, being well up on the vertical. Checking with the MFJ
analyser, I was shocked to find it resonant at around 20.25Mhz- this would
be consistent with the reflector acting as a director, but with very poor
efficiency. So, we took the whole thing down, and double checked everything,
finding no errors. Now something even more puzzling. Checking further HF
with the MFJ, I find sharp "resonances" around 63 and 106Mhz, being roughly
the third and fifth harmonic of an in band 21Mhz resonance, but which does
not occur at the fundamental frequency, as described above.
I have some nice fibreglass spreaders for the final job, but there's clearly
no point in proceeding until I can make the prototype work. Oh yes
everything checks out FB in EZNEC which indicates that the dims. ARE
correct, but the &^%$£*** thing just doesn't want to play
Any advice appreciated.
73
Clive
GW3NJW
_______________________________________________
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless
Weather Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any
questions and ask for Sherman, W2FLA.
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
_______________________________________________
See: http://www.mscomputer.com for "Self Supporting Towers", "Wireless Weather
Stations", and lot's more. Call Toll Free, 1-800-333-9041 with any questions
and ask for Sherman, W2FLA.
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|