On 4/22/20 8:00 AM, Grant Saviers wrote:
Good point. Modeling Yagis and optimizing with Auto-EZ is instructive
in how much one change changes a lot else. Watching the automatic
optimizer try to find the several dimensions of target values shows
this. Having designed a two band interlaced 30/17m I gained enormous
respect for those designers with proven performance multi-band designs.
A lot of hours at the PC. Plus I wonder about how designers got it
right in the "old days". My solution to the mount modeling problem is
all insulated elements.
In the "old days" people would make all sorts of claims, regardless of
whether the antenna worked well or not. If the sunspot number is >200,
given 24 hours you can probably "work the world", and things like crummy
F/B or gain that's 4 dB less than you think don't make a lot a difference.
That's why ARRL started requiring either range measurements or a model
in ads that claim gain.
Beyond that, lots, and lots, of cut and try. If it's your *job*, e.g. at
NBS, in pre-computer days, you've got some theory to work with, so
you've got analytical equations for mutual impedance as a starting point
(and tables for the Si() and Co() integrals you need).
Then, you make up a test matrix and systematically vary lengths and
spacing and make lots of measurements. That's what a "good" antenna
range has over a "not so good" - the ability to run the AUT up and down
quickly. If you can run an azimuth pattern in, say, 10-15 minutes, and
then change lengths in 10 minutes with another 5 minutes to run up, and
5 minutes to bring down, you can get a dozen tests done in a day.
A couple months later, you'll have the data you need to produce an
"optimized" design.
You'll have previous knowledge of the range, so you know how high to put
the AUT so that the "ground reflection" effects are known. At JPL, the
two longest ranges (1200 and 3000 ft) that are regularly used are over a
canyon between the transmitter and receiver and both have adjustments of
the height of both ends. But still, there are anomalies when someone
piles a bunch of steel light poles in a flat spot half way between the
end points.
https://mesa.jpl.nasa.gov/1200_Foot_Range/index.html
We haven't done many HF antenna measurements on the JPL range, though.
UHF (400 MHz) and up, for the most part.
One can see why modeling is popular.
Grant KZ1W
On 4/22/2020 06:38, Chuck Dietz wrote:
Hi Tom,
I don't think you are taking the right approach. I don't think you can
just
"tune" one element of a Yagi antenna. You need to measure all the element
sections and placements including the element mounts and run them
through a
computer design program. Changing any dimension will affect the other
dimensions to some degree. If it is a commercially sold Yagi, the file
will
already be available and can be used to optimize the antenna.
Chuck W5PR
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|