Kevin Normoyle wrote:
> On 7/18/2010 8:18 AM, john@kk9a.com wrote:
>> BTW, it is pretty easy to design a 2el
>> yagi with 50 ohm impedance eliminating the need for the match.
>>
>> John KK9A
>>
>
> 40M 2L 50 ohm impedance? with reasonable gain/f/b/swr bandwidth? I don't
> think so, or I couldn't do it
> (with my boom length and full length elements)
> Am I missing something?
>
> Actually I did create one by accident. If you off-center feed the DE,
> you can get a 50 ohm match without too much loss in the other metrics.
> I thought that was interesting. I saw someone proposed that elsewhere,
> and it actually does work. It would be possible mechanically, but there
> is a loss in gain.
>
> Does anyone have a 40M 2L with 50 ohm input, with no matching
> components? Well I guess a moxon does it, and maybe some dual-driven
> things? what am I missing? I thought a simple 40M2L can't, as far as I
> tried.
There's really nothing special about "matching components" vs "antenna
feed method" to get a particular feed point impedance. most of the feed
schemes are, in effect, a lumped or distributed matching network that
happens to use the elements as part of the L or C, rather than a
discrete component.
Likewise, there's all sorts of schemes to make a physically smaller
antenna (drooped tips, linear loading, hats, etc.), all of which have
fairly subtle differences in performance from each other.
It really comes down to mechanical and loss considerations. Air
dielectric is pretty low loss. Aluminum is slightly higher loss than
copper, etc. But unless you get really exotic, they're all about the
same (as in fractions of a dB).
If you live in an area with strong winds and ice, you're going to choose
a physical configuration that's different from someone in the relatively
benign Los Angeles basin.
There are tradeoffs between variations in feedpoint Z, F/B (and to a
lesser extent gain) over frequency, so if these are important to you,
then you'd pick a design that minimizes the variations you find
particularly obnoxious.
One thing that I think modeling is particularly useful is checking the
sensitivity of the design to small misalignments and changes. Just in
looking at the dozen or so Yagi antennas I see on my daily commute, I'd
say that NONE of them have all the elements in the same plane. If the
proposed whiz-bang design only works with perfect alignments, then, in
practice, it's not going to be all that wonderful.
This is something that is very hard to discern with an actual antenna in
A/B type tests (because relatively few people have a calibrated antenna
range with no multipath, etc.).
I've started doing some measurements of the feedpoint Z of a HF mobile
whip on my car. So far, I've done the Z with the car sitting still, and
with a string pulling the whip back to the "driving at 60 mi/hr"
orientation. But what'd I'd really be interested in is the actual
variability as you drive in traffic.. how much does it change when
you're next to that semi?
I'll have to recruit my daughter to drive while I sit shotgun with the
laptop and VNA, but I suspect that there's fairly large variations, and
those variations dominate all the back and forth about sizes of loading
coils, base vs midpoint loading, etc.
_______________________________________________
_______________________________________________
TowerTalk mailing list
TowerTalk@contesting.com
http://lists.contesting.com/mailman/listinfo/towertalk
|