To all,
I have received a number of very interesting replie to my questions.
Some of the issues are still open, but there is a lot of reference material
that I must yet sift through. All of your thoughts are appreciated. Below
is the gist of a note that I sent back to the chap that I had been
corresponding with, who desires to build an 80/160 inverted L with the trap
located at the top of the 80 meter vertical section.
Thanks again for all your help.
73 and happy holidays, Dallas W3PP
-------------------------------------------------------------------------------------------------------------------------------------
You have created a monster! Back in the 60's I wrote an article for the
PVRC bulletin about building and using coaxial traps. From that article,
Bob Sommer (can't remember the call) wrote a QST article that gave
construction details for coaxial traps. I then wrote a Turbo Basic program
(DOS) that calculated trap data from various input data. My approach has
always been empirical rather than theoretical. Now with the fancy
analytical tools available, a more theoretical treatise is relatively
simple.
I Googled coaxial traps and came up with some interesting data from
W8JI, W4RNL and VE6YP. An inquiry on the Antennaware reflector also yielded
some interesting points of view, and reference to several other works that
are useful in trap design.
Several interesting views were presented:
1. A 3 dB loss figure would result in tremendous heat generated in the
trap. This just doesn't happen. Cebic (W4RNL) relates the loss to radiated
signal and shows that it's in tenths of a dB and is primarily related to the
shortening of the antenna. The difference in radiated signal for lump
component traps vs. coaxial traps was insignificant.
2. The major difference in coaxial and lump component traps is as a
result of dielectric losses in the capacitance of coax as opposed to a
better dielectric constant with discrete capacitors. This is manifest in a
trap with lower Q.
3. Several authors (including VE6YP) indicate that the turns to length
ratio should be near .45:1. This value (they say) results in higher Q as a
result of less coax length on a larger form. Somewhere, I had made an
assumption (in my program) that a 1:1 ratio was best, and I flag designs
that fall outside of .67:1 and 1.33:1. Not sure why I made that assumption,
but my best recollection is that it had to do with maximizing the Q. That
may have been from standard solenoid theory.
4. The data yielded by the VE6YP program tracks very closely with mine.
The 3 major differences that I note are that I use a longer pigtail
(although my program allows that length to be changed whereas his does not),
I have adjusted the mean diameter of the coil to take into account a slight
compression of the coax. And finally, I use the actual jacket thickness as
an input criteria, whereas he does not (at least there is no place to input
that parameter).
5. Almost all commenters, and references that I have read, indicate
that coaxial trap loss is highly overstated. I think the real question is -
how does trap loss affect radiated signal. That is difficult to assess, and
Cebic comes closest with his theory of shortened radiator.
One additional thought that I had for your installation was that the
trap loss might actually work in your favor. Trap loss has little or no
effect on your radiated signal on the 80 meter portion of the antenna. The
trap is at a current null on 80. On 160, if there is significant loss
(which I really doubt) it would tend to reduce the horizontal radiation,
probably a desirable thing.
From all of this, I have gained a lot of insight into modeling
techniques for traps, both coaxial and lumped. I am up to my ears in
antenna design projects, but I have managed to devote some time to this old
favorite, and will continue as time permits.
_______________________________________________
Antennaware mailing list
Antennaware@contesting.com
http://lists.contesting.com/mailman/listinfo/antennaware
|