[TowerTalk] FW: Moxon rectangles

[Jim Lux]

----- Original Message -----
From: "K4SB" <k4sb at bellsouth.net>
To: <towertalk at contesting.com>
Sent: Friday, March 25, 2005 8:20 PM
Subject: Re: [TowerTalk] FW: Moxon rectangles


> Tom Rauch wrote:
>
> > I spoke against including a "resonance" function in the
> > MFJ259. The reason is you cannot measure resonance of an
> > antenna through a length of feed cable unless you know a
> > great deal about the feedline.
> >
> > All those who think they are measuring resonance of an
> > antenna when they measure zero X at the end of a
> > transmission line without understanding special
> > qualifications are assuming incorrectly. The only exception
> > would be if the antenna was 50 J 0 at resonance, in which
> > case the feedline could still introduce false zero X
> > crossings.
> ------
> This may be a case of a simple ommision on Tom's part or an assumption
> that the rest of us know as much as he does.
>
> However, if you use an electrical 1/2 wave line connected to the
> antenna, the impedence of the antenna will reflect exactly at the MFJ.
>
> As a matter of course, I keep good quality RG58 lines for all bands
> just for that purpose. I might also add that it does not matter what
> the characteristic impedence of the test line is. Also, while
> constructing it, if you use a high multiple ( such as
> 140 mHz ) for a 40 meter line, you get a much sharper dip.


Mind you, making a piece of coax that is 180 degrees (or an exact multiple)
for an entire band is a bit of a challenge.(except for the trivial case of
zero length and flipping the terminals)  Take 3.5 to 4 MHz as an example.
The length would be 12.5% shorter at the high end, compared to the low end.
At any frequency other than the one for which the line was cut, it will
create some impedance transformation, which might just happen to make a
reactive load at one end look resistive at the other.

And as Tom points out, it gets a bit tricky to back this out (unless you
have something much fancier than a MFJ antenna analyzer.. most network
analyzers can do it).  Numerically, if you put 86.6 feet (about 180 degrees
at 3.75 MHz) of RG58 feeding a 25+j0 load, at the bottom end (3.5 MHz)
you'll see 28 - 7.4j  and at the top end (4.0 MHz) you'll see 29+7.2j

Trying another example (using XLZIZL), I set up a series resonant LC  at the
end of the same 86.6 ft of RG-8.  Holding the L constant at 100 uH and
changing the C from 25 to 15 pF here's what we get
LuH  CpF  fresonant fzeroX
100 25 3.18 MHz   betw 3.8 and 3.85
100 20 3.56 MHz   betw 3.55 and 3.6 and betw 3.85 and 3.9 (i.e. two
crossings of zero x)
100 18 3.75 MHz   betw 3.75 and 3.8 (pretty close, because that's what the
line is cut for)
100 15 4.11 MHz   btw 3.6 and 3.65 AND another X zero crossing somewhere
above 4.0


Looks like Tom is right.. quite the pain to get it right, and distinctly
non-intuitive.

However, IF you know what frequency you want to resonate at, AND you cut
your transmission line for that frequency, then the zero X will occur at
that frequency (and probably some others).  Just don't expect to be able to
tune any arbitrary dip frequency with a single piece of coax.