[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: R: R: phase shift with coax

Domenicl (i8cvs) wrote:

> If the impedance of each antenna is 50 ohm and if i use two 50 ohm
> feedlines of any lenght but one 1/4 wavelenght longer than the other
> i get only the phase shift for circular polarization but not the
> impedance matching.

I agree with you.  But this was not what I was suggesting be done.

> Infact connecting both lines in parallel to a "T" connector i get an
> impedance of 25 ohm.

Again, we agree.

> In order to match 25 ohm to a 50 ohm feed line i need a transformer
> made by a line 1/4 electrical wavelengt long wich impedance is SQR
> ( 25 x 50 )= 35 ohm and i have to made it with tubing or using two
> 75 ohm lines each 1/4 electrical  wavelenght long connected in
> parallel.

I agree, and this is what the dimensions I quoted do.

If D is the inner width of the outer, square aluminum tubing, and
if d is the outside diameter of the inner, round brass tubing, then
the impedance of the "transmission line" consisting of the round
tubing centered inside the square tubing is:

Z = 138 log (1.08 D / d)

For the figures I quoted ("standard" available sizes, which we
will show to be close, but not perfect to give us the desired
impedance), we get:

Z = 138 log (1.08 x 0.875 / 0.500) = 138 log(1.89)

Z = 138 x 0.27646 = 38.15 ohms. 

Pretty close to 35 ohms.

Note that with the cable method, you are transforming 50 ohms at
the antenna to 100 ohms at the two ends of the T connector, so
that the parallel connection of two 100 ohm feedpoints becomes
50 ohms again.  And that requires that the 1/4 wavelength cables
be Sqrt(50 x 100) = Sqrt(5000) = 70.71 ohms, which is pretty close
to 75 ohms, but again not exactly right.

> This is more complicated than to match the impedance using two 75
> ohm lines each 1/4 electrical wavelenght long as used with the
> original phasing harness on the A144-20 T antennas.

In the final analysis, it's more work to do it with the power divider
(unless you just buy one!) because you have to fabricate the divider,
and still have to fabricate the two 50 ohm cables that differ by
1/4 wavelength, versus fabricating (at a minimum) two 1/4 wavelength
75 ohm cables and one 1/4 wavelength 50 ohm cable.

The advantage is presumably lower loss and better impedance matching
because of the inherent low loss of the air dialectric transmission
line vs. the low-velocity-factor, higher loss polyethylene dialectric
of the RG-59, the fact that you might not know the REAL velocity factor
of the piece of RG-59 you buy, making the 1/4 wavelength sections the
wrong length, and the unknown impedance and loss characteristics of the
T connector (some may be good, but I've heard stories of some T
connectors working really poorly).  But at these frequencies, I doubt
there would be a noticeable difference unless the T connector was one
of the real bad ones, assuming everything else was done correctly.

A very reputable US manufacturer of high-performance VHF/UHF/Microwave
antennas, M Squared, recommends the power divider approach that I
described when building stacked arrays of antennas.  But they also
use 75 ohm coax with high-quality, weatherproof F connectors on the
coaxial baluns of their folded dipole driven elements, on all four
bands of their Yagi's that I own (50, 144, 222, 432 MHz).  When
stacking their HO-Loop antennas, they offer both options, the 75 ohm
coax method and the 2-port power divider method.  When I bought my
pair of SQLoop antennas from them years ago (the predecessor of the
current HO-Loop), they said that they most commonly sold stacked
pairs with the coax cable impedance transformer and closer than
optimal spacing because it was less expensive, and gave acceptable
performance.  They offered me the option of building my stacked pair
with the 2-port power divider, LMR-400 phasing cables, and an
optimal stacking distance, and that's what I went with.  At the
time, I had no Yagi nor any way to rotate one, so I wanted the
extra little bit of performance on 2M SSB that the more expensive
solution provided.  There's no way I could claim that the difference
was measurable, since I didn't buy one of each configuration.  But
it was a neat way to get a horizontally polaraized omnidirectional
antenna system that served me well before I got a tower and rotatable
directional antennas installed.

So it's all a matter of personal choice.  (I really hate to assemble
RF connectors onto coaxial cables, but I could be talked into either
method myself, depending on circumstances.  How's that for indecisive?)

73 de KB0ZEV
Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org