Hi Bob,
On 3/18/2020 7:18 AM, Bob McGraw K4TAX wrote:
On point #2, the balanced feed line is twisted at my installation. This
is one advantage to the vinyl covered line. This makes essentially a
twisted pair which Bell Labs many years show is a way to minimize
induction into the feed line. In transforming from balanced to
unbalanced, I use two "baluns" at the shack end of the feed line. One
is a 4:1 dual core balun operating as a wide band transformer and the
second is a 1:1 balun operating as a balanced to unbalanced
configuration. It also provides a significant amount of common mode
current rejection. This has been verified by measurement of the common
mode current on the shield of the short jumper between the balun and the
ATU. {See https://www.dj0ip.de/balun-stuff/ }
Several points. First, twisting applies ONLY to coupling INTO the line
-- that is, the DIFFERENTIAL circuit.
Second, to be effective for that, distance between twists must be a very
small fraction of the wavelength. For example, the high twist rate
(called the "lay" by cable engineers) of CAT5 prevents very good
rejection of differential mode crosstalk into the low HF spectrum.
Measuring common mode current at the shack end of a line tells us
NOTHING about common mode current anywhere else on the line, because,
like any antenna, current varies along the wires that form it. In the
common mode circuit, the feedline is a single wire connected to the
center of the antenna on one end.
With no choke and no transformer in the shack, the connection is to
"ground," whatever that is. That causes the current to be near a maxima,
depending on what "ground" is, and current on the wire varies along the
wire (the line) like with any other antenna. A quarter wave up from the
maxima it will be a current minima and a voltage maxima; a half wave up
is another current maxima. This is not a transmission line effect, it is
an antenna effect; VF ~ 0.98, the VF of a wire with insulation.
An effective choke and/or a transformer at the shack end forces a
current MINIMA at that point, with maxima and minima along the line up
to the intentional antenna. A "balun" that is inductive or capacitive at
the frequency of interest will modify that current distribution; so will
a transformer that has capacitance between windings.
On 3/18/2020 7:18 AM, Bob McGraw K4TAX wrote:
> As to the research N7WS, I've read his paper several times. In fact,
> a fresh copy is on my desk. Due to test equipment limitations, he did
> his wet/dry line measurements at 50 MHz and higher with a short piece of
> feed line. I did my measurements between 1.8 MHz and 30 MHz with 100
> ft of feed line. The line was measured, for wet/dry purposes, in a
> matched impedance configuration between 1.8 MHz and 30 MHz. I agree
> with his data on velocity factor changes but can not agree on increased
> loss data.
Wes's choice of VHF is quite reasonable, and on a percentage basis, IS
representative of loss at MF and HF. Because loss for lines using
non-ferrous conductors increases approximately as the square root of
frequency, it's easier to measure at VHF, but the curves for all such
cables will be approximately parallel to each other. This can be seen in
the plot on page 2 of this link. All of the lines are approximately
parallel except for the kW twinlead (CCS conductors) and the coax with a
CCS center.
http://k9yc.com/Coax-Stubs.pdf
As to your tests between antennas at your QTH -- what matters is signal
to noise on RX, and the differences are likely to be in the 3-9 dB
range, depending on what noise sources are around you. To observe them,
we must do A/B comparisons on many WEAK signals, switching between the
antennas many times on each signal. Where this matters is copying weak
signals -- we won't hear differences on stronger ones. And it only
matters if your QTH is noisy.
73, Jim K9YC
______________________________________________________________
Elecraft mailing list
Home: http://mailman.qth.net/mailman/listinfo/elecraft
Help: http://mailman.qth.net/mmfaq.htm
Post: mailto:[email protected]
This list hosted by: http://www.qsl.net
Please help support this email list: http://www.qsl.net/donate.html
Message delivered to [email protected]
