On Sun, 04 Apr 2010 20:46:21 GMT, Owen Duffy wrote:

Bob wrote in
:

Anyone know the velocity factor of JSC #1317 450 ohm line, 18 AWG?
Googling seems to give a variety of answers, and it's not posted at
the JSC site.

Wes, N7WS, measured some Wireman lines similar to that above. His
measurements indicated Zo quite different to nominal, and velocity factor
around 0.9.

For applications where velocity factor is important, eg the 'matching
section' of a G5RV, I suggest you measure the actual cable.

I'm plugging the velocity factor figure into Cecil's program for
optimum feedline lengths on a multiband dipole, IMAXMIN.EXE. Given the
approximate nature of this kind of feed, a ballpark figure is probably
okay.

Bob
k5qwg


Wes's data is included in TLLC (http://www.vk1od.net/calc/tl/tllc.php).
Your cable has similar stranding to Wireman 551, but velocity factor will
depend on the detail of the dielectric extrusion and punching. If JSC is
the manufacturer, they may even be the source of Wireman lines, in which
case Wes's data may be directly applicable.

I have reservations about the adequacy of copper cladding on the cable
such as yours at the lower end of HF.

Owen

Bob wrote in
:

....
I'm plugging the velocity factor figure into Cecil's program for
optimum feedline lengths on a multiband dipole, IMAXMIN.EXE. Given the
approximate nature of this kind of feed, a ballpark figure is probably
okay.

Bob,

Have you seen my article "Optimum length of ladder line" at
http://vk1od.net/blog/?p=949 ?

Owen

On Apr 4, 4:19*pm, Bob wrote:
I'm plugging the velocity factor figure into Cecil's program for
optimum feedline lengths on a multiband dipole, IMAXMIN.EXE. Given the
approximate nature of this kind of feed, a ballpark figure is probably
okay.

Yes, given all the variables, adjusting the final length, sometimes by
a few feet (depending on wavelength) is almost always required to
achieve system resonance. Remember that this approach is designed to
eliminate the tuner and therefore eliminate tuner losses and it is
designed to be used with a 1:1 current-choke-balun. Owen's comments
are certainly valid for systems using antenna tuners and 4:1 baluns.
In fact, if one chooses a ladder-line length halfway in between my
"good" (current maximum) and "bad" (voltage maximum) lengths, one will
obtain the odd 1/8 wavelengths points that are recommended for use
with 4:1 baluns. Those points result in a ballpark impedance in the
neighborhood of Z0 +/- jZ0/4, e.g. 400+j100 ohms. For those who
understand a Smith Chart, a picture is worth a thousand words.

http://www.w5dxp.com/smith.htm
--
73, Cecil, w5dxp.com

On Mon, 5 Apr 2010 06:27:54 -0700 (PDT), Cecil Moore
wrote:

On Apr 4, 4:19*pm, Bob wrote:
I'm plugging the velocity factor figure into Cecil's program for
optimum feedline lengths on a multiband dipole, IMAXMIN.EXE. Given the
approximate nature of this kind of feed, a ballpark figure is probably
okay.

Yes, given all the variables, adjusting the final length, sometimes by
a few feet (depending on wavelength) is almost always required to
achieve system resonance. Remember that this approach is designed to
eliminate the tuner and therefore eliminate tuner losses and it is
designed to be used with a 1:1 current-choke-balun. Owen's comments
are certainly valid for systems using antenna tuners and 4:1 baluns.
In fact, if one chooses a ladder-line length halfway in between my
"good" (current maximum) and "bad" (voltage maximum) lengths, one will
obtain the odd 1/8 wavelengths points that are recommended for use
with 4:1 baluns.

The more I look at it, the odd 1/8 wavelengths is probably the way I
will go, connecting to my tuner's 4:1 balun. There will be a 130 foot
flat-top, and the 450-ohm feedline length can be somewhere between 50
to 100 feet or so. Tnx for the input!

Bob
k5qwg

Those points result in a ballpark impedance in the
neighborhood of Z0 +/- jZ0/4, e.g. 400+j100 ohms. For those who
understand a Smith Chart, a picture is worth a thousand words.

http://www.w5dxp.com/smith.htm

Bob wrote in
:

The more I look at it, the odd 1/8 wavelengths is probably the way I
will go, connecting to my tuner's 4:1 balun. There will be a 130 foot
flat-top, and the 450-ohm feedline length can be somewhere between 50
to 100 feet or so. Tnx for the input!

I guess then that you didn't look at the article I quoted.

Typical T match ATU's are lossier on capacitive loads than on inductive
loads.

The odd eighth wave rule of thumb is a popular one. But, alternate odd
eight waves (on a resonant load) assures the highest ATU losses for the
given SWR.

These rules of thumb, and there are plenty that are conflicting, are
usually given without explanation of why they work. We are a gullible
lot!

The same occurs with 4:1 voltage tuner baluns which anecdotal evidence
suggests assist match of a wider range of loads. There is good reason to
think that the mechanism behind this is that their own loss assists, and
it is an inefficient work-around for another problem.

Owen

On Mon, 05 Apr 2010 20:33:47 GMT, Owen Duffy wrote:

Bob wrote in
:

The more I look at it, the odd 1/8 wavelengths is probably the way I
will go, connecting to my tuner's 4:1 balun. There will be a 130 foot
flat-top, and the 450-ohm feedline length can be somewhere between 50
to 100 feet or so. Tnx for the input!

I guess then that you didn't look at the article I quoted.

Actually, I did look at the article.

It mentioned the voltage maximum problems, the current maximum
problems, and then said, "Is there a better option?" And I don't
understand the few sentences that follow that query. In other words, I
don't understand the solution -- i.e. "line lengths around 135
degrees longer than voltage maximum" :-)

Bob
k5qwg


Typical T match ATU's are lossier on capacitive loads than on inductive
loads.

The odd eighth wave rule of thumb is a popular one. But, alternate odd
eight waves (on a resonant load) assures the highest ATU losses for the
given SWR.

These rules of thumb, and there are plenty that are conflicting, are
usually given without explanation of why they work. We are a gullible
lot!

The same occurs with 4:1 voltage tuner baluns which anecdotal evidence
suggests assist match of a wider range of loads. There is good reason to
think that the mechanism behind this is that their own loss assists, and
it is an inefficient work-around for another problem.

Owen

Bob wrote in
:

....
It mentioned the voltage maximum problems, the current maximum
problems, and then said, "Is there a better option?" And I don't
understand the few sentences that follow that query. In other words, I
don't understand the solution -- i.e. "line lengths around 135
degrees longer than voltage maximum" :-)

The location of voltage maxima depends on the load on the line. If you
were to plot the impedance at various lengths of line, it is highest (and
purely resistive) when fed at a voltage maximum. As the line is lengthed,
that impedance becomes capacitive, and lower, eventually becoming lowest
(purely resistive again) at the current maximum (90° longer than the
point of voltage maximum). Increasing the length further, impedance
becomes inductive and increases eventually becoming highest at the next
voltage maximum. At a point of about 135° longer than the voltage
maximum, the impedance presented to the T match is in the region where it
is most efficient. Alternatively, you could state this as 45° shorter
than a voltage maximum.

This is not your odd eighth wave (from a resonant load) rule, because
that also encourages the capacitive region where losses are higher.

Owen

On Apr 5, 3:33*pm, Owen Duffy wrote:
Typical T match ATU's are lossier on capacitive loads than on inductive
loads.

How about typical CLC Pi-Net ATUs?
--
73, Cecil, w5dxp.com