[TowerTalk] Determining coax length?

Ray, W4BYG w4byg at att.net
Wed Nov 13 07:37:10 EST 2013

Great stuff!  Thank you for taking the time to do this and share it with us.
Your graphs and program are very instructive.
Ray, W4BYG 

-----Original Message-----
From: TowerTalk [mailto:towertalk-bounces at contesting.com] On Behalf Of Dan
Sent: Wednesday, November 13, 2013 3:31 AM
To: towertalk at contesting.com
Subject: Re: [TowerTalk] Determining coax length?

As K9YC said, coax VF changes as the frequency changes.  The TLDetails
program allows you to plot VF vs Frequency for any of about a hundred
different types of line, including Bury-Flex.  Here's a sample plot for two
different types of RG-58, one more lossy than the other.  The higher the
loss, the more the VF drops off at lower frequencies.


Also, the OP wanted to know how critical it is to have equal coax lengths.
Using AutoEZ I put together a model with two generic 10m 4-element Yagis
stacked at 35 and 70 feet (1 WL and 2 WL on 10m).  Each is initially fed
with 100 ft of Bury-Flex from the respective feedpoints back to a simulated
StackMaster II, modeled as an ideal transformer.  From there an additional
run of 50 ft of Bury-Flex goes back to the rig.

All of those dimensions are adjustable.  Here's a screen grab of the AutoEZ
"Variables" sheet.


And here are all the "non-antenna" parts of the model.


The Zo/VF/Loss values for Bury-Flex were automatically entered using this
dialog window.  All three characteristics will change if the frequency is


Then I used this dialog to calculate what 30 electrical degrees would be in
feet, using the VF of Bury-Flex at 28.1 MHz.


Finally, I set up a "variable sweep" to change the length of line to the
upper antenna by steps equivalent to 30 degrees.  That means I'm simulating
a mismatch in the line lengths, starting with no mismatch and ending at 180
degrees of mismatch.


After calculating all the rows this is what the "Patterns" sheet shows.  You
can use the spin button to "watch the movie" of the changing patterns as the
mismatch in line lengths grows.


To demonstrate use of the spin button, here's an animated gif.  The changing
value of variable "H" can be seen in the lower-right corner.  The outer ring
has been frozen so that the pattern magnitudes as well as shapes may be


Documentation and download links for both TLDetails and AutoEZ may be found


If you'd like to experiment with the free demo version of AutoEZ here's the
model file used in the above study.  (Note that you can't just click on the
link to open the model.  Save it to your computer then open it from within
AutoEZ, as explained in Step 3 of the Quick Start guide.)


Because of the large number of segments you won't be able to use the demo
version to calculate this model.  However, you _can_ modify the model
geometry by manually changing one or more variable values, perhaps changing
the antenna heights and/or the coax lengths to match your situation.  You
can then use the "View Ant" button to view the modified model, which means
that AutoEZ will create a temporary .ez format file, pass that file to
EZNEC, and invoke the EZNEC viewer.  You can then use the EZNEC "FF Plot"
button to show the pattern for that particular set of variable values.

As mentioned, the antennas in this model are just generic 10m Yagis that I
happened to have on hand.  If you have an EZNEC file for your particular
antenna please contact me off-list and I'll be happy to help you create a
"stacked, fed, and variable-ized" model for use with AutoEZ.  And that
applies even if you are using the demo version.

Dan, AC6LA

TowerTalk mailing list
TowerTalk at contesting.com

More information about the TowerTalk mailing list