[TowerTalk] Hunting for a feedline fault

[David Gilbert]

Several weeks ago I laid four 150 foot lengths of LDF4-50A Heliax from 
my tower to just outside my shack.  I used quality Heliax-to-UHF 
connectors at both ends of the Heliax, and connected to flexible Davis 
RF BuryFlex coax to snake through twisty portholes to get into the shack.

Everything looked good at low power, but when I recently put 1500 watts 
to my 80m Inverted-V that has an SWR barely above 2:1 without the 
"tuner" (a simple 2 uH series inductor at the output of the amplifier) 
things went bonkers after about ten seconds.  The SWR went sky high and 
has stayed high now even at low power.  Clearly something in that line 
zapped because switching to a different run of Heliax (same length) has 
given no problems at all over several hours of heavy running.

The most likely problem would be with the connector at one end or the 
other of the Heliax, but since I had already buried the Heliax there was 
always the possibility that one of the damn rocks on this hillside 
crushed the Heliax at some other point in the line.  The question became 
how to narrow down the likely position of the fault.

This would obviously be a good application for a TDR, and my FA-VA5 has 
that capability when connected to a computer running the free DG8SAQ 
software, but for some reason the USB port on my FA-VA5 doesn't seem to 
function even though I've installed the correct driver.

The FA-VA5 does, however, manually read complex impedances quite 
accurately (and in my opinion is a steal for the price), so tonight I 
took R ± jX readings every MHz from 2 MHz to 30 MHz and recorded the 
values.  I then fired up TLW, the free transmission line software that 
comes with the ARRL Antenna Book, set the Cable Type to LDF4-50A and set 
the impedance box to "Input" instead of "Load". I inputted each set of 
values I had recorded (frequency, resistance, and reactance) and then 
checked "Graph", which plots the resistance and reactance the full 
length of the line.

As might be expected the lowest frequency readings were not terribly 
consistent, but for each reading from 4 MHz all the way up to 30 MHz the 
graph told me I had a short between 23 and 24 feet from the shack end of 
the coax.  There was no other graph point that was the same for every 
frequency .... only the zero resistance at 23.5 feet.  I'm pretty sure 
it is more than coincidence that 20 feet is my best estimate for where 
the BuryFlex meets the Heliax and I'd bet a lot of money that I botched 
the connection there.

This technique is probably well known to most folks here, but it hadn't 
previously occurred to me and I thought I'd pass it on in case any 
others aren't aware of it.  It wouldn't be definitive if the problem is 
neither open nor short and instead something in between (or if it was 
intermittent), but for hard faults it should be helpful.

73,
Dave   AB7E