There are multiple ways to do TDR. I like the way that this unit does
it -- an inverse FFT of a frequency sweep.
http://sdr-kits.net/VNWA3_Description.html
To expose small perturbations in the feedline or system, make the sweep at
VHF/UHF. To see only the more gross defects, sweep from about 50-150 MHz.
To understand this, remember that a linear frequency sweep will spend more
time in the high octaves than the low ones, so the greatest contribution
to the display will be that higher octave, whatever you have chosen.
Actually, to see the more gross defects, we would look at a lower frequency.
I don't want to imagine what my 160 stuff would look like at VHF. :)
A system can have a 500 ohm transmission impedance bump 1 foot long on 160
meters and it just doesn't matter. The general rule is if a reasonable
impedance discontinuity is less than one degree long, it will not upset the
system. SO 239's, for example, are about 35 ohms in the female's joining
spring part (the males are nearly perfect). The effect of that bump is
nearly immeasurable below 100 MHz.
We all know a one foot long chunk of wire that might be 400-800 ohms surge
impedance barely changes SWR and adds immeasureable loss between the coax
and a vertical base on 160. Same reason. Although the wire is a "major"
impedance bump, it is electrically not too long.
What we cannot tolerate on 160 are resistive series connections and low
resistance shorts shunting the system, or cross coupling from sharing common
currents.
On 160 meters, if we simply measure RF voltage across the input of a line
while sweeping low frequency, recording the repeating frequency of voltage
minimums, we can find the distance to any cable or connection problem
affecting the system by more than a few dB. Little lumps and bumps at VHF
might locate a future issue like a chewed shield, but for an existing signal
loss they are just a distraction.
If a shield develops high resistance 800 feet from my house, the high series
resistance will cause a repeating voltage null 800/492 = 1.626 MHz apart.
If I swept the line and saw repeating nulls spaced 3 MHz apart, I would know
an issue existed 492 / 3=164 electrical feet away. If the cable was .85 vf
, the issue would be 164*.85 = 139.4 feet away physically. This would be
true for an open or a short.
All the software and refinement does for locating major existing issues
affecting level is eliminate the use of a calculator. Any SWR measuring
device, or even a simple voltage indicator, could do the job. I can find a
bump affecting receive levels with a Heathkit VF1 VFO and a 1N34 diode and
meter about as well as I can with a TDR on 160. :)
_________________
Topband Reflector Archives - http://www.contesting.com/_topband
|