[TowerTalk] TH-7 Test Results

Ward Silver Ward Silver <hwardsil@WOLFENET.com>
Tue, 21 Jul 1998 20:30:47 -0700 (PDT)

Without including the entire discussion...

This is a valid question - why, even with high VSWR, did we include the
TH-7 and TH-6/7 data?

The suspicious thing about the 10-meter VSWR was not so much that it was a
little high, but that it was also flat, while the other two bands showed
normal curves.  Subjective, true, but to the experienced eye, indicative
of possible problems.  As it turned out, both the TH-7 and TH-6/7 had VSWR
curves that were shaped quite similarly, although the TH-7 was uniformly a
"half-point" higher.  Since the gain and front-to-back looked completely
reasonable on *both* antennas, we decided to go with the data and point
out the higher-than-expected VSWR.  Actually, most of the antennas showed
VSWR a bit higher than customary, or we probably would have been more
aggressive on the subject.

The VSWR correction applied (pg. 19) accounts for the extra losses in the
line and is routinely applied to any kind of gain measurement, so it
shouldn't be surprising that it is used in our report.  For the most part,
the corrections are minor as shown in the data sets of Appendix C.

This could be confusing as it's not discussed explicitly except by the
protocol in passing (6.1.2, pg. 37).  The signal source is held at a
constant power output during all signal strength tests.  The power was
read on the wattmeter.  VSWR was characterized in an earlier step.  So
there's no correction required for variations in signal source output.

There are caveats liberally sprinkled throughout the report in order to
warn anyone making a buying decision against taking the results as the
be-all and end-all.  As mentioned at least once - this is one antenna and
one site in one test.  I'd love to have the time and resources to have
done a half-dozen of each antenna...however...

As we point out in the report, one of the most useful activities any
interested parties could undertake would be to repeat the tests to see how
much variation is observed.  In that way, we could begin to improve the
protocol and measurement techniques.  It's a good thing to have the

73, Ward N0AX

FAQ on WWW:               http://www.contesting.com/towertalkfaq.html
Submissions:              towertalk@contesting.com
Administrative requests:  towertalk-REQUEST@contesting.com
Problems:                 owner-towertalk@contesting.com
Search:                   http://www.contesting.com/km9p/search.htm