I just got my copy, too, and the report is terrific! Great job, Steve & Co.
I have some minor bones to pick, but highly recommend this report to all,
especially those intending to purchase a commercial tribander. The $15 + S/H
is cheap for the quality of this research. I really hope that others will be
inspired to conduct similar tests.
Some comments:
1. The way I read it, the only place that the 2.2 dB and 6 dB subtractions
were used was to adjust the EZNEC data modeling the patterns of a
theoretical 2-element yagi. This was done because EZNEC outputs in dBi, and
presumably the authors wanted to scale those numbers to the same range as
the actual test figures, which are comparisons to a dipole at 50 ft. In
theory, if you start with an isotropic (free-space omnidirectional)
radiator, a dipole in free space will show 2.2 dB gain over it at the peak
of the forward and rear lobes. Put that dipole over real ground, and in
theory you get another 6 dB due the groundwave reflection adding to the
skywave. The test results are comparisons to a dipole over real ground, so
in theory you would have to add 8.2 dB to compare them to an isotropic
radiator. Remember, that's in theory and has nothing to do with the real
world, where your mileage may (and will) vary. The report also says that
since most manufacturers present their figures in dBi, you would have to
subtract the theoretical 8.2 dB to compare the test results with the
manufacturer's claims (*if* they use dBi.) This may be of passing interest,
but as the report points out, may not be very useful. The real meat of the
report is the *relative* performance characteristics of the antennas under
test. All that dBi stuff is irrelevant to the comparative data.
2. A small nit about the actual test procedures: The protocol calls for
assembling the antennas according to the manufacturers' specifications, but
evidently a Force-12 B-1 "bead" balun was used with the TA-33, Skyhawk,
TH-7DX and TH-7DX (6). Although it shouldn't make any difference at all, I
would have preferred to see the tests conducted with the antennas configured
exactly as supplied (or recommended) by the manufacturers. For example, I
don't think a balun comes with the TA-33, and the TH-7DX comes with a BN-26
or can be used with a BN-4000B (like I do.) I think it's unlikely that the
average TH-7 owner is going to use a Force-12 balun.
3. Some antennas have different tuning settings, such as CW, phone, or "DX"
(somewhere in the middle.) The report doesn't specify which settings were
used for each antenna, and it would be nice to know that.
4a. The reported VSWR bandwidth results for the TH-7 are troubling. Judging
from the shape of the reported curves, the antenna was tuned for the
standard settings, which are optimized for phone. If you compare the
reported curves with Hy-Gain's published curves, the curves reported for 20M
and 10M look pretty much like the corresponding published curves, except
that they are raised considerably higher. The 15M curve looks correct for
about 3/4 of the band, but should start to tilt upward towards the high end
of the band. It, too, is raised considerably higher than Hy-Gain's version.
4b. I think the average reader might incorrectly conclude that Hy-Gain has
misrepresented the VSRW bandwidths for the TH-7, or at least that one cannot
expect broadband VSWR performance. I'm sure that the author's didn't intend
this, but they never really comment on what might have caused the VSWR to
deviate from the published specs. They simply comment that the VSWR was
higher than they expected on 10M, and that a TH-6 that had been converted to
a TH-7 had lower VSWR on all the bands than the stock TH-7 (but still higher
than Hy-Gain's specs for the TH-7.) The way it's written, one might conclude
that all TH-6's that have been converted to TH-7's will have better VSWR
bandwidth performance than stock TH-7's, but surely that's not correct! I
don't know the details on the TH-6 conversion and whether the resulting
antenna is identical to a TH-7, but I would have felt more comfortable
reading about either a properly performing stock TH-7 and converted TH-6,
instead of a defective TH-7 and a converted TH-6. Was the latter offered
because the former was broken? They don't say. The report says that VSWR
tests were used to make sure that nothing was wrong with each antenna, but
the VSWR curves for the TH-7 clearly do not come close to those specified by
the manufacturer. Why wasn't the discrepancy explained? Why wasn't this
antenna either fixed, eliminated or replaced?
4c. Just for reference, my own VSWR test results track so closely with
Hy-Gain's published curves that it's downright scary (my TH-7 is tuned for
the "DX" setting, and my test results closely match the published "DX"
curve.) There are some minor variations in magnitude, but the shapes of the
curves are almost identical. I get a 2:1 bandwidth of over 350 KHz on 20M,
over 450 Khz on 15M, and 1.7 MHz on 10M. When I was researching which
tribander to buy, I received VSRW numbers from many other owners that
correspond to my test results. Also, the ARP VSWR test results for the TH-7,
appended to Steve's report, agree with mine. I think that's pretty
compelling evidence that something must have been wrong with the TH-7 tested
by Steve's group. [Protocol: I took my measurements today with an Autek RF-1
on my own TH-7 at 50 feet on a tubular crankup (no guys.) I measured at the
base of the tower, with a little less than 80 feet of RG-213 between the
analyzer and the TH-7 (OK, 20 feet more loss, but the setup was pretty close
to the report's.) I confirmed all of the readings with an MFJ-259, then
confirmed again with 100 watts from the shack, measuring with the venerable
Nye-Viking RF-003 Power Monitor. Since I have 260 feet of LM400UF between
the shack and the RG-213 that goes up the tower, the SWR values were
considerably lower. However, the shape of the curves was virtually the same
as with the SWR analyzers.]
4d. Should we also be suspicious about the gain results for the TH-7,
especially on 10M? Intuitively, I think the extra element should boost the
gain at least a little bit over 20M and 15M. Yes, the traps and coax could
be lossier at the higher frequency, but it still doesn't seem right. I know
that a VSWR compensation figure was used, but the high VSWR on all bands,
especially 10M, still makes me suspicious about all of the results obtained
for this antenna. I think that's the downside of not explaining what was
wrong with the test TH-7's VSWR. In commenting on the high VSWR on the TH-7
(and the more normal curve for the converted TH-6), the report recommends
that adjustments be made to lower the VSWR, but this will have no effect on
performance. A few pages back, the report says that VSWR behavior, gain, and
F/B are closely related. I know this has to do with design tradeoffs, but if
the VSWR curves aren't following the manufacturer's curves, isn't it
possible that the gain and F/B won't either?
4e. The TH-7 is one of the top products in the large tribander market. The
VSWR anomaly, and the fact that nothing was done to explain it,
unnecessarily undermine the credibility of the report. I think something was
wrong with the TH-7 under test and that in all fairness, it should be fixed
and retested.
Did I mention that I have a TH-7...? ;-)
Otherwise, excellent report.
73, Dick, WC1M
--
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
|