Aloha,
Following my confession/ignorance in not retuning my amp
during the tests yesterday, Jon wrote, in part:
> Yeah, you gotta retune for different loads.
> There will be SOME variation because, as others point out,
> the SWR meter is really designed to optimally work in a
>relatively 50 Ohm environment. Yet, in some ways that's
>contradictory because if you always had a 50 Ohm
> environment, why would you need an SWR meter?????.......
Per this point....
Rcv'd a very nice note from Bruce Knox, W8GN of RF Applications,
builders of my P-3000 meter, and the niffty new VFD meter. He
was stimulated by my remarks/ seeming problems using the
P-3000. In part, he said that the P-3000 should be viewed as
an "operating aid", and not a lab instrument. In fact, without seeking
his permission even, here is what he wrote to me, in part:
"Been watching your input on the VSWR discussion. Please
remember, we only have an 8 bit a/d and once we end up
doing the f+r/f-r and scaling, then displaying rounded results
to the nearest 10th, you'll inevitably see changes in VSWR
as you change power. I have also noted that the coupler
does indeed respond differently under changing power conditions
when loads are not 50 + j0 (perhaps up to .3 higher depending on
the reactance of the load). Even in to a nice dummy load, you can
see a 0.1 or 0.2 change just based on internal math and rounding.
I guess we'd need to up the design a bit for the unit to be used as a lab
instrument. As you know, it's mainly intended as an operating aid."
RF Apps various meters can be examined at:
http://www.rfapps.com
So, I apologize to Bruce and the company as my remarks were
off target from the intended use of the meter. I was really befuddled
by the increasing vswr, and am sure all was compounded by my
attempts to do these experiments up at 10 meters, where every couple
feet of coax brings a significant change in Z when the load is both
reactive and not 50 ohms either.
Jon, KE9NA, goes on to comment about Walt Maxwell's words:
> It sounds like the author (Maxwell) described very well what
>needs to occur for maximum power transfer. To restate,
>the amplifier must be conjugately matched to the impedance
>of the antenna system at the output of the amplifier. .....
> If a conjugate match does not occur for any reason then some
> power is reflected back to the amp and essentially is dissipated as heat.
So, today, repeated yesterday's efforts, but this time retuning,
and also replacing the Palstar AT4K tuner into the system. The
AT4K has a huge dual wafer switch to allow selection of completely
bypassing the tuner, connecting the rig to a dummy load, and
choosing either of two antennas, or a balanced antenna. Info/phtos
about the AT4K are at: http://www.palstarinc.com/palmain.htm .
Also performed the tests at 10 meters, then again down on 40 M.
Found that, indeed, more power was delivered by the linear
when it was retuned/load cap adjusted from the settings for
dummy load, or matched antenna system configuration(matching
attained with the AT4K tuner). However, as resulted yesterday,
significantly MORE drive power was required to the amp to
yeild equivalent watts into the non-matched 2.3 vswr antenna
system. This was true both with my beam at its' worst spot
on 10 meters, and with my rotating dipole at a frequency
on 40 M where it also gave 2 + vswr, though different
R and X values from those up on 10 with the beam.
Recall the beam Z values were R=10, X=20, swr=2.3 at
28055; the 40M dipole Z values are R=32, X=21, swr about 2.2.
On 10 meters, even with retuning the amp output circuits when
driving the "bare" 2.3 swr antenna system, I still had to
increase the drive power from 50 to get 1500 watts into
the dummy load, to about 65 watts to get 1500 watts
forward indicated into the un-matched beam antenna,
while also reading 250 watts reflected. So net out the
antenna was about 1250 watts. Yesterday, recall, 140 watts
drive was needed to net out 1500 watts from the antenna with
some 300 watts reflected.
Using the AT4K tuner, with which I could not quite reach
a perfect R=50, X=0 condition, it actually required ONLY
36 watts drive to read 1500 forward indicated on the Bird,
with about 100 watts reflected because of the "imperfect"
tuned match. So net up and out the antenna, about 1400
watts. Very interesting point here was the lessening of
required drive power below that needed into the
dummy load, why?
Repeated everything on 40 meters, where wavlength
and line run lengths should be of no matter, as they
clearly were up on 10 M.
Practically the exact same results:
Into the dummy load with 1500 watts and 52 watts of drive.
Zero reflected power indicated usning the 100 watt Bird
slug in the reflected position.
Into the AT4K matched dipole, 1500 watts forward indicated,
only 34 watts drive required, and found 55 watts being reflected.
Net out the dipole, 1445 watts or so.
Into the un-matched dipole with 2.2 swr, 1500 watts forward
indicated, 100 watts drive required, and 250 watts showing as
reflected back with best re-tuning of the amp for highest output
power under these conditions - still took the 100 watts drive.
My conclusions: Yes, you can drive an un-matched antenna
system, of probably up to 3:1 vswr or so without using a
tuner. But there is a cost: clearly higher dissipation
as more drive seems to always be required, at least that
is what I have found. If the amplifier has the suds and
you also have a strong power supply, you can put enough in
to get enough out, after subtracting the reflected power from
the indicated forward power to achieve the full legal 1500 watts
up and out the antenna. Many of today's amps available
to us can handle these operating conditions, as has been
suggested already by Dan, especially the rugged Henry
amps.
I have decided to continue to use the AT4K tuner. Seems
to be a lot easier on the linear and the system. Would
make no difference to your operation on the other guy's
receiving you; you would have extra steps in a band
QSY with the need to readjust the tuner as well as amp
tuning (unless you have one of the auto tune amps, now
Alpha Power even has an auto tuner built in to the new
model 87A, I believe). I do not contest seriously, only pick out
needed new "entities", etc. so do not need super quick
band QSY.
Found this all to be interesting, and I learned some things.
Hope you did also.
And I really wonder why less drive power is required into
the matched system than into the dummy load for the
same indicated forward power; especially since there
is zero, or close to it, reflected power when the
Bird resistor is the load ?
73, Jim, KH7M
--
FAQ on WWW: http://www.contesting.com/ampsfaq.html
Submissions: amps@contesting.com
Administrative requests: amps-REQUEST@contesting.com
Problems: owner-amps@contesting.com
Search: http://www.contesting.com/km9p/search.htm
|