Welcome to the Ten Tec group Ed. I appreciate the time you have
taken to help us understand the effort the League goes to to get solid
performance data on our gear. I knew it was a daunting task but had
no idea how much so. Thank you!
In the referenced post you bring up another matter that I have been
thinking about for some time. See below..............
On 6 Mar 2003 at 17:47, Hare,Ed, W1RFI wrote:
SNIP
>
> But I remember one year I was mobile with my Ten Tec Omni D during
>the ARRL Phone Sweepstakes. I parked right under the W1AW
>antenna array and was able to work 'em on phone while they were
>blasting away on CW, at least on 20 meters. That rig is still my
> main rig at home, btw. :-)
>
In some ways I think the design of our gear may be going backward.
Take Ed's comment about his Omni D for example. The Omni has a
double tuned narrowband RF amplifier using real inductors and real
capacitors. The arrangement uses rack driven slug tuned coils and
fixed value capacitors, ala Collins and Drake. The balance of the front
end is pretty conventional with a bouble balanced diode ring mixer and
a strong post mixer amplifier, then on to the noise blanker gate and the
IF filtering. One can peak the desired signal and reject lots of other RF
energy WITHIN THE BAND being worked. So......only a limited
amount of interfering signal power ever reaches the mixer. Relatively
speaking, the mixer and following stages have a pretty easy task in
staying linear because of the limited power being applied.
Now ratchet forward in time to rigs like the Corsair. The Corsair uses
a triple FIXED TUNED BANDPASS filter before the diode ring first
mixer and an even stronger post mixer amplifier. BUT, all the signal
energy in the band reaches the mixer. In the interest of convenience,
we have now given the mixer and subsequent stages a more difficult
task in staying linear.
Moving forward in time again we come to general coverage rigs like the
Paragon and most of the imports and what do we find? We find a fixed
tuned front end filter that is often many megahertz wide! To their credit
Ten Tec did a good job with the Paragon. I think at the time it was
tested at the ARRL lab, it demonstrated a very strong front end but the
test [if I understand what I think I know] only included two frequencies.
In the real world, there are hundreds of signals hitting the mixer if it is
proceeded by a wide first filter. So, in the interest of getting general
coverage, we have now given the mixer and following stages a very
difficult task in staying linear!
Then we have the RX-320 which has almost no front end filtering.
Everything from 100 Khz to 30 Mhz hits the first mixer! I would have
never given that arrangement a chance of working but actually it's not
too bad under most conditions. I doubt it would have worked in the
environment Ed described above though.
I believe Ten Tec, Patcomm and Elecraft reached a good compromise
with the Omni V and VI, Scout, Elecraft K2 and Patcomm 500 and
9000, all ham band only transceivers. But I suspect even those are a
compromise.
>From the viewpoint of signal handling ability it seems to me a receiver
with a dual or triple tuned narrowband first filter using real inductors and
capacitors, not varicaps and low Q surface mount inductors; and a
state of the art mixer and post mixer amplifier and roofing filter would
be an outstanding "real world" performer!
Which brings me to my point/question in this already too long post.
Are we going backward in our receiver design, Orion aside for the
moment?
Since the tests run at the League, and all other tests I have seen only
use two tones, are we subjecting the front end to what it will actually
see when is service? Does the second order intermod test tell us
everything we need to know? I'm not smart enough to frame an
intelligent question here but it seem intuitive to me that the more
unwanted signal power that reaches the mixer, the more problems we
are going to have with distortion products. How about a test using a
broadband noise source of known power in a controlled wide
bandwidth, combined with the test signals? The noise maybe should
be notched at the two test signal frequencies. Would that get us any
closer to real world performance figures or am I just wondering off into
the tall grass? :-)
OK, back to our regurarly scheduled program..............
--
73 de Lee WA3FIY
wa3fiy@radioadv.com
|