[TowerTalk] slipping every so slightly off-topic

[Jim Jarvis]

Tom, TT:

While I agree generally that receivers tend to be non-linear,
and their AGC response curves tend to be non-logarithmic, as
well.... I'm going to pick a small bone.

Tom wrote:

>That why radios using DSP systems as primary selectivity are
> so poor within roofing filter bandwidth (look at the 756PRO
> or TS2000) and it's why PSK people whine when anyone comes
> near them with more than 20 watts into a wet noodle.
-0-

You incorrectly lump the 756PRO in the ts2000 category.

The reason that an unmodified PRO demonstrates distortion and
an apparent reduction of close-in IMD is NOT lack of primary
selectivity.  The roofing filter bandwidth is +/- 7.5KHz from
center frequency.  All signals within the roofing filter are
AM demodulated to create the noise blanker drive signal.  This
is amplified in a chain which is powered from the 8V rail.

The primary audio channel in the receiver is ALSO powered from that
8V rail.  It turns out that the noise-amp chain was inadequately
decoupled from the 8v rail...such that loud signals within the roofing
filter will modulate the DC powering the primary audio.  The result is
a low-frequency modulation of the audio output baseline.  (Thumps, burbles,
am detected voice and data images.)

K1KP developed the hypothesis.  Together we decided upon a test, which I
implemented in my 756PRO.  Results were published in 2001 in QST.  You
haven't
lived until you've heard a signal interfering with itself!

All it took was a 1000uF cap hung on the 8v supply, to eliminate the
problem.
The PROII's have several 330uF capacitors, and a larger inductor in
the feed to the noise-amp, eliminating the problem.  It required a board
redesign to fit them in.

As for PSK people, they tend to whine because they're using SSB bandwidths,
and relying upon the sound card to provide filtering.  Receiver AGC's go
nuts with inband large signals.   In addition, there is still a population
of
overdriven/nonlinear transmitters.  On PSK, the PRO will provide solid copy
of
invisible signals, using 50Hz bandwidth.

All of which drifts away from the "using a receiver as a measurement tool"
discussion.
LPL's technique is still the best....use a real wattmeter, hold S meter
reading constant
and vary the tx power.  eliminates rx behavior altogether.

N2EA
jimjarvis at ieee.org