[Central-Electronics] RE: Vintage Ham Shack for Sale - Includes 100V

[Radra]

Hello Jim,

Sounds like you are serious about cleaning up your signals.  My experience
with the 100V was similar to what you encountered with the 20A except the
performance of the 100V is better than what you are encountering.

I did achieve about 55 dB suppression of the unwanted sideband in both LSB
and USB at 1 kHz.  The four electrolytic capacitors in the modulator were
replaced with a matched set.  The modulator was aligned per the "SIDEBAND
SUPPRESSION ADJUSTMENT" procedure in the 100V manual.  I employed a spectrum
analyzer instead of an oscilloscope to directly examine the RF output.  The
audio input signal was 0.5 Vp-p.  By careful adjustment of the AF Ratio and
AF Bal pots and the two 8 MHz phase shift coils, I could just achieve 55 dB
suppression in both LSB and USB modes.  I don't expect this adjustment to be
stable because the two pots are not mechanically stable.  A better solution
would be to replace both pots with fixed metal-film resistors in combination
with multi-turn pots as you did.  I didn't do it because I wanted to retain
the original 100V design.

The 55 dB value was achieved at 1 kHz and cannot be maintained across the
audio spectrum of 300 to 3000 Hz.  For example, at 1170 Hz, the suppressed
sideband level drops below 65 dB which is the transmitted noise floor.  At
2000 Hz, only about 48 dB of suppression is achieved.  Not only is
suppression a function of audio frequency, it is also a function of RF
output level.  Residual carrier will mix with the desired sideband signal to
generate its complement because the IF and RF amplifier stages are not
perfectly linear.  This mechanism is commonly referred to as spectrum
regrowth and is a prime consideration in the design of linear amplifiers for
cell phones.

Achieving a high level of suppression or isolation within a transmitter
requires that a number of factors be considered and optimized.  Providing
adequate shielding is difficult when low-level audio and high-level RF exist
in close proximity.  I think, given the design of the 100V, that a more
reasonable suppression target would be something in the low 40's dB across
the audio spectrum.

According to the panel meter, carrier suppression exceeds 60 dBc.  But
according to the spectrum analyzer, carrier suppression was on the order of
only 45 dBc or so.  I believe the discrepancy lies in the point where the
measurement is made, modulator verses final amp.  Probably what is happening
is that carrier is reinserted due to leakage.

Jim, I assume you are using an oscilloscope per the manual and not a
spectrum analyzer.  That may explain why you cannot observe better sideband
suppression.  The dc power within the 100V contains 60 Hz and its odd
harmonics and apparently the same is true for the 20A.  Because the
oscilloscope is time-domain, you observe the combination of power supply
artifacts (hum), signal (test tone) and suppressed sideband; the high-level
hum obscures the low-level sideband signal.  Use of a spectrum analyzer
enables one to readily distinguish between signal, sideband and hum because
they are displayed in the frequency domain.  A strong hum component is
clearly evident riding on the suppressed carrier.

Regarding HF antennas, a while back, I designed a communication shelter for
the Air Force.  Their goal was to disseminate weather information within a
couple hundred miles of the shelter.  I employed a Harris antenna which, as
I recall, looked like a two element beam.  Its spacing above ground was such
that energy from the antenna combined with the ground reflection so that it
was directed vertically.  Ionspheric bounce then returned it to the earth
within a radius of a couple hundred miles of the shelter.  This may be a
version of the "cloud burner" to which you refer.  I don't think the home
owners' association would permit me to erect one here.

Lyman K7LH