|[AMRadio] Infinite impedance detector vs diode,
k4kyv at charter.net
Fri Sep 5 01:33:18 EDT 2008
Interestingly, the above slew rate discussion explains the fallacy of using
a diode in series with the modulation transformer secondary to prevent
overmodulation of a plate-modulated final.
The exact same slew-rate phenomenon caused by the discharge rate at the
output in a diode detector also occurs in a plate modulated final when a
diode is placed in series with the modulation transformer secondary to
achieve negative peak clipping.
Think of "X" in figure 1 as the modulation transformer secondary in series
with the rf final power supply. D1 is the clipper diode, which may be
placed in series with either the top end or the bottom end of the modulation
transformer secondary. C1 is the rf by-pass capacitor in the final, plus
any additional capacitance in the splatter filter. R1 is a resistor
equivalent to the class-C final.
When a negative peak from the modulator exceeds the DC plate voltage on the
final, thus causing overmodulation in the negative direction, the diode
should cut off, but the discharge rate of the capacitances associated with
any splatter filter circuitry and the RF by-pass capacitor in the final, is
too slow to follow the instantaneous descent of the negative peak, so the
diode becomes reverse biased and turns off prematurely. The voltage on the
capacitors in the final continues to bleed off at a slower rate while the
negative peak from the now-disconnected modulator resumes its upswing, and
the descending discharge voltage on the capacitor(s) eventually "collides"
with the rising positive voltage output from the modulator. This results in
the identical sharp point on the tip of the wave form of the negative peak,
as shown by the green trace in figure 9 and the red and black traces in
Because of this, even with a good low pass splatter filter (or maybe a
better term would be modulation transformer destroyer) between the
modulation transformer and the final, this waveform distortion may cause the
splatter to be worse than it would be if there were no series diode in the
circuit and the final were simply allowed to be overmodulated.
More information about the AMRadio mailing list
This page last updated 20 Oct 2017.