Diodes (was 'Re[2]: [HBR] HBR -- Part 8')

[Walt Hutchens]

Dan said:
> I know that some receivers reduce the filament voltage on just the
detector
> with a series resistor.  One example is the Hallicrafters SX-100 which
uses
> two 6.8 ohm resistor in parallel for the 6BJ7 triple diode.   Also, the
Knight
> R-100 used a 3.3 ohm resistor to the triple diode  6BC7.  I wonder if this
may
> have been for the benefit of the noise limiter  section because the Drake
2-B
> used an 8BN8 for the noise limiter and first audio  amp powered from 6.3
> volts.  I always wondered what benefit that  the lower filament voltage
gave.
> Maybe you can shed a little more  light on the subject.

I don't know, but (assuming that these sets all have diode detectors and
AGC)
I will speculate that this is the same issue I experienced with the
diode in the AGC circuit: electrons bombarding the plate make it
negative, causing an AGC voltage even with no signal.  Lowering the
filament voltage would reduce the effect.

The output current capability would be reduced but for AGC that's 'who
cares?'  There would be an increase in plate resistance and a loss of
Gm for an amplifier triode in the same envelope but that's probably
unimportant in most receivers.

I can use a shunt resistor on the 19JN8. Although I won't do that long
term because the 1st IF shares the envelope, it would give an idea of
the magnitude of the reduction.

It could be that this is how that commercial receiver using no cathode
resistors, got away with it.

I'll bet if I can find my Radiotron Designers Handbook, it discusses
this issue.

I don't think this is likely to be a noise issue, since shot noise is
a microvolts-size issue and by the time the signal reaches a diode
detector or AGC rectifier, it's in the volt range.

Interesting question.  I didn't know about those receivers with
under-voltage diodes.

Walt
KJ4KV