Hi,

I've been playing with a circuit that I believe has some
potential, the lineup is as follows.

I'm turning an 0.5V DC offset, 0.5V audio signal at
1000 hz to modulate an RF 5V VCO where 0V-1Mhz, and 1V-2Mhz
The VCO drives a switched capacitor cell (1000pf) to create a
simulated resistance. This simulated resistance is the top
resistor for a series voltage divider, the bottom resistor
of which is 1M.

I send 20V to the voltage divider, the output of the voltage
divider is sent to a low pass filter to retrieve the audio
signal, in millivolts, with a slight DC component.

* C:\Program Files\LTC\SwCADIII\Draft1.asc
R3 N005 0 1e6
C1 N004 N005 1000p
V7 N006 N008 SINE(0.5 0.5 1000)
S4 N003 N004 N002 N001 MYSW
S2 0 N004 N001 N002 MYSW
R2 N001 0 1e12
V1 N003 0 20V
A1 N006 N007 0 0 0 0 N010 0 MODULATOR space=1e6 mark=2.2e6
V2 N007 N008 5V
R1 N009 N005 1000
C2 N009 0 1µ
R4 N008 0 1e12
R5 N002 N010 1000
C3 N002 0 330p
..tran 0 .001 0 .01
..model MYSW SW(Ron=1e-12 Roff=1e12 Vt=1.0 Vh=0)
..backanno
..end

There's at least a couple amazing things about this circuit:

1) If I adjust the top frequency in the VCO slightly, I have
a DC reading frequency meter. If I scale the VCO and DC output
correctly I could read 2.100 to 2.150 Mhz reading as .100 to .150
Volts, and so can display frequency digitally without needing the
expense of a frequency counter, only a cheap DMM module.

2) If I place an RF filter between the VCO and the switched
capacitor, I can decide what range of angles I want the 1000hz
audio signal to appear.

For example, if I have a lowpass RC
filter of R=1000 Ohm and C=330 pf, then only that portion of
the 1000 hz signal for a small range of angle surrounding the
negative peak of the input will appear on the output.

If I have use a highpass filter instead, I will eliminate the
entire signal except for the angles surrounding the positive
peak.

3) Suppose I use an arbitrary voice signal instead of the 1000hz
tone. I could use a lowpass filter to eliminate portions of the
signal going above an arbitrary voltage, meaning that I have the
beginnings of an AGC control or noise blanker. I could use a
pass filter to eliminate portions of the signal going below an
arbitrary voltage, meaning that I can eliminate an arbitrary portion
of the noise floor.

The only circuit I am aware of that has properties remotely similar
to this is an FM discriminator, and I think I am going into unknown
territory here...

Opinions?

The Eternal Squire

(yes, I have a call, but I prefer my privacy)

The Eternal Squire wrote in message ...
....
1) If I adjust the top frequency in the VCO slightly, I have
a DC reading frequency meter. If I scale the VCO and DC output
correctly I could read 2.100 to 2.150 Mhz reading as .100 to .150
Volts, and so can display frequency digitally without needing the
expense of a frequency counter, only a cheap DMM module.

Seems like a complicated way to do it, and it will have only the
stability of whatever analog components you're using. Why bother,
when a counter can be as cheap, fewer smaller parts, and have crystal
stability? See also various F to V converters if you really must do
it with analog; I think Jim Williams and Bob Pease have both published
some good work on them.