I am glad to read the other questioning responses.  I
was beginning to wonder if I was missing something.  The only way
one can invert frequencies is with some sort of product
detector/chopper/mixer, etc.

        I built a primitive broadcast automation system back in
1975 using a 567 as the tone decoder chip for the cuing system
at 25 HZ with no low-pass filter and it false-tripped all the
time.  The 567 will respond to frequencies that are 3 times the
oscillator frequency with the effect being that the output may
chatter but usually not fully go low.

        I fixed that problem with a circuit using a hex inverter
or a couple of stages of 7400 logic plus a NE555 timer chip.  I
think you could chuck both those chips and use a 12C508 today
with the idea being that your program must look for the 567 to go
low and stay there for a prescribed amount of time.  If it goes
back up, then the whole process starts again.

        My old cuing system worked perfectly after the extra
timing circuit was added.  I suspect I would have also had great
luck with a low-pass filter and maybe better selection of the
low-pass filter and timing components on the 567, itself.

        If you are going to use a NE567, be prepared for a bit of
cut and try time to really get it right.

        Also, if you use a 12C508 to handle the timing and
shaping fixes I was discussing, it may be possible to become more
creative with the programming as your 12C508
has enough program memory to more than handle what it needs to do
to emulate the RC timer circuitry using the NE555 I originally
described.

        I don't know if the 567 is considered to be obsolete as I
have not tried to buy any in many years, but I think the 12C508
signal shaper routine I described could turn a flaky and somewhat
marginal system in to a much better one.  Oh yes, the whole thing
can fit in to a 16-pin DIP socket with the only external
components being the resistors and capacitors for the 567 and a
bypass cap for the 12C508.

Martin McCormick WB5AGZ  Stillwater, OK
OSU Center for Computing and Information Services Data Communications Group

Olin Lathrop writes:
>Huh!!??  You're not really serious, I hope!
>
>Just in case someone out there might get confused, multiplying the signal
>by a constant of -1 does not change its frequency content or the relative
>amplitudes of different frequencies in any way.

--
http://www.piclist.com hint: PICList Posts must start with ONE topic:
[PIC]:,[SX]:,[AVR]: ->uP ONLY! [EE]:,[OT]: ->Other [BUY]:,[AD]: ->Ads