pseudo-random shift generator

[David Halliday (Volt Computer)]

The circuit in question ( ETI 4600 page 20 ) couple be expanded without
too much problem.  The only question is that you would then be having to
deal with as many as ten ICs with their power and real-estate
requirements as well as the resistors and capacitors needed for the
clock generator.  A traditional noise source just requires a few
resistors, capacitors, a noisy transistor and one op-amp.

You don't double the circuit or cross-couple it, ( there is only one
input and it is just a fixed frequency clock signal ) you would have to
extend the length of the PSR-SG and add more taps to the feedback
circuit.  Easy enough to do and, if it was made long enough, it wouldn't
suffer from periodicity problems.

Still, the traditional noise sources are quick, easy, simple and
small...




> -----Original Message-----
> From:	jc at lynx.bc.ca [SMTP:jc at lynx.bc.ca]
> Sent:	Monday, September 15, 1997 11:12 PM
> To:	synth-diy at horus.sara.nl
> Subject:	pseudo-random shift generator
> 
> >A minor comment about the digital noise source - it is generated by a
> >pseudo-random shift generator. The problem is that the period repeats
> >every few seconds - this is 99% of the time not a problem but in
> cases
> >where you have long periods of nothing but noise, the repetative
> nature
> >is very obvoius.  It is possible to build longer length PSR-SG's but
> the
> >chip count gets a bit much compared to the traditional back-biased
> >analog transistor unit.
> >
> >David Halliday
> 
> I'm interested in these circuits a great deal myself, perhaps you
> could
> detail the circuit you're using (text description or gif file ?!) ...
> I was
> wondering and suspect that you could generate longer sequences by
> cross
> coupling two such circuits, as in a dual feedback arrangement ... by
> this I
> mean two complete random generator circuits except the outputs are fed
> back
> to the inputs from there or something (possibly further nored or
> exored
> using intermediate bits) ... doubling the circuit wouldn't be
> objectionable
> complexity-wise I wouldn't think.
> 
> jc
> 
> /////////////////////////////   vivaAnalog
> \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
> 
> /////////////////////   http://www.lynx.bc.ca/~jc/
> \\\\\\\\\\\\\\\\\\\\\\
> 
>      Vacuum tube modeling - Great guitar players - Traynor amp
> schematics
> 
> //////////////////////////////////
> \\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\
>