SV: Re: [sdiy] X8 squarewave multiplier.

[Michael Ruberto]

Another simple trick is to build a 20 step bar graph using LM3914s as per 
datasheet. Instead of LEDs on the outputs you tie each out high and then run 
them in sequential pairs to 8 two input XOR gates. Mix the outs of all XORs 
together for the multiplied output. Should retain 50% duty cycle quite well.

M. A. Ruberto


>From: karl dalen <dalenkarl at yahoo.se>
>To: Harry Bissell Jr <harrybissell at prodigy.net>,        Dave Kendall 
><davekendall at ntlworld.com>, synth-diy at dropmix.xs4all.nl
>Subject: SV: Re: [sdiy] X8 squarewave multiplier.
>Date: Wed, 18 Jan 2006 02:41:03 +0100 (CET)
>
>It is infact very simple! (:-) .
>
>Take a triangle then run trough a comparator sort of
>windowing, that will extract a square of 4x frequency,
>then dubble edge detecting that you get two edges for
>every one inserted, thats 8x. Think in terms of "phases"
>instead of zero crossings when dealing with the triangle!
>
>Or do a sawtoth multiplyer 8X is easy,just watch spurious edges!
>
>--- Harry Bissell Jr <harrybissell at prodigy.net> skrev:
>
> > The pulse rising/falling edge technique does not work beyond
> >   2X.  You NEED a square wave for it to work.
>
>Yes it does, a dubble edge detector on a square does 4X!
>Just time the Rc nets properly and it will cover 0 to 20khz
>easy!
>
>Asuming we are talking edges and not 50% squares here!
>
>Reg
>KD
>
> >   It could be made to work with one-shots, but only at a single
> >   frequency. Abandon this idea early....
> >
> >   H^) harry
> >
> > Dave Kendall <davekendall at ntlworld.com> wrote:
> > > It's a hell of a lot easier to start out with an 8X square wave and 
>divide
> > it
> > > down to 1X than to generate an 8X from a 1X signal.
> >
> > I'd worried about that, not ever having seen frequency multipliers 
>anywhere
> > in synth diy sites.....
> >
> > > Off the
> > > top of my head the way I'd do it is leading and falling edge triggers
> > > triggering one shots of the proper period. You'd have to have 3 levels 
>of
> > > those though.
> >
> > I guess there might be a risk with successive trigger delays?, and 
>changing
> > voltages probably would complicate matters even further. Hmmm...
> >
> >
> > Dave
> >
> >
> >
> >
>