Full wave rectifier as Saw-to-Tri converter

[Bissell, Harry]

Here is a saw/tri converter. It requires only two well matched resistors to
give "perfect" matching (but it will have a glitch unless your sawtooth has
an infinitly short fall time (oh yes!!!). Maybe the troubles you were having
were due to the ramp not being centered around exactly 0 volts, that will
also cause the peaks not to match. The Absolute value circuit I show is
bulletproof and way superior to the schematic of the "full wave" rectifier
you posted. BUT, it does not allow gains other than unity, and does not
allow filtering by putting a cap in the negative feedback of the second
op-amp (but, it doesn't have the crappy gain matching of the fast full-wave.
I always have to trim the gains.)  Try this circuit, you can eliminate the
DC offset stage if you don't need it, and you can do the output however you
like, so at a minimum, its  just a dual op-amp. The circuit was originally
from an EDN or Electronics Design Ideas arount 1983 or so, if I find the
original I'll credit the designer.  ;-)  Harry Bissell <<Triwave2.gif>> 

> -----Original Message-----
> From:	Roy Tate [SMTP:roytate at ionet.net]
> Sent:	Thursday, February 11, 1999 1:08 AM
> To:	synth-diy
> Subject:	Full wave rectifier as Saw-to-Tri converter
> 
> I recently tried out a few of the saw-to-tri converter schematics
> I've collected, and in most cases, I wasn't happy with the results.
> The switching point would either have a spike to +15 or -15 V or
> the inverted section would not meet the non-inverted section of the 
> wave.
> 
> After looking at Walter Jung's OpAmp Cookbook, page 249
> (figure 5-17 - see attachment), and noticing that Rick Jansen's
> wave shaper uses this circuit, I tried it out.  Given a +/-5V saw
> input, it generates a positive 5V triangle with a 1 volt offset
> between the positive and negative halves of the peak.  The 
> resulting triangle is ...
> 
>       |\        |\    
>      /  \      /  \   
>     /    \    /    \  
> \  /      \  /      \  /
>  \/        \/        \/
> 
> This sounded okay, but I would rather have a "perfect" triangle,
> as it will make sine wave shaping easier, and will not sound as
> bright as a dirty triangle.  So, I experimented with the circuit
> and finally came across a tweak that makes the output triangle
> nearly perfect.  I simply replaced the 10K resistor (R2) with a 20K
> trimmer and dialed it up to 11.3K Ohms (R2 goes between the 
> inverting input and output of the second op-amp.)  Any resistor 
> between 11.2K and 11.5K would probably give satisfactory results,
> but a trimmmer allowed me to line up both peaks.
> 
> Obviously, to use this in a synth, I would add an output buffer 
> with a gain of 2 and a negative 5V offset.  So 3 op-amp sections 
> are required to make a usable synth module.  By the way, I used
> the TL072A (the low offset version).
> 
> Regards,
> 
> Roy Tate
> roytate at ionet.net
> http://www.ionet.net/~roytate << File: SAW-TRI.GIF >> 
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Triwave2.gif
Type: image/gif
Size: 26787 bytes
Desc: not available
URL: <http://synth-diy.org/pipermail/synth-diy/attachments/19990211/a0ae0617/attachment.gif>