Fast analog sw. as VC resistor (0/1)

[brad sanders]

Hello again, everyone.

I've been working to *complete* the VC ADSR generator idea I littered
this list with last week. I still don't have a complete breadboard, as
I've changed the MUX to the MUCH better DG409, and have to wait for
samples to arrive from Maxim before I can wire it all up.

However, I wanted to share the design (this far) with the members of
this list, and get some feedback on the complexity. I've never seen a
complete ADSR gen based on the SSM chips, so I have nothing to compare
with my design. 

Is this circuit substantially more complex than a COMPLETE circuit
based on the SSM chip? Admittedly, this design COULD be made simpler
by removal of the output filter and input protection circuitry - but I
personally HATE this idea being, as I am, of the "bulletproof" academy
- in fact, this is one reason I specifically chose the Maxim DG
series: they have some SERIOUS input protection, obviating the need
for more external protection. 

I've included a shrunken schematic. It's ugly, but it's small. Soon as
I get the whole thing complete I'll add a revamped "tech" section to
my own web space and include full details. I've also been working on a
VC 4 channel mixer/dual VCA/dual four quadrant multiplier (four
inputs, two outputs; make whatever you want of it) which uses just
three chips: a quad opamp, a 339, and yet another Maxim DG series
switch. It's a bit more complex than a basic LM13600 based dual VCA,
but it's MUCH more flexible. Again, I'll put complete details on this
at my webspace - probably early next month.

And, for those who asked: here's the parts list for the circuit as it
exists in this rev (0.9). The passives are all incredibly cheap;
C_adsr should be of good quality and LOW ESR, and C_cv_filt is
essentially a S/H cap and so should be a halfway decent film - but the
rest of the circuitry can be just about anything, (Note, 'tho, that in
ANY HF switching circuitry like this you should use LOW ESR power
supply bypass caps; I use Sanyo OSCONS, 'tho Panasonic HFQs might be a
bit cheaper.)

Qty  References                                             Value  
-----------------------------------------------------------------
 3  R_opt,R_adsr,R_pullup4                                    1K  
 1  R_hyst                                                    1M  
 4  Rp1,Rp2,Rp3,Rp4                                          100  
 8  Rg,R_div,R_lp1,R_lp2,Rg_div1,Rg_div2,R_pullup1,          10K  
    R_pullup3
 1  R_bias3                                                  47K  
 4  Rpd,R_bias1,R_bias2,R_pullup2                           100K  
 6  Cp1,Cp2,Cp3,C_lp1,C_lp2,C_opt                          100pf  
 2  C1,C2                                                  4.7uf  
 1  U1                                                     DG409  
 1  U2                                                     LM339  
 1  C_cv_filt                                             0.01uF  
 1  C_adsr                                                0.33uF  
 1  U3                                                   LMC6082  

Many thanks go to Mark Smart and Christopher List - and to Daniel,
whose "challenge" spurned me to start this whole mess. Looks like I
might be able to make my own multiphonic VC synth after all - WITHOUT
those rare and 'spensive chips!

Oh: does anyone HAVE a fatman? I'm wondering how the LM13600 based VCF
performs. It's a pretty simple circuit; should be inexpensive enough
to duplicate over four or eight modules.


> http://radioactive.home.ml.org <-------------------------------