Monday, June 28, 2010

Sinewave Fun...

Well slowly more and more of the 921B VCO comes to life. I am having fun with the Sinewave shaper today and getting a reasonable shape coming out but there are still quite a lot of harmonics... It seems like the amplitiude of the Triangle wave is just too low to get a good Sinewave. The non-linearities of a transistor curve are exploited to get the shape and there is an input amplitude adjustment but I have it hard at maximum to get the best output. The other two adjustments in the circuit balance the two sides of the sinewave and both mimize the harmonics within their travel... I am also getting a fair amount of high frequency "hash" on the waveform from a lack of decoupling capacitors! Will try and clean that up tomorrow then on to the really intersting part -- the Sync section... Not like any sync circuit in any other music synth that I have seen before! As I progress through this I am really tempted to make a clone of this which is more 21st century, use better matched pairs and slightly less obscure parts and newer multipler etc. I suppose then it isn't really a clone... Other intersting things are the output levels. Most modern VCO's are adjusted for 10V peak-to-peak on everything where as this is a lot lower and adjusted for more of an equal audible loudness. By that I mean that the sinewave is 1.8V p-p, the triangle is 1.4Vand the pulse 2.2V -- not sure if anybody else does that? Also seems odd that the pulse is so much more...
Posted by Picasa

Monday, June 21, 2010

It Lives!


Really thought that I must get back onto the Moog 921B VCO. I was having a LOT of trouble getting a dual JFET... any dual JFET not necessarily the right one. Getting a through hole one from Mouser came in at a staggering $118 US! I then found a surface mount part that is similar in nature and managed to completely stuff up the footprint... I then did a fair bit of simulation with LTSpice and discovered that the offending section of the circuit is little more than a very high impedance buffer. Since this thing was designed long before FET input op-amps this was effectively what it was trying to achieve. I ended up removing a bunch of diodes and resistors and swapped the OTA for a CA3140 and added two jumpers --Hazaar! the result a sawtooth! I will tidy up the layout a bit so that it can be done either which way but I seriously doubt that it is going to make much sonic difference....

Tuesday, June 15, 2010

ARP Odessey VCO

Not really sure why I need another VCO... but anyway we have one now! Well actually two! The FET's were all from what was hanging around but still seem to work. I didn't get too caught up in matching the transistors as I probably need a VCO for other stuff...