David G. Dixon
dixon at interchange.ubc.ca
Wed Feb 18 08:48:06 CET 2009
Synthesizers, eh? ...
Well, I've just finished testing my very first circuit design, a
triangle-core VCO with log and linear FM, pulse with PWM, saw and sine. I'm
very happy with its performance, and I did it all with just three chips: one
LM13700 and two TL074s. The triangle is perfect (thanks, Thomas Henry!),
the sine has as little as 1.4% THD (according to simulations; it sounds
pretty nice, too! Thanks, Ray Wilson!), there is only a barely perceptible
glitch (on the scope) in the middle of the ramp (which may disappear
entirely when I move the circuit off of the breadboard), and the pulse is
pretty square! Tomorrow I'll lay out the pc board for home etching. I plan
to build at least two of them.
I thought of one little innovation today which is very simple: I put in a
switch to select between the triangle and the ramp for the input to the
pulse wave comparator. When the resulting pulses are mixed with the sines
or triangles from the same VCO, they morph them in different ways. I'm not
sure if the results really sound very different, but it is a very simple
thing to do, so I'll probably keep it. Has anyone else tried this?
I've also completed the design and simulation of a dual-OTA-based
four-quadrant multiplier which is actually pretty linear (unlike the useless
single-OTA design in the LM13700 datasheet which is impossible to bias
properly). The carrier signal is fed to the opposite inputs of both halves
of an LM13700, and one complete TL074 is used to buffer and invert the
modulating signal and sum the currents from the OTAs. With the twist of a
pot knob, it can be an inverting or non-inverting two-quadrant multiplier
(VCA) at the ends, or a four-quadrant multiplier (ring-mod) in the middle,
all with nearly unity gain. I've laid it out on a breadboard, but haven't
tested it yet, because I don't have an LFO to use with it. I'll throw
something simple together tomorrow and give it a whirl with my VCO. This
was all inspired by trying to figure out how to put a Grenader gizmotron
(i.e., -1 to +1 adjustable gain opamp) under voltage control.
One detail which has come out of this is the importance of putting buffers
after attenuators in order to maintain a linear gain response with respect
to the attenuating pot setting when feeding a modulating signal into an
inverter or other device with finite input impedance. It is probably worth
a few extra opamps in any design to ensure this linear response.
Finally, I've also been educating myself on the finer points of filter
design (transfer functions, Laplace transforms, zeros and poles, etc) in
preparation for designing my own VCF later this month.
If it sounds as if I'm reinventing the wheel, well, I guess I am. But, it's
an awfully good way to learn electronics! I highly recommend it!
David G. Dixon
Department of Materials Engineering
University of British Columbia
309-6350 Stores Road
Vancouver, B.C. V6T 1Z4
"PERFECTA FINGAMUS SERVIAT NATURA"
The information in this email and in any attachments is confidential and
intended solely for the attention and use of the named addressee(s). It
must not be disclosed to any person without the writer's authority. If you
are not the intended recipient, or a person responsible for delivering it to
the intended recipient, you are not authorized to and must not disclose,
copy, distribute, or retain this message or any part of it.
> -----Original Message-----
> From: synth-diy-bounces at dropmix.xs4all.nl [mailto:synth-diy-
> bounces at dropmix.xs4all.nl] On Behalf Of Tom Wiltshire
> Sent: Tuesday, February 17, 2009 10:19 AM
> To: synth DIY
> Subject: [sdiy] Synthesizers...
> ...ones we built ourselves. Anyone fancy a chat about that?
> Surely someone's done something interesting.
> Myself, I'm still playing with the DCO design that I posted a problem
> with the other day. Now it's going properly, I've been sorting out
> the amplitude calibration, and looking at the way that the Rhodes
> Chroma uses a ramp plus a PWM pulse wave to mimic the effect of two
> detuned sawteeth. Apparently these two are equivalent, but I'm yet to
> prove it to myself to my satisfaction.
> Synth-diy mailing list
> Synth-diy at synth-diy.org
More information about the Synth-diy