ECE4893A: Electronics for Music Synthesis (Spring 2011)
Electronics for Music Synthesis
When: MWF 12:05-12:55 – Where: Van Leer C240
Office: Van Leer 276
Phone: 404-385-2548 (but it’s better to try reaching me through
e-mail)
E-mail:
lanterma@ece.gatech.edu (by far
the best way to reach me; please include “EMS” somewhere in the subject of class-related
e-mail so I can find it quickly)
Course website:
users.ece.gatech.edu/~lanterma/ems11
Prerequisites: ECE3040 and ECE3041 (with concurrency allowed,
i.e. you can
be taking ECE3041 this semester and be OK). Basically, I need some
familiarity with op amps, diodes, and transistors, and I need to make sure that
you have had some experience using a scope by
the time you will to use one in EMS. (Oscilloscopes are
introduced pretty early in ECE3041, which is
what makes the “concurrency” part OK)
Website for previous offering:
EMS (Spring 2010) –
will help give you a feel for what the class is like
<!– The photo: Chris Garyet testing his voltage-contolled “wah”
circuit
(from Spring 2006) –>
<!–
–>
(Note: I often abbreviate the class title as “EMS.” This usage of “EMS”
should not be confused with
“EMS” as in Electronic Music
Studios, the makers of many classic synthesizers such as the Synthis.)
Aaron’s SDIY Pages
- Synth
DIY Datasheet and Ap Note Collection - Music synthesis
patents collection, with some brief commentary - Current
modular synth manufacturers – look through these to get ideas for your
own new designs, ideas for packaging, etc.Homeworks
- Homework 1 (due Wednesday, Feb. 9,
at the start of class) - Homework 2 (due Friday, Feb. 25, at
the start of class) - Homework 3 (due Friday, March 18 at
the
start of class) - Homework 4 (due Friday,
April 22 at the
start of class)
Lectures
Dear readers from outside the class:
If you find these
lectures useful, please consider making a small donation (maybe $25 or
thereabouts, although any amount is appreciated)
to the Georgia Tech Foundation earmarked to go towards
my synthesizer research; the funds will go towards parts and equipment
for student projects.
Here
are instructions on how to donate.Since this is a lecture/lab class, I will only lecture for 2/3 of the class
periods, and that lecturing will be “front loaded,” i.e. I will lecture for
the first 2/3 of the class, and the last 1/3 of the class you will be just
working on your final projects, with me dropping by the lab to help out as
much as I can.If you look at the lectures from the
Spring 2010 version
of
the course, you will get a pretty good idea of what we will cover in the
future.- 1/10: Snowpocalypse, part 1
- 1/12: Snowpocalypse, part 2
- 1/14, Session 1: 40 Years of Music Synthesis in 40 Minutes
(video; sorry I accidentally
put in a memory card that was too small, so this cuts off part way through.
Also, I compressed the discussion a bit since the semester got to a late
start. Here are links to equivalent lectures from the 2010 version
of the class: part 1,
part 2.) - 1/17: No class (MLK Holiday)
- 1/19, Session 2: Demo of a Modular Synthesizer
(sorry, camera ate video file) - 1/21, Session 3: Circuit theory review, part 1
(video) - 1/24, Session 4: Circuit theory review, part 2
(video) - 1/26, Session 5: Circuit theory review, part 3
(video) - 1/28, Session 6: Voltage Controlled Amplifiers, part 1:
Operational Transconductance Amplifiers
(video) - 1/31, Session 7: Voltage Controlled Amplifiers, part 2:
Linear Current Sources
(video; alas, I had an
epic brain glitch and wrote an NPN resistor on the board, which makes
no sense in the current source circuit. It should have been a PNP,
and the error shoud have jumped out at me on several dozen occasions.) - 2/2, Session 8: Voltage Controlled Oscillators – Sawtooth Cores
(video) - 2/4, Session 9: Exponential Current Sources – Basics
(video) - 2/7, Session 10: Exponential Current Sources – Temperature Compensation
(video; I was not feeling well
that day, particularly during this lecture, so this is not me at my best.)
A tutorial on exponential convertors and temperature compensation, by
Rene Schmitz (this is the description my lecture was based on)- Tempo
Equations, by Ian Fritz (a much more detailed analysis)
- 2/9, Session 11: Voltage Controlled Oscillators – Triangle Cores
(video)- Buchla 259
Vintage Modular Oscillator youtube demo - Buchla
259 schematics,
on 4 pages:
1,
2,
3,
4. In class, we looked at the triangle core of the “Principal
Oscillator,” is in the upper right corner of page 2. - The
Buchla
258 is an older, simpler oscillator with a similar
triangle core: 258C,
258A,
Mark Verbos’
258 variant. (Mark
keeps an excellent Buchla
Tech blog.) - The new Buchla 261e
is quite similar to the Buchla 259. Here’s a
youtube demo of the 261e:
Simple 261e Melody - Buchla 259
vs. 261e Audio Comparisons on electro-music - Alessandro
Cortini,
who used to play synths for Nine Inch Nails, sold
his
Buchla 259 a while back. Alessandro has been making music on his
Buchla 200e system
under the name
blindoldfreak.
- Buchla 259
- 2/11, Session 12: Simple Waveshaping Circuits, part 1
(video)- Related to circuits shown in class:
- R. Williams,
Triangle to Sine Conversion with OTAs - M.H. Miller,
Triangle
to Sine Conversion (Nonlinear
Function Fitting), ECE414 Notes - R.G. Meyer, W.M.C. Sansen, S. Lui, S. Peeters,
The Differential Pair as a Triangle-Sine Wave Converter,
IEEE J. of
Solid-State Circuits, June 1976, pp. 418-420. - G. Klein,
Accurate Triangle-Sine Converter,
IEEE International Solid-State Circuits Conference, Digest of Technical
Papers, Volume X, Feb. 1967, pp. 120-121.
- R. Williams,
- Other ideas:
- H. Hassan,
FET
Differential Amplifier as a Tri-Wave to Sine Converter,
Proc. 36th Southeastern Symposium on System Theory, 2004, pp. 427-430 - Z. Tang, O. Ishizuka, H. Matsumoto,
MOS Triangle-to-Sine Wave Convertor Based on
Subthreshold Operation, Electronics Letters,
Vo. 26, No. 23, Nov. 8, 1990, pp. 1983-1985.
- H. Hassan,
- Related to circuits shown in class:
- 2/14, Session 13: Simple Waveshaping Circuits, part 2
(video) - 2/16, Session 14: Complex Waveshaping Circuits, part 1
(sorry no video from this year; here’s
an anologous video from 2010)- Demo:
Adaptation
of the timbre circuit from the Buchla Music Easel (fun part starts at
around 2:20)
- Demo:
- 2/18, Session 15: Complex Waveshaping Circuits, part 2
(sorry no video from this year; here’s
an anologous video from 2010 – alas,
I was sick when I did this lecture and sound quite hoarse) - 2/21, Session 16: Single-pole OTA-C filters
(video)- Oberheim/Rossum patent,
Circuit for Dynamic Control of Phase Shift - ARP patent,
Frequency
Sensitive Circuit Employing Variable Transconductance
Circuit
- Oberheim/Rossum patent,
- 2/23, Session 17: 4-pole filters with feedback
(video)- Note that most of the linear analysis on the Moog ladder filter, as
presented in papers and webpages listed under Session 20B, applies to
4-pole-with-feedback filters built with OTA-C sections. (An
analysis of the nonlinearities would show differences.)
N-Pole Filter Circuit Having Cascaded Filter Sections – careful,
the resonance feedback path drawn on the first page is in error! It should
be going to the negative terminal of the first OTA on the left.
- Note that most of the linear analysis on the Moog ladder filter, as
- 2/25, Session 18: Examples of 4-pole OTA-C filters with feedback, part 1
(video) - 3/9, Session 19: Examples of 4-pole OTA-C filters with feedback, part 2
(sorry no video from this year, since I forgot the camera; here’s
an anologous video from 2010) - 3/11, Session 20A: Demo of the MOTM 440 Voltage Controlled Filter (an
SSM2040-style circuit, made with discrete transistors)
(video) - 3/11, Session 20B: Transistor & Diode Ladder Filters, part 1
(video; alas, the way I drew the
audio input coupling is misleading. I may be misleading in using the word
“misleading,” since a more accurate word might be “incorrect.” Anyway, there’s
a resistor chain that sets bias points on the the top (rightmost) four of the five
transistors, and
the audio input is coupled to the first transistor through a capacitor. The way
I drew the ladder makes it look like the audio input is part of that resistor chain,
which isn’t correct; I should have used a little half-circle “jump” to jump
one wire over the other to show that
the base of the first transistor
isn’t part of the resistor chain that sets the DC voltage
bases of the top four transistors.)- Moog patent,
Electronic High-Pass and Low-Pass Filters Employing the Bass to Emitter Diode Resistance of Bipolar Transistors - T. Stilson and J. O. Smith,
Analyzing
the Moog VCF with
Considerations for Digital Implementation,
Proceedings of the 1996
International Computer Music Conference, pp. 398-401. - A. Huovilainen,
Non-Linear
Digital Implementation of the Moog Ladder Filter, Proc. of the
7th Int. Conf. on Digital Audio Effects (DAFx’04), Naples, Italy, Oct. 5-8,
2004. - T.E. Stinchcombe,
Analysis of the Moog Transistor Ladder and Derivative Filters,
Oct. 25, 2008 - M. Civolani and F. Fontana,
A Nonlinear Digital Model of the EMS VCS3 Voltage-Controlled Filter,
Proc. of the 11th Int. Conf. on Digital Audio Effects (DAFx’08), Espoo,
Finland, Sept. 1-4, 2008. - T.E. Stinchcombe’s page on
Diode
Ladder Filters - T.E. Stinchcombe’s
Filter pole animations
- Moog patent,
- 3/14, Session 21: Transistor & Diode Ladder Filters, part 2
(sorry no video from this year, since I forgot the camera; here’s
an anologous video from 2010) - 3/16, Session 22:
Second-order filter properties
(video)- T.E. Stinchcombe’s
Filter
pole
animations (see the bottom of the page for second-order examples)
- T.E. Stinchcombe’s
- 3/18, Session 23: State variable filters (theory)
(video)- MIT 2.161 Signal Processing notes,
Op-Amp Implementation of Analog Filters (see Section 2) - R. Johnson,
Programmable
State-Variable Filter Design For a
Feedback Systems Web-Based Laboratory - Daycounter, Inc. Engineering Services,
State
Variable Filter Design Equations
- MIT 2.161 Signal Processing notes,
- 3/28, Session 24: State variable filters (examples)
(video) - 3/30, Session 25: Sallen-Key filters (theory)
(video) - 4/1, Session 26: Sallen-Key filters (examples)
(video)- Demo:
Adaptation of the lowpass gate
circuit from the Buchla Music Easel - Demo: Thomas White’s
Buchla
Lopass Gate 292 Clone - T.E. Stinchcombe’s page on
The
Korg35 Chip, the MS-10 & MS-20 Filters, Clones and Links - T.E. Stinchcombe,
A Study of the Korg MS10 & MS20 Filters, August 30, 2006
<!–
- Demo:
- 4/18: Session 25: Survey of digital music synthesis; wrap-up
(video)
–>
References
We will draw material from numerous sources: book, articles, patents,
and particularly schematics and descriptions posted on websites. Think of
google as the main class text. Here’s some
good ones:- Hal Chamberlin, Musical Applications of Microprocessors, 2nd
Edition, Hayden, 1982; if you
get just one book, this is THE book to get. Although it has
“microprocessors” in the title, it has a superb section on analog circuits.
NOS (New Old Stock – meaning old, but unused) copies are available for
purchase from Jeff Dec ($50 + shipping); e-mail
jdec@mindspring.com - Barry Klein, Electronic Music Circuits, SAM, 1982. Long out
of print, but photocopies can be purchased directly from
barry.l.klein@wdc.com - V. Valimaki and A. Huovilainen,
Oscillator and Filter Algorithms for Virtual
Analog Synthesis, Computer Music Journal, Vol. 30, No. 2, 2006,
pp. 19-31. - T. Stilson,
Efficiently-Variable Non-Oversampled Algorithms in Virtual-Analog
Music Synthesis, PhD Thesis, Standford University, June 2006.
Grading
Final letter grades
will be based on a series of written homeworks,
a few simple, enjoyable lab assignments,
two in-class quizzes, and
the quality of a final project in
which you will design and build a module for a modular synthesizer. The
final project will permit (and encourage) you to make extensive use of
various existing schematics you might find on the web, in textbooks, or
elsewhere. Details about the final project will be posted at a later date.The homeworks
are intended to be instructive and enlightening, and in particular
get you looking at schematics of real synthesizers that have been
in production, and not “textbook” problems. I try to avoid giving
anything resembling “busywork.”The labs will be brief (less than an hour), fun, and not
have lengthy 3041/3042 style reporting requirements. We will not do many
of these; probably just two, or three at the most. They will be intended
to help you get your “feet wet,” so you will have some more hand-on hardware
experience before jumping into the final project. (Previous versions of
the class just had the final project without any earlier labs. One of the
most common suggestions I received from students was to put in some small
lab components earlier in the semester
so students would feel more confident going into the final
project.)The first quiz
will be given about 1/3
through the class, and the second will be given about 2/3 of the way through
the class. Both will be closed book. The first quiz
will focus on basic
facts about circuits and electronic facts that a designer needs to have
“at their fingertips,” without having to stop and look up, in order facilitate
a smooth creative workflow. I will provide extremely detailed
information about
what I will ask on that quiz. The second quiz will probe what kind of
intuition you have developed concerning the class material; the
questions will be more qualitative in nature (for instance: if the value
of resistor X is increased, will the frequency of this oscillator go up
or down?), in the sense that they will not require tedious calculations with
precise numeric results. Each quiz will be weighted like a homework
assignment.
(I used to do
only one quiz, but, curiously enough, students told me
I should give more quizzes!) There will be no usual written final exam given
during final exam week.I consider the final project
to be the most important thing in the class;
hence, your course grade will max out at whatever your project grade
is,
e.g., if you do B work on the homeworks, etc., but turn in an A project,
you might get an A for the class, or you might get a B;
but if you do A work on everything else but turn in B level project,
your grade won’t be an A.This is a small class, and I will work with you very closely in helping you
with your final project. By the end of the semester, I will have a pretty
accurate feel for what concepts you understand and what concepts you
don’t. - Homework 1 (due Wednesday, Feb. 9,