From Oscillators to
Mixers to Sound
In order to better understand the
Soundgin — or any sound-generating
device — it’s handy to know a little
about sound in general. Sound is
waves that travel through some type of
medium, like air or water. The size of
the waves determines its amplitude, or
volume, and how close the waves are
from one another determines its
frequency. The farther apart the waves,
the lower the frequency, and vice versa.
When humans speak, a curious little doodad in the throat called a larynx
vibrates, and as air from the lungs
passes by the larynx, sound waves are
produced. These waves travel through
the air, and anyone nearby hears us
when the waves enter their ears. A
similar process is involved in making
sound from a guitar string (the string
vibrates, disturbing the air around it) or
a saxophone (a wooden reed vibrates
as air is blown over it).
The Soundgin doesn’t use a larynx,
string, or reed. But it does use the same
general principles of making waves.
Sound production begins in one of its six
oscillators. These oscillators are akin to a
larynx, except the Soundgin has six of
them. As such, it is said to be a six voice
sound synthesizer. The oscillators are
grouped in two sets of three; each set is
routed to its own mixer, so that the
signals from the oscillators can further
be controlled. More about this in a bit.
Finally, the outputs of the two
mixers are combined into a final audio
output. This output is connected to an
amplifier to increase the level of the
signal produced by the Soundgin chip.
Most any audio amplifier will do.
The Soundgin prototyping board uses
the LM386 sound amplifier, which is
inexpensive, easy to use, and requires
few external parts.
The proto board, shown in Figure
1, can be powered by a nine volt battery, and has connections for a DB9 serial cable, power, and speaker terminals.
From the amplifier, the signal is sent
to a speaker. Until now, the “sound”
from the Soundgin has just been electrical impulses. The speaker turns those
impulses into physical vibrations that
move air. Our ears pick up that air movement and perceive it as sound.
A Closer Look at
The heart of any sound-making
module is its oscillators, and the
Soundgin is no exception. As noted
above, Soundgin consists of six independent oscillators. Each oscillator is a
separate sound synthesizer, or voice.
Any voice can be used by itself, though
it is common to use them in combination to produce more elaborate sounds.
For example, the output of one oscillator can feed into another to produce a
slow rising and falling siren effect.
Each voice consists of a frequency
generator, an amplitude modulator, an
envelope generator, and a ramping/
The frequency generator controls
the pitch of the sound. The lower the
frequency, the lower the pitch. If you
picture a piano keyboard, the keys to
the left have a decreasing frequency,
so they have a lower pitch. The keys to
the right have an increasing frequency,
so they have a higher pitch.
Not all sounds are at the same
volume level, and any sound-making
device needs a way to control the overall level of any oscillation. Each of
Soundgin’s oscillators contains an
amplitude (volume) control that can be
varied from 0% — which is no signal
output at all — to 100%.
Sound is often characterized by
the shape of the waves that produce it,
not just the frequency of those waves.
The shape of the wave affects the harmonics of the sound — harmonics can
be said to be the interaction of the
wave with itself and other sound waves
being produced at the same time.
Soundgin produces eight types of wave
shapes, including sine wave, square
wave, and saw tooth.
The envelope generator is a kind of
dynamic volume control that changes
over time. The envelope is characterized by four distinct phases: Attack,
decay, sustain, and release — often
referred to simply as ADSR. These phases are graphically shown as a series of
ramps. Attack indicates how fast the
oscillation comes to full volume. Decay
is how fast the oscillation ceases.
Sustain is a constant volume of the
sound after the decay until the note is
released; and release is how quickly the
sound fades out. These parameters —
along with the wave shape — define the
timbre of the sound. It’s what helps
makes the oscillation sound like
something familiar to us — a piano or a
violin, for instance. Each of these instruments has a peculiar ADSR envelope.
Soundgin provides further sophistication in the sounds it produces by
using separate ramp and target controls for each oscillator. These controls
have a similar function to the ADSR
envelope, but are more robust — and
trickier to use. When the oscillator’s
amplitude target is set, the amplitude
of the oscillator will move towards the
target at the rate determined by a
special transition value, until the oscillator’s amplitude has reached the target.
One use of this feature is to make
various rising and falling sound effects,
like phaser guns or alien bio sounds.
Mixing and Matching
You can create a multitude of
sounds simply by setting Soundgin’s six
oscillators to some frequency, adjusting
the amplitude and wave shape, and
having them run completely independently. Whether such a mix would sound
pleasant is another matter! And it misses a lot of potential of Soundgin as a
complex sound and effects generator.
As noted above, Soundgin allows
you to control one oscillator with another. In music synthesis, this patching technique is used to produce an infinitely rich
flavor of sounds. On the average analog
synthesizer, any oscillator can be patched
to any other oscillator. The Soundgin chip
does not provide for unlimited inputs and
output patching, but rather it predefines
the patching between specific oscillators.
This technique greatly simplifies the
programming you’ll need, yet you are still
able to produce a wide variety of sound
and music effects.
Recall the Soundgin has six oscillators, and that these oscillators are separated into two sets of three each. The
oscillators for the first set are labeled A1
through A3; the oscillators for the second set are labeled B1 through B3. The
functionality for both sets are identical.
Oscillator 2 (A2 or B2) can control the
amplitude modulation of Oscillator 1 (A1
or B1). Similarly, Oscillator 3 (A3 or B3)
can control the frequency of Oscillator 1.
SERVO 01.2007 35