Paia 4700/S User manual

A GUIDE FOR
USING
THE
PAIA
4700/S
SYNTHESIZER
SYSTEM
i
mae
<t
ps9
EPH
cal
:
TITUS
<i
rir i :
res |
e
"|
©
1976
PAIA
Electronics,
Inc,
1020
W.
Wilshire
Blvd.
»
Okla,
City,
OK
73116

There has never been a musical instrument that was concieved overnight and
released to the world in an immutable state the next day, each instrument has under-
gone change and refinement to bring it to its present condition, The same is true of
electronic music but the newness of this field coupled with the technology explosion
has caused its development to be compressed in time. Whereas the piano has taken
centuries to evolve, electronic musical instruments first appeared only a little over
four decades ago,
Most of these first instruments were little more than exercises in technology
but some were designed to overcome shortcomings in existing instruments, For
instance, the piano keyboard is one of the most powerful musical operating systems
available but it has one outstanding drawback in that while it provides the musician
easy access to all twelve notes of the equally tempered musical scale it prohibits
him from using the infinity of musical pitches between those twelve notes. By its
very nature it eliminates the possibility of any easy glide from one musical pitch to
another ( glissando ).
One electronic instrument designed to overcome this weakness was the Martinot,
The Martinot is similar to modern organs in that a standard keyboard is used to con-
trol an electronic oscillator built around a frequency determining capacitance/induct-
ance tank circuit, The inductor is tapped at points that produce frequencies correspond-
ing to the chromatic scale; a rather straightforward, if somewhat simplistic, approach
to electronicly generating a musical scale. In addition to the keyboard there is a
finger ring attached to a slider that controls another oscillator. When properly adjusted
this second oscillator produces the pitch corresponding to the keyboard key adjacent to
the position of the ring, The combination of keyboard and slider allows the musician
to glissando from one note to another or add vibrato with a simple move of the hand
without sacrificing the operating ease of the keyboard.
The ondioline was a contemporary of the Martinot but is significant because it
was the first electronic musical instrument to use something other than a sine wave as
its basic tone, In the Ondioline a relaxation oscillator controlled by the keyboard
produced a sawtooth wave which in turn activated several frequency dividers. The
output of the oscillator and frequency dividers were combined using much the same
techniques used in some modern organs so that the instrument was capable of
generating a great variety of sounds. Observers report that a skilled operator could
come close to making the Ondioline talk,
While the Martinot and Ondioline were both designed in France, America's
contribution to freeing the musician from the restrictions of the keyboard was probably
the most outstanding ~ not to mention bizarre. A Theremin has no visible means of
control at all and is played simply by moving the hand in relation to two metal plates
or rods, Inside the instrument are two high frequency oscillators, one shielded from
any external influences and the second arranged so that the plate or rod forms part
of the frequency determining inductance/capacitance tank circuit. The outputs of
these two oscillators are combined in such a way that an audible tone that is the
difference between the two frequencies is produced, As the performer's hand is
brought closer to the sensing antenna the difference in the two frequencies increases
and so does the pitch of the tone, A second circuit allows the performer's other hand
to determine the volume of the sound produced, Since there are no frets or keys to

to
provide
visual
or
tactile
clues
to
the
pitch
a
Theremin
will
produce,
it
is
a
very
difficult
instrument
to
play
-
but
loads
of
fun,
SYNTHESIZERS
The
first
equipment
that
would
come
close
to
meeting
our
current
definition
of
a
synthesizer
was
built
by
Dr.
Harry
Olson
during
the
early
1940's*
Produced
under
the
auspices
of
the
RCA
Labs,
the
RCA
Mark
I
and
Mark
II
Synthesizers
were
some-
thing
to
behold,
The
Mark
I
has
been
disassembled
for
some
time
now
but
the
Mark
IT
still
exists
and
is
currently
being
leased
to
Columbia-Princeton
Electronic
Music
Center,
it
neasures
17
feet
long
by
7
feet
high
and
is
valued
at
anywhere
between
$250,
000 and
one
and
a
half
million dollars
depending
on
who
you're
talking
to.
The
average
performer
might
be
a
little
disappointed
in
the
Mark
0
today
because
even
if
there
were
some
way
to
transport
it
to
a
gig,
he
would
find
when
he
got
there
that
he
couldn't
actually
perform
a
number,
The
Mark
II
was
simply
not
capable
of
real-time
operation,
each
characteristic
of
the
sound
the
instrument
was
to
produce
was
laboriously
calculated
and
plotted
ahead
of
time
and
the
result
punched
into
a
roll
of
paper
tape,
When
it
came
time
for
the
Mark
II
to
do
its
thing
the
tape
was
fed
in
-
like
a
very
large,
very
expensive
player
piano
-
and
the
results
recorded
on
a
multi-
track
disc,
(
early
40's,
remember,
recording
tape
wasn't
so
hot
in
those
days,
)
When
all
the
parts
of
a
number
had
been
recorded
-on
the
separate
tracks
of
the
disc
they
were
re-recorded
on
another
disc
from
which
a
master
was
made.
You
might
think
that
about
the
only
thing
that
the
Mark
I
and
Mark
II
did
that
was
of
any
consequence
was
add
the
word
Synthesizer
to
our
vocabulary
but
that's
not
the
case
at
all,
They
were
significant
first
of
all
becuase
they
were
the
first
to
put
it
all
together
as
far
as
electronic
music
production
was
concerned,
All
the
oscillators,
amplifiers
and
filters
needed
in
one
place
at
one
time
and
best
of
all
some
means
-
no
matter
how
cumbersome
-
of
controlling
them
ail,
Secondly,
they
were
the
first
instruments
to
utilize
white noise
sources
as
part
of
an
electro-musical
instrument,
White
noise
will
be
covered
in
detail
later,
for
now
it
should
suffice
to
say
that
without
it
sounds
like
snare
drums
and
cymbals,
to
mention
only
two,
are
impossible,
Don't get the impression that electronic music cannot be produced without a
synthesizer,
that's
not
true,
Imagine
that
you
are
in
a
laboratory
with
all
sorts
of
electronic
equipment
such
as
oscillators,
filters,
amplifiers,
modulators,
tape
re-
corders,
etc,
You
turn
on
one
of
the
tape
recorders
and
set
the
oscillator
for
the
pitch
you
want,
twiddle
the
knobs
of
the
amplifier
to
shape
the
loudness
contour
and
play
with
the
filter
knobs
to
adjust
timbre,
It
only
takes about
six
hands
and
a
couple
of
minutes
but
when
you're
through
you've
got
a
whole
note
recorded
on
the
tape.
Repeat
the
process
often
enough
and
you've
got
a
whole
string
of
notes.
Of
course,
the
tempo
is
not
right
and
the
notes
may
not
be
in
the
right
sequence
but
you
can
fix
that
by
snipping
the
tape
apart
and
editing
out
all
the
junk
before
splicing
it
back
together
again
to
produce
the
desired
melody,
Now
you
go
back
and
do
the
same
thing
for
bass,
rhythm
and
all
the
other
parts.
About
the
only
thing
you
can
say
for
this
technique
is
that
it
should
certainly
give
you
a
feeling of accomplishment. Considering the complexity of the process even such
monstrosities
as
dogs
barking
out
the
tune
of
"Away
in
a
Manger"
can
be
forgiven
~-
all that knob twiddling has to do something to a person's mind,

In the early 1960's Dr, R. A. Moog ( recognize the name?) began developing
and producing a line of electronic music synthesis equipment that revolutionized
the field, The feature that made the Moog equipment such a quantum jump in ease
of operation sounds almost ridiculously simple, but its implications are so far
reaching that it must be stressed; THE KEY PARAMETERS OF THE PROCESSING
ELEMENTS ARE A FUNCTION OF THE SUM OF SEVERAL CONTROL VOLTAGES
RATHER THAN THE POSITION OF A KNOB,
As
an
example
of
the
operating
ease
of
voltage
control
let's
see
what
it
does
for
a relatively
simple
processing
element,
an
amplifier.
As
we
shall
see
a
little
later,
one
of
the
things
that
contributes
most
to
the
way
an instrument
sounds
is
the
manner
in
which
its
sound
builds
up
and
dies
away.
When
using
the
classical
tape
splicing
technique
these
characteristics
have
to
be
duplicated
manually
for
every
note by
turning
the
volume
control of
the
amplifier,
Even
though the
Mark
II
allowed
for
automatic
control of
the
amplifiers,
information
still
had to
be
punched
into
its
programming
tape
for
each
individual
note.
With
voltage
control
the
job
of
setting
the
correct
time
varying
amplifier
gain
can
be turned
over
to
an
automatic
electronic
function
generator
circuit
that
produces
a
repeatable,
pre-set
voltage
waveform
each
time
a key is
pressed.
This
voltage
is
then
used
to control
the
amplifier,
The
musician
sets the
function
generator
to repro-
duce
the
characteristics
of
some
real or
imagined
instrument
and
the
electronics
will
produce
that
characteristic
for
each note
he
plays.
If
he
desires
a totally
different
sound
it's simply
a
matter
of
re-setting
a
couple
of
knobs.
Summing
the
control
voltages
allows
the
performer
to produce
more
than
one effect
from
a
single
process—
ing
module.
If,
in
the
above
example,
the
operator
decides
to
add
a
low frequency
amplitude
modulation
(
tremolo
)
to the
sound,
he
needs
only
to sum
a
second
volt-
age
that
is
changing
at
the
rate of
the
desired
tremolo
into
one of
the
remaining
amplifier
control
inputs,
As the
control
voltage
varies
up
and
down
so
does
the
gain
of
the amplifier
and
the
volume
of
the
sound.
THE SOUND OF MUSIC
Anyone can make weird noises on a synthesizer simply by randomly making
connections and pushing buttons, It's even fun for the first hour or so, until you
begin to think of specific sounds you want to make and can't, If we're going to learn
to use a synthesizer rather than just play with it it's important that we understand
what sound is and what makes one sound different from another.
If your knowledge makes the following discussion seem trite, read on anyway.
We have to start somewhere and if nothing else you can probably find something to
disagree with.
Sound travels as waves, waves of pressure in the air. A vibrating string
displaces the air around it and the air molecules that the string moves in turn
bump into and move other molecules, All the things that these sound waves can
bump into and be reflected off of and the effect that this has on the original wave
are beyond the scope of our discussion, The only thing relevant to the subject at
hand is that if a man is present the pressure of the waves will finally cause a
deflection of his eardrum which in turn will vibrate three small bones inside his
ear which will in turn cause a disturbance in a fluid medium which in turn excites
the auditory nerves which in turn causes the man to say '' Hey, listen to that aed
Whether he fills in the blank with "noise or "music" is personal preference,

The
thing
that
vibrates
to
produce
the
sound
doesn't
have
to
be
a
string.
It
can
be
a
synthetic
or
organic
membrane
as
in
a
drum,
a
vibrating
reed
as
in
the
wind
instruments
or
the
lips
of
the
musician
as
in
the
brass
instruments,
Most
important
to
us,
it
can
also
be
the
cone
of
a
loudspeaker,
When
a
recording
of
a
musical
instrument
is
made
a
microphone
converts
the
air
pressure
waves
into
exactly
analogous
electrical
voltage
waves,
If
you
were
to
graph
the
vibrations
of
the air
and
the
"vibrations"
of
the
voltage
side
by
side
they
would
be
identical
except
that
one
would
be
measured
in
volts
and
the
other
in
dynes
per
square
centimeter
-
or
something.
When
these
voltage
variations
are
re-played
through
an
amplifier
and
loudspeaker
they
are
converted
from
electrical
back
into
sound
energy,
If
all
the
links
in
the
chain
have
been
faithful
in
their
recording
and
reproducing
functions
the
pressure
waves
generated
by
the
loudspeaker
will
be
exactly
the
same
as
those
originally
generated
by
the
musical
instrument
and
the
two
will
be
indistinguishable,
Since
the
thing
that
an
amplifier
and
loudspeaker
works
with
is
not
really
sound
but
an
electrical
analog
of
sound;
and
since
it
is
possible
to
electronically
generate
any
imaginable
voltage
waveform
(
difficult
in
some
cases,
but
possible
),
it
seems
only
logical
that
at
some
point
sounds
should
be
generated
not
by
physical
musical
instruments
but
by
synthesizing
their
electronic
analog
and
then
converting
that
to
sound,
PITCH,
DYNAMICS,
TIMBRE
There
are
really
only
three
characteristics
that
determine
what
a
musical
instrument
will
sound
like:
pitch,
dynamics
and
timbre,
Of
the
three,
pitch
probably
requires
the
least
explanation.
Pitch
and
frequency
are
two
words
from
two
different
technologies
that
describe
the
same
thing.
When
an
engineer
or
technician
speaks
of
261
Hz,
they
mean
that
the
thing
they
are
referring
to
is
vibrating
261
times
per
second,
When
a
musician
mentions
middle
C
he
is
also
talking
about
something
that
is
vibrating
261
times
per
second.
If
the
musician
is
dealing
with
conventional
instruments
he
is
probably
talking
about
a
string
or
reed
but
if
he
is
working
with
an
organ
or
synthesizer
he
is
likely
referring
to
the
same
thing
that
the
technologists
were
talking
about,
the
frequency
of
the
changes
of
an
electrical
waveform,
The
human
car
is
more
sensitive
to
changes
in
pitch
than
any
other
musical
parameter,
The
intensity
of
a
sound
has
to
be
cut
significantly
before
a
listener
experiences
any
decrease
in
loudness
but
a
skilled
musician
can
tell
when
a
musical
semi-tone
deviates
by
as
little
as
3%
of
the
interval
between
that
note
and
the
next
higher
tone.
Dynamics
is
a
broad
term
that
refers
to
the
time
varying
intensity
characteristics
of
the
sound;
how
fast
it
builds
up
and
how
fast
it
dies
away.
The
length
of
time
required
for
a
sound
to
build
up
to
its
greatest
intensity
is
called
attack
time
and
this
one
parameter
conveys
more
information
about
the
way
an
instrument
is
played
than
any
other.
If
the
attack
time
is
very
short
the
instrument
will
be
in
the
percussion
family
where
the
vibrating
member
is
immediately
excited
to
its
maximum
amplitude
by
the
deforming
action
of
being
plucked
or
struck
with
a
hammer
or
mallet.
If
the
attack
is
relatively
slow
then
the
instrument
is
probably
in
the
reed
or
bowed
string
groups
where
the
action
of
the
exciting
force
-
the
wind
or
bow
of
the
performer
-
takes
a
short
time
to
fully
excite
the
vibrating
element.
4

If you forget about the talent factor for a moment the primary purpose of the
musician in playing most instruments is to serve as an energy source. The
performer pumps energy into the system (instrument) and the system dissipates it
in some way, usually as either sound or heat, I know what you're probably thinking.
Heat? Yes, heat; if you were able to accurately measure the temperature of a drum
head you'd find that it gets hotter as you pound on it, The energy that is converted
to heat can be thought of as being lost since it does not contribute to the primary
object of producing sound,
Very interesting, right? But what has this to do with the sound of a bassoon.
Just this, another important characteristic of an instrument is its release time.
That is, how fast the sound dies away, Release time is directly related to how much
of the energy goes into heat and how much into sound, A vibrating string, for instance,
is as close to lossless as you can get and its release time is very long. The stretched
membrane of a drumhead on the other hand is very lossy and as a result the release
time of drums is very short.
Reed instruments have a short release time because the reeds are relatively
lossy and don't continue to vibrate for very long after the musician stops adding
energy. Brass instruments have the shortest release time because the performer
can
force
his
lips
to
stop
vibrating
and
the
column
of
air
in
the
instrument
is
very
lossy.
Sustain time is the interval in between attack and release, the steady state
response of the instrument, As is obvious, percussion instruments have zero
sustain time - as soon as the attack is finished there is no more energy input so
it's downhill the rest of the way. Instruments that have some continuous energy
input from the performer, in the form of bowing, blowing or even pedaling in the case
case of some organs, can sustain as long as the energy holds out.
Though
attack,
sustain
and
release
are
the
primary
phenomena
of
dynamics
there
is
one
other
condition
that
is
common,
When a
percussion
instrument
is
struck
very
hard
the
vibrating
member
will
deform
beyond
the
point
at
which
a
smooth release
is
possible,
in
effect
more
energy
is
put
into
the
system
than
it
can
handle,
with
a
resulting
overload.
Under
these conditions
the
system
(
string, membrane
or
whatever)
will
rapidly
get
rid
of
the
excess
energy,
With
the
"overload"
dissipated
the
vibrating
element
will
continue
to
dissipate
the
remaining
energy
in
a normal]
fashion,
The
result
is
an initial
rapid
attack
immediately
followed
by
a
decay
time
which
is
then followed
by
a
normal
release.
In
a natural
instrument
it
would
be
all
but
impossible
for the
decay
time
to
be
followed
by
a
sustain
interval
but
with a
synthesizer
this
is
simple.
We can graphicly illustrate the conditions discussed by plotting the overall]
intensity of the sound versus time as shown in figure 1. Since these graphs are
drawn to show the peak amplitude of the sound at any given time and therefore
"contain" the sound they are often referred to as envelopes.
It is pretty obvious that as important as dynamics is, it doesn't account for all
the differences between the sounds of instruments. For instance, the trumpet and
french horn are both brass instruments with approximately the same attack, sustain
and release characteristics. They even overlap as far as pitch range is concerned
but there would be little danger of mistaking the blaring, brassy sound of the trumpet
for the mellow, muted tones of the french horn. These differences come about

+ +
a wo
mol o>
oe. J
» r
= eae
a (A) v9 (B)
= =
Oo o
time + time +
+ t
wv
3 =
2 2
2 (c) f= (D)
oO o
time + time +
FIGURE 1. Amplitude envelopes for (A) Percussion (B) reeds
(C) attack-decay-release (D) attack-decay-sustain-release
because no musical instrument produces a tone that is composed exclusively of a
single frequency, Each note is composed of a number of different frequencies, and
the number and amplitude of the various components are what gives each instrument
its distinctive timbre,
The concept that a single musical pitch
can be made up of more than one frequency
can be confusing and needs further attention.
The sine wave shown in figure 2 is the basic
building block of any imaginable accoustic or
electrical wave. It is the only waveform that
is composed entirely of a single frequency
and, more importantly, any waveform can be
built up using nothing but sine waves, FIGURE 2. Sine wave
To illustrate this look at figure 3, Here
we have two sine waves drawn in dotted lines
which are labeled "A" and ''B", As you can
see from the drawing, waveform"B" goes
through two cycles in the time that it takes
waveform "A" to complete a single cycle.
Waveform "B" is therefore twice the freq-
uency of "A" and is said to be the second
harmonic of the fundamental frequency "A""
If we draw another wave that was three times
FIGURE
3.
Fundamental
and
the
frequency
of
"A"
it
would
be
the
third
énd. harmonic harmonic, four times would be the fourth
harmonic, five times the fifth and so on,

If at every point in time we sum together the amplitudes of waveforms A and B
the result is the waveform shown by the solid line, Note that while the new wave is
shaped differently than either A or B it has the same frequency ( and consequently
pitch) as the fundamental frequency A, If third, fourth, fifth and higher order
harmonics were added into this wave the result would continue to change shape but
the frequency would remain the same,
It is not necessary that every harmonic of a fundamental frequency be included
in a wave and indeed the most musically interesting sounds have certain harmonics
deleted, The square wave shown in figure 4 is a good example. It is difficult to
imagine that the sharp-edged wave
illustrated could be built up from
smoothly changing sine waves but it can
as shown in the progression of diagrams
figure 5 (a) through (c). In (a) fundament-
al frequency is added to its third harmonic
producing the waveform shown by the solid
line, In (b) the fifth harmonic has been
added to the result of (a) to produce the
new solid waveform and in (c) the seventh
FIGURE 4. Square wave harmonic has been added to all the rest.
You can see that the trend as higher order
harmonics are added is to steepen the
sides of the square and flatten and reduce
the ripple in the top. When enough har-
monics have been added the result will
be a square wave, Notice in particular
that not all harmonics are added together
for a square wave, only the odd harmonics
(3rd, Sth, 7th, etc.) are included,
MAKING WAVES
Now that we have a pretty good idea
of why instruments sound the way they do
we can looking at ways of duplicating these
sounds using electronic circuits,
The first method of electronically
producing a desired waveform is called
additive synthesis and the technique should
be obvious from the discussion of harmonic
structure, Several oscillators provide a
source of sine waves of various harmonic-
ally related frequencies and combinations
of the outputs are summed together to
build up the desired waveform. By chang-
ing the amplitudes of the sine waves
practicaly any waveform can be easily
produced, One of the problems with this
system is keeping ali of the oscillators
tuned so that they are multiples of one
figure 5.

another,
Most
electronic
organs
that
use
additive
synthesis
systems
get
around
this
problem
by
using
one
oscillator
for
the
highest
frequency
component
desired
and then producing
the
other
frequencies
using
a
chain
of
frequency
dividers.
The
technique
used
in
synthesizers
is
called
subtractive
synthesis
and
can
be
thought
of
as
just
the
opposite
of
additive
synthesis.
Rather
than
summing
together
the
frequencies
you
do
want,
you
start
off
with
a
source
that
is
already
rich
in
harmonics
and
then
remove
the
ones
you
don't
want.
This
may
seem
a
rather
strange
way
to
get
from
here
to
there
but
there
is
an
excellent
biological
precedent
for
subtractive
synthesis,
the
most
versatile
musical
instrument
of
all-
the
human
voice.
There
are
other
reasons
for
using
subtractive
synthesis
than
just
pleasing
mother
nature,
If
we
are
going
to
be
consistent
in
our
design
of
a
line
of
voltage
controlled
equipment,
then
everything
should
be
voltage
controlled,
including
the
oscillators,
Designing
a
voltage
controlled
sine
wave
oscillator
is
not
impos-
sible
but
it
is
difficult.
Then
there
is
the
need
to
lock
all
the
sine
wave
oscillators
to
precise
multiples
of
the
fundamental
frequency,
and
a
mixing
system
to
combine
all
the
harmonics
in
the
proper
amplitude
relationships.
As
you
can
see,
a
system
of
this
type
would
be
quite
cumbersome
and
tedious
to
operate.
From
a
technological
standpoint,
it
is
much
easier
to
electronically
generate
a
complex
ramp
or
square
wave
than
a
sine
wave,
And
with
the
many
recent
refinements
in
filter
circuits,
subtractive
synthesis
is
definitely
the
better
route
to
take,
Since synthesizers operate with
harmonic rich waveforms as their pri-
mary
signal
source
there
is
no
need
to
/WV
MM
Ln
start out with a sine wave at all, The triangle} ramp {square
VCO's supplied with most synthesizers
provide a variety of waveforms each of
which provides different harmonic
structures, Common practice is to use
a relaxation oscillator to generate a
voltage ramp which is then converted
to
triangle
and
pulse
waves
using
simple
shaping circuits, In some cases the
triangle will also be shaped into a sine
wave, These waveforms and their
harmonic contents are listed in table 1.
In order to use subtractive synthesis
we need some means of getting rid of the
harmonics we don't want and to do this
we use filters. A filter is quite simply
an electronic gadget that eliminates a
single frequency or group of frequencies, Table 1.
Harmonic content of
triangle, ramp and square wave
n=3,142 n2=9,872

Figure 6 shows diagramatically a repre-
sentation of the frequency response of a low
pass filter, This drawing shows that as the
frequency of the signal being fed to the input
of the filter increases the amplitude of the
filter's output falls off, Note that the filter frequency >
does not change the frequency of the input FIGURE 6.
signal, only the amplitude, If the input is
a complex waveform the filter will of course
change the signal's shape as it attenuates the
higher frequency components but that is, after
all, what we're after,
output >
Low-pass filter
output >
Figure 7 shows the frequency response
of
a
high
pass
filter,
In
this
case
the
amp-
litude of the output falls off as the input
frequency
decreases,
FIGURE
7.
High-pass
filter
frequency >
Notice
that
in
both
of
these
filters
the
response
curve
is
flat
either
up
to
or
beyond
some
definite
frequency.
This
is
the
frequency
at
which
the
filter
begins
to
"take
hold"
and
is
designated
the
cutoff
frequency
or
fo.
One
other
important
parameter
associated
with
low
pass
and
high
pass
filters
is
the
roll
off
rate,
ordinarily
measured
in
units
of
db/octave.
This
sounds
complicated
but
it's
really
not,
A
decibel
(db
)
is
a
measure
of
electrical
level
and
when
you're
talking
about
voltage,
a
change
of
6
db.
corresponds
te
halfing
(
if
-
6
db.)
or
doubling
if
(@
6
db,
)
the
original
reference
level.
Octaves
are
of
course
frequencies
that
are
double
some
reference
frequency;
thercfore
a
low
pass
filter
that
"rolls
off
at
6
db,
/octave
simply
means
that
every
time
the
frequency
is
doubled
the
output
of
the
filter
falls
by
1/2.
Figure 8 shows the frequency res-
ponse of a band pass filter, As the dia-
gram implics, a band pass filter attenuates
all frequencies above and below a certain
frequency while allowing the frequency of
interest
(
or
frequencies
close
to
it
)
to
aes
pass
without
being
effected.
The
frequency
frequency
>
that is allowed to pass without attenuation FIGURE 8. Band- Fi
is
quite
logically
called
the
center
frequency
oo
Band=pass
“filter
and
is
also
designated
f,,
There are
parameters that can be used to specify
how well the filter does its job of rejecting
frequencies outside of its pass band but
none of them are very easy to understand
and for our purposes we will confine our—
selves to speaking of the 'Q" (quality) of
the filter, The higher the 'Q" the greater
the
frequencies
outside
the
pass
band
will
frequency
>
be
attenuated,
FIGURE
9.
Notch
filter
As the frequency response curve of
figure
9
shows,
you
can
think
of
a
notch
filter
as
being
the
opposite
of
a
band
pass
filter,
Instead
of
allowing
frequencies
around
the
center
frequency
through,
the
notch filter blocks these and allows all others to pass,
output >
output >

CONTROLLERS
It is about time that we looked at a problem that has plagued instrument makers
since the first caveman beat on a hollow log - how to control the instrument in such
a
way
that
you
realize
its
full
potential,
With
most
conventional
instruments
the
control system is obvious, You control some of the elements of the dynamics by
how hard you blow, pick, or strike the instrument and you control the pitch by the
positions of your hands and/or lips, Timbre is in most cases a quality of the
instrument and is therefore beyond the control of the performer,
This is not the case with a synthesizer; you have at least the theoretical
capability of controlling and varying every characteristic of the sound. Some
characteristics you can pre-set by the position of a knob and some you can turn
over
to
automatic
function
generating
equipment,
Some
parameters
are
varied
with
a
manual
controller
such
as
a
keyboard
and
some,
unfortunately,
you
wind
up forgetting about because there are no more controllers available,
Before
examining
some
of
the
types
of
controllers
that
are
available,
make
sure
that
you
have
firmly
implanted
in
your
mind
that
a
controller
for
a
synthesizer
does
only
one
thing;
it
provides
a
voltage
proportional
to
some
parameter
that
is
physically
changed
by
the
performer.
While
in
most
cases
the
voltage
produced
by
the
controller
will
subsequently
be
used
to
set
the
pitch
of
a
VCO,
this
will
not
always
be
the
case,
Depending
on
the
sound
being
produced,
the
controller
may
also
be
used
to
set
the
center
frequency
of
a
band-pass
filter,
roll
off
rate
of
a
low
pass
filter
or
any
number
of other
things.
KEYBOARDS
When
used
to
control
a
piano,
a
keyboard
is
one
of
the
greatest
inventions
of
all time,
When
used
with
a
synthesizer
it
is at
best
a compromise,
Musicians
are
used
to
keyboards
being
connected
to
polytonic
instruments,
that
is,
instruments
that
are
capable
of
playing
as
many
notes
at
one
time
as
the
number
of
keys
being
pressed
down,
With
most
basic
synthesizers,
this
is
not
the
case,
Basic
keyboards
are
by
design
a
monotonic
controller
capable
of
producing
one
control
voltage
output
at
atime,
This
means
the
synthesizer
is
to
be
played
like
a
saxaphone
or
trumpet
-
one
note
at
a
time.
Some
intermediate
line
synthesizers
use
a
clever
switching
arrangement
to
produce
two
notes
at
a
time,
but
this
is
still
far
from
a
true
polytonic
system.
Development
of
micro-
processors ( a computer in an Integrated Circuit ) will be the technological step
which will finally allow development of fully polytonic keyboards - playing more
than one note at a time, and having individual voicing for each additional note being
played,
Since the electronic organ has become commonplace, performers have gotten
used to the idea that they can't control the dynamics of their instrument by varying
the striking force on the keys, this is also true of most synthesizers. Other than
triggering signals that are generated when any key is pressed, the only control
voltage that most keyboards produce is proportional to the location of the key
being activated. One manufacturer has a keyboard that is an exception to this
rule; in addition to the standard control voltage it also generates two voltages
proportional to the velocity of the key as it is pressed down and the final pressure
on the key as it is held down, This is a significant improvement since it allows
the performer to directly influence three musical parameters by pressing a single
key,
10

Strangely
enough,
the
original
objection
to
a
keyboard
that
was
mentioned
in
the
first
part
of
this
booklet
(
unavailability
of
pitches
between
semi-tones
)
is
not
a
great
problem
on
a
synthesizer,
Most
keyboards
provide
a
''pitch''
knob
that
allows
some
variation
in
tuning
of
the
instrument
and
many
provide
for
an
automatic,
variable
rate
glissando,
In
spite
of
its
drawbacks,
the
standard
keyboard
has
one
big
thing
in
its
favor
-
familiarity.
It
is
similar
to
a
thing
that
the
musician
already
knows
how
to
use
and
re-training
time
is
therefore
reduced,
LINEAR CONTROLLERS
These
are
electrically
and
mechanically
the
simplest
of
all
controllers.
Most
consist
of
a
long
strip
of
electrically
resistive
material
with
a
voltage
applied
to
each
end,
The
potential
difference
between
the
two
ends
distributes
evenly
along
the
length
of
the
strip
so
that
the
voltage
between
any
point
and
electrical
ground
is
proportional
to
the
position
of
that
point
on
the
strip,
When
the
performer
presses
on
the
controller,
a
conducting
metal
band
makes
contact
with
the
resistance
element
and
picks
off
the
voltage
present
at
the
point
of
contact,
Linear
controllers
are
generally
not
intended
as
substitutes
for
keyboards
for
a
number
of
reasons,
First,
it
is
technically
difficult
to
automatically
pro-
duce
a
trigger
pulse
whenever
the
controller
is
pressed,
This
function
has
to
be
performed
manually
with
a
seperate
switch
that
must
be
closed
for
each
note
or
run
that
is
to
be
played.
Secondly,
using
a
linear
controller
for
pitch
is
like
playing
a
fretless
instrument
such
as
a
violin,
it
requires
considerable
experience
to
know
what
pitch
is
going
to
be
produced
at
a
given
location.
These
devices
come
into
their
own
when
used
in
conjunction
with
a
keyboard,
In
this
application
they
can
provide
an
auxillary
control
for
some
parameter
other
than
pitch,
like
manually
sweeping
a
filter
or
controlling
the
amount
of
noise
mixed
into
a
sound,
The
control
voltages
produced
by
this
unit
can
also
be
summed
into
one
of
the
VCO
control
inputs
to
produce
a
manually
controlled
glissando
or
vibrato
or
can
be
used
with
a
VCA
to
give
manually
controlled
tremolo,
FOOT PEDALS
Foot
pedals
allow
you
to
control
additional
musical
parameters
with
your
feet.
They
are
similar
to
the
expression
pedals
on
electronic
organs
except
that
instead
of
controlling
only
the
volume
they
can
be
used
to
control
filters,
oscillators
or
amplifiers.
Anything
you
can
say
about
linear
controllers
applied
to
foot
pedals,
they're
intended
to
be
used
in
conjunction
with
a
keyboard,
JOY STICKS
These
are
the
wackiest
controllers
imaginable
and
as
you
would
expect
are
similar
to
the
joy
sticks
used
in
airplanes,
The
biggest
thing
going
for
this
type
of
controller
is
that
it
offers
the
possibi-
lity
of
directly
controlling
four
musical
parameters
simultaneously.
One
parameter
could
be
controlled
by
moving
the
stick
forward
and
backward,
another
by
moving
the
stick
from
side
to
side,
a
third
control
voltage
could
be
generated
proportional
to
vertical
motions
(
along
the
long
axis
of
the
stick
)
and
a
fourth
porportional
to
Nn

12
the
rotation
of
thehandle,
If
you
like
you
could
even
put
a
switch
on
top
to
control
such
vital
functions
as
self-destruct.
A joy stick scems like a valid concept but anyone that could use one properly
probably wouldn't be able to communicate with earth people,
ENVELOPE GENERATORS
Envelope generators are automatic controllers that electronically gencrate
atime varying voltage as pre-set by the positions of knobs or sliders, An
envelope
generator
ordinarily
responds
to
a
trigger
pulse
by
generating
an
electric-
al
waveform
that
rises
to
some
pre-set
valuc
in
a
pre-set
time,
sustains
that
level
as
long
as
the
trigger
pulse
is
present
(
or
for
a
pre-set
time
in
some
cases)
and
then
falling
back
to
zero
in
a
pre-set
time,
Some
envelope
generators
are
capable
of producing the attack, decay, sustain, release type functions discussed earlier,
The output of the envelope generator can be used in the same ways any other
control
voltage
source
should,
but
these
items
find
their
most
common
application
in controling dynamics and time varying timbral qualities of a sound,
A low frequency oscillator can also serve as a control voltage source to
provide cyclicly varying voltages for vibrato, tremolo or filter sweeping,
SEQUENCERS
Sequencers fall under the category of control devices much like keyboards.
The main point of a sequencer is that it is programmed ahead of time to provide
a specific pattern of control voltages. On command, the series of voltages
appears at the output, and this pattern can be repeated many times with high
consistency, Standard analog sequencers are programmed via potentiometers
to provide the desired voltage for each stage, After programming, the module is
turned on either manually or from remote trigger signals, and the internal clock
steps through the sequence to "read" the programmed voltages, Digital tech-
nology has aided in the development of digital sequencers which differ in one
main area, Rather than manually programming the voltage for each stage, an
input is provided to accept contro! voltages from a keyboard or other control
device. As the keyboard is played, the various control voltage steps are trans-
lated into a digital word and stored in a memory system, On command, the
sequencer reads its output information from the memory rather than from a
string of potentiometers. These devices are still costly, but as computer technology
increases the cost will continue to drop,
Although sequencers are usually thought of as being used to control an
oscillator ( repeating bass lines, high speed arpeggios ctc,) many interesting
effects can be obtained by using them as controllers for filters, VCA's or any
other modules requiring a control voltage.
DIGITAL COMPUTERS
As short as two years ago, thoughts of digital computers being interfaced
with an electronic music synthesizer were reserved for well funded college
music departments or government subsidized artistic research programs, Since
then, computers have grown smaller, more inexpensive, and basically more
accessible, Electronic hobbyists are delving into these machines as they did
with Ham radio ten years ago, Realizing a computerized music system is

practical today, Initially, we will seé computers being used to control the analog
sound generating and processing modules of the synthesizer, Then, as prices
decrease, fully computerized synthesizers will be seen which will generate any
waveform, changing waveform, amplitude, and other parameters as you play -
depending on how the computer is programmed. Eventually, we may even see
computer music systems which can "listen" to a sound and reproduce the same
sound under operator command. Or how about a machine that monitors the
operators brain waves and vital signs, and composes music specifically to fit
his moods /thoughts. The list is endless.
THE EQUALLY TEMPERED SCALE
This is as good a place as any to bring up the subject of the equally tempered
or chromatic scale, As anyone who is reading this knows there are 12 semi-
tones in each octave of the chromatic scale, 7 naturals labeled A through G and
5 accidentals that are designated as either sharps or flats of the naturals, With
two exceptions the sharp of one note is identical to the flat of the next highest note,
the exception being that there are no accidentals between B and C, or E and F so
that B# is the same as C and F» is the same as E,
For each octave increase in the musical scale the frequency of the note
doubles so that since middle C corresponds to 261.6 cycles per second the next C
above middle C is 523.2 cycles per second,
Somewhere back in antiquity ( around the time of J, 8. Bach ) some genius
decided that since there are 12 semi-tones to the ocatve and each octave doubles
the frequency, each note should be related to the note directly below it in the
scale by a factor of the twelfth root of two. Just in case you're not used to
working out the twelfth root of numbers in your head, this translates to 1. 059 times
times the frequency of the note directly below it. The significance of this is that
as pitch increases, the difference between adjacent notes in the scale also increases,
All this may seem like academic trivia until you realize one point, All volt-
age controlled uscillator designs produce a device whose output frequency is directly
proportional to the control voltage, identical control voltage changes produce
identical frequency changes.
An example will most readily demonstrate the significance of these facts,
Suppose that we have a keyboard that produces a control voltage of . 625 volts
when its lowest C is pressed, The voltage corresponding to the next C up is
quite logically 1.25 volts but don't fall into the trap of thinking that the voltage
corresponding to the third C is 1.25 plus . 625 or 1. 875 volts because it's not,
it should be twice the voltage required for the second C or 2 X 1, 25 = 2.5 volts.
Many synthesizer designers use an electronic conversion device to get
around this difficulty, This device converts a linear controller output voltage
(lv. for the first C, 2v. for the second, 3v. for the third C, etc.) to the
octavely related voltage required by the VCO,
Unfortunately, the exponential converter circuits ( as these devices are
known ) are not only expensive but also quite often they tend to drift so that
even for a fixed input voltage the output voltage ( and of course pitch of the VCO)
wanders from one value to another,
13

A simple means of getting around this is to have the keyboard generate
octavely related voltages in the first place. This eliminates the need for a
separate exponential converter on each oscillator and filter. And when you're
talking about 3 or 4 VCOs and a couple of filters, this is a big savings in cost
and drift problems,

THE 4700/8 AS A SYSTEM
The 4700/S is an assortment of the previously discussed modules which has
capabilities competitive with commercial systems selling for several times the
price, In fact, there are very few, if any, synthesizers in the "under $2000"
bracket that include a sequencer as a standard function. Also, many low cost
filters provide only low pass output, or a choice of low pass or band pass. Rarely
do
you
see
three
simultaneous
filter
outputs
available.
The
module
complement
of the 4700/S is such that the sequencer can control a full synthesizer by itself,
and you still have a full array of modules for use with the keyboard.
Another concept of much debate is the modular patchable system versus the
pre-patched
switch
operated
system,
There
is
no
argument
that
the
normalized
synthesizers fill a very useful role - specifically the musician who wants to make
ultra-fast changes on stage while he is performing, or someone who wants to
add some new sounds to their repertoire without the need to fully understand
what they are doing. But for those of us who want to really learn about the nature
of
sound,
the
patchable
system
can't
be
beat.
Being
able
to
utilize
any
module
in
any
configuration
draws
every
drop
of
performance
out
of
the
circuitry.
After
mastering
the
operation
of
the
modules,
most
any
sound
that
you
can
imagine
can
be
duplicated
on
a
patchable
system.
Further,
the
modular
concept
implies
an
obsolescence
proof
system,
for
as
your
ideas
grow
you
can
expand
your
system
with
a
few
more
modules
rather
than
selling
the
old
system
and
replacing
it
with
a
bigger
better
model.
Most
people
find
that
after
using
a
patch
system,
it
doesn't
take
that
much
longer
to
insert
a
few
cords
than
it
does
to
flip
several
switches,
Thus,
more
and
more
patchable
systems
are
being
seen
on
the
road
with musicians,
The
following
section
contains
several
examples
of
patches
for
the
4700/8
system,
Starting
with
these
sounds
will
help
explain
the
uses
and
capabilities
of
the
modules,
After
initially
setting
up
the
patch,
experiment
with
altered
settings
on
the
controls
and
listen
to
the
changes
that
occur.
This
type
of
experimentation
is
very
helpful
in
becoming
more
familiar
with
the
system
and
how
it
works,
The symbology used in the following diagrams to represent the various
modules is fully explained in the PAIA manual ''A Schematic Symbology For
Synthesizer Patching Arrangements" which is provided with this manual.
We
have
chosen
to
represent
synthesizer
patches
in
this
manner
because
many
people
physically
arrange
their
modular
systems
in
different
configur-
ations.
The
symbology
system
is
universal
and
can
be
adapted
to
any
synthesizer.
Best
of
luck
in
your
experiments
with
the
4700/S;
welcome
to
a
whole
new
world
of
sound,
23

EFFECTS
USING
WHITE
NOISE
AS
A
SIGNAL
SOURCE
The
randomness
and
wide
frequency
content
of
white noise
makes
an
ideal
basis
for
many
sounds
both
musical
and
special
effect.
The
most
obvious
is
simulating
the
wind
or
the
sound
of
the
surf.
The
noise
output
of
the
2720-5
Control
Oscillator
Noise
Source
is
patched
to
the
4730
filter
input.
The
band-
pass
output
is
the
output
of
the
patch,
Asa
control
source
for
the
filter
you
can
use
a
randomly
tuned
sequencer
with
a
lot
of
glide,
an
envelope
generator
with
long
attack
and
release
times,
the
keyboard
with
long
glide
time,
or
the
control
oscillator
output.
Each
of
these
controllers
gives
a
different
effect,
but the
common
drawback
is
that
they
aren't
completely
random
like
Mother
Nature's
wind
and
surf,
To
increase
randomness,
feed
more
than
one
of
these
controllers
into
the
three
control
voltage
inputs
of
the
filter,
The
random
sequencer
pattern
can
be
occurring
continuously
while
the
keyboard
output
is
user
controlled
to
make
the
wind
sweep
higher
or
lower
depending
on
the
location
of
the
key
pressed,
Additionally,
an
envelope
generator
can
be
triggered
by
the
keyboard
to
cause
an
instantaneous
gust
of
wind
settling
back
to
the
level
determined
by
which
key
was
pressed,
Increasing the Q of the 2720-5
filter
causes
a
narrower
4730
band of frequencies to be
passed
through
the
filter.
OUTPUT
With Q at maximum and
the keyboard as the only
control source, you can
generate
a
melodic
wind
KEYBOARD
H
and
actually
play
melodies.
OUT
Baki!
[so
|
t
Using
an
envelope
gen-
ro
erator
to
control
a
filter
with
H
low
Q
setting
and
range
switch
|
ADSR
[7
‘
at LOW,
some
very
effect-
ive thunder and explosion LFO
effects can be added, Also,
try the low pass output for
these
type
of
effects.
The
arias
envelope
generators
will
most approximate an explo- is
sion with
a
minimum
attack
Rigure!d
=
D/SURF
and long release.
(SINGLY OR
COMBINED)
Using the noise source
through a VCA which is con-
trolled by an envelope gener-
ator will produce hi-hat
cymbal effects. Set the
envelope attack at minimum
and the release at about 20%.
Patch the keyboard pulse trigger
to the input of the envelope gen-
(USED ASA
erator. Every time you hit a key
the cymbal sound will be produced, aces SOURCE
OUTPUT
Figure 2 - CYMBAL
24

To
add
more
expression
to
the
sound
keep
your
hand
on
the
release
control
of
the
envelope
generator
and
vary
the
setting
for
different
beats.
This
gives
the
effect
of
the
hi-hat
cymbal
opening
and
closing,
A
keyboard
controlled
snare
drum
can
be
conceived
which
allows
the
user
to
play
a
melody
on
the
drum.
The
keyboard
control
voltage
output
sets
the
pitch
of
the
VCO
to
duplicate
the
"strike
tone"
normally
provided
by
the
tuned
drum
head,
The
triangle
or
sine
VCO
output
is
mixed
with
white
noise
at
the
mixer
and
processed
through
a
VCA,
The
envelope
generator
for
the
VCA
should
have
minimum
attack
and
decay,
maximum
sustain
and
medium
release,
Triggering
the
envelope
generator
with
the
pulse
keyboard
trigger
source
can
be
filtered
prior
to
being
mixed
with
the
strike
tone.
Experiment
with
assorted
combinations
of
filter
output
and
Control
sources
such
ag
key-
board,
envelope
generator,
or
fixed
bias
(as
outlined
in
the
string
patch
OUT
Figure
3
-
MELODIC
SNARE
DRUM
4710
OUT
COMMENTS:
4782:
Pitch
-
Set
to
upper
octave,
Glide
-
Off.
4720:
Initial
Freq.
-
Around
30
to
50%.
4740:
Attack
—
20%.
Decay
-
Minimum,
Figure
4
-
FLUTE
Sustain
-
Maximum,
Release
-
20%,
This
is
a
basic
patch using
the
soft
sine
wave
of
a
VCO
to
duplicate
the
sound
of
a
flute.
The
medium
attack
and
release
on
the
envelope
duplicates
the
effect
of
the
build-up
and
collapse
of
the
air
column
in
the
flute,
The
envelope
generator,
being
triggered
by
the
step
trigger
from
the
keyboard,
will
generate
a
sustained
signal
as
long
as
a
key
is
depressed.
The
balanced
modulator
is
being
used
as
a
VCA
for
this
patch,

{2) 4720
OUT
Figure 5 - FUNK BASS
COMMENTS:
Keyboard: Set to low octave,
4720's: Both VCO's use ramp output.
VCO's tuned in unison at lowest setting.
4711: Mix two inputs equally.
4740: Attack - Minimum.
Decay - 20%.
Sustain - 60%.
Release - 30%
4730: Lowpass Output.
"Q'' Control - 50% to 100%.
Initial Freq. - Below 50% if range switch is "thigh".
Maximum if range switch is "low'.
This
patch
gives
the
funky
"wow"
bass
sound
made
popular
by
Stevie
Wonder,
the
Bee
Gees
and
many
others.
The
4740
variable
output
can
be
changed
manually
while
you
are
playing
for
added
expression
and
variation,
(214720
OUT
4740
Figure 6 - CHIMES ( AND OTHER METALLIC SOUNDS )
26

COMMENTS:
Keyboard: High octave, no glide,
4740: Attack - Minimum.
Decay - Minimum.
Sustain - Maximum.
Release - 50 to 100%.
4720's:
Tune
triangle
VCO
to
an
augmented
fourth
above
the
pulse
VCO.
Pulse
VCO
pulse
width
-
30%,
This
is
a
general
purpose
patch
for
bells.
The
settings
shown
will
provide
a
sound
of
tubular
chimes
or
church
bells, Experiment
with
various
permutations
of
this
patch,
How
about
a
chime
with
sustain?
To
do this,
use
the
keyboard
step
output
rather
than
the
pulse,
Also
play
with
the
4740
to
get
chimes
with
slow
attack,
etc.
The
biggest
variables
in
this
patch
are
the
VCO's.
(3) 4720
BIAS SUPPLY
4740
Figure
7
-
STRINGS
COMMENTS:
Keyboard:
Pitch
control
-
high
range.
No
glide,
Vco's:
Tuned
in
unison,
pulse
output.
Initial Freq. - Approximately 30%.
Initial Pulse Width - 10 to 20%.
LFO (Low Frequency Oscillator):
Approximately 10 Hz.
Very small output amplitude - 10%.
Mixer; All signals mixed equally.
Filter: Low pass output; Range- maximum; Q Control - 50%.
ADSR: Attack ~ 20%.
Decay - 30%.
Sustain - 60%.
Release - 50%.
Bias Supply (See Text) - Approximately 3 volts.
27