An Electronic Music Primer

Author(s): John David Weinland

Source: Journal of Music Theory, Vol. 13, No. 2, (Winter, 1969), pp. 250-275
Published by: Duke University Press on behalf of the Yale University Department of Music
Stable URL: http://www.jstor.org/stable/842989
Accessed: 07/08/2008 09:39

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=duke.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact support@jstor.org.

http://www.jstor.org
250

An

Electronic Music

Introduction

Composing in an electronic music studio combines the functions

of composer and performer, since not only are sound sources
specified and organized but, in the process, the finished musi-
cal entity is performed as well. This paper will elucidate the
composer-studio interaction in terms of its consequences for
both the compositional and performance aspects of electronic
music realization. A brief outline of study will be presented
with the intended goal of informing students about the operation
of electronic music studio equipment, practical studio tech-
niques and exercises, and guidelines on compositional method.
Although this information is primarily relevant to the so-called
'classic' analog electronic music studio, much of it is general
enough to be applied to semi and fully- automated studios as well.
 251

Primer

JOHN DAVID WEINLAND

This paper will first outline some preliminary cautions on stu-
dio operations, then will deal with equipment operation, basic
studio skills, and finally with compositional techniques.

I. Preliminary Cautions

A. Aural fatigue is a depression in the normal of

response
the ear to sound. In severe cases, such depression in hearing
sensitivity (an attenuation in the apparent volume of a given
sound) may be accompanied by tinnitus ('ringing' in the ears),
headache, and tiredness. Aural fatigue is particularly danger-
ous inasmuch as exposure to sounds loud enough to cause this
condition may also lead to hearing loss or deafness. Sounds
which cause and aggravate aural fatigue are sounds of a pri-
252

marily steady-state nature -regardless of dynamic level - and

aperiodic sounds at high dynamic levels of a harsh, surprising
nature. Both must be assiduously avoided, and frequent breaks
of a few minutes in the course of a working period are advisable
to minimize aural fatigue. Symptoms of this fatigue include a
marked decrease in the apparent background noise level (traf-
fic noise, etc.), increased irritability, headache, and drowsi-
ness.

B. Beginning students should not use earphones until they are

thoroughly familiar with the equipment in the electronic music
studio. The chance of accidentally connecting a pair of ear-
phones to an unusually loud sound source is great and such
sounds may cause temporary or even permanent damage to the
ear. Even the advanced student should be careful to remove
earphones from his head before connecting them or disconnect-
ing them from any other piece of equipment.

C. It is virtually impossible to damage most equipment found

in electronic music studios short of physical violence done on
the mechanism. There are three exceptions: (l)loudspeakers,
(2) tape recorders, and (3) amplifier outputs.

A loudspeaker is an electro-mechanical transducer consisting

of a thin membrane (usually paper) set in motion by a coil of
wire which is attached to the membrane and placed in a mag-
netic field. The amplitude of the sound depends on the distance
travelled by the membrane (or its excursion), which in turn
depends on the amount of current flowing in the coil of wire.
Loud sounds require greater excursions of the membrane than
do soft ones and sounds of an excessive amplitude will move
the membrane to the point where it tears. Such loud sounds
are often incurred when the composer attempts to use pure, or
sine, tones lying at the ends of the audio spectrum; i.e., below
100 Hz. and above 10,000 Hz. Inasmuch as human hearing
response is weaker at these extremes than in the middle of the
audio range, corresponding increases must be made in the
volume of such extreme sounds so that they may be heard as
being as loud as tones in the mid-range. It is these increases
of amplitude at either hearing extreme as well as exceedingly
loud percussive sounds which may damage loudspeakers.

Since most professional studio tape recorders have exposed

moving parts, it is important that these areas be kept clean
and free of oil, cigarette ash, and dust. Tape recorder heads
are very delicate and nothing should contact their surfaces ex-
cept tape, or an alcohol-moistened cotton swab (for cleaning
 253

purposes). Grease pencils, used for marking tapes, should,

under no circumstances, touch any of the heads.

Inputs and outputs of most studio equipment are of a low-voltage

nature and no damage can arise (usually) from accidental in-
correct connections between pieces of equipment. Amplifier
outputs, on the other hand, put out significantly more power at
higher voltages than other equipment. Amplifier outputs should
only be connected to loudspeakers or earphones.

D. Cleanliness is of paramount importance when working with

tape, tape recorders, and associated equipment. The student
should accustom himself to washing his hands before and, at
odd moments, during work. Body oils accidentally deposited
on the ernulsion of magnetic tape cause a reduction of high-
frequency response, losses of signal (called 'drop-outs'), and
poor-quality tapes in general. All working areas should be
kept oil and dust-free.

II. The Equipment

The production of electronic music has associated with it four

categories. These are: (A) Listening systems; (B) Sound gen-
eration; (C) Sound modification; (D) Sound storage. These
functions are performed by the following types of equipment:

(A) Listening systems: Amplifiers, loudspeakers, ear-

phones.
(B) Generation: Electronic oscillators (or generators),
white-noise generators, and sounds replayed on tape
recorders.
(C) Modification: Filters of various kinds, envelope con-
trol devices, mixers, reverberation devices, varia-
ble-speed tape recorders, direct physical modification
of the tape.
(D) Storage: Tape recorders.

Evidently, these categories overlap in some areas. Tape re-

corders, for example, may be used as 'generators' or sound
sources in the playback process. Tape recorders are also
used as modification devices in some cases (e.g., variable-
speed recording, sound-on-sound, and tape recorder rever-
beration).

All stuidos, as well, have some facility for the interconnection

of these devices. Equipment interconnection is provided for
254

either by a patch-panel (or Patch-board), or by an array of

switches (called a switch-board). A patch-panel is a series of
electronic terminations - which are the
inputs and outputs of
the various machines in the studio- arranged as a series of
'jacks' (or connecting points) on a board. Interconnection is
brought about by connecting one 'jack' to another by means of
a passive conducting wire with 'plugs' at each end (a patch-
cord). A switch-board uses switches to accomplish the job of
the interconnecting wires - or patch-cords - in the patch-board
system. With this system, the operator flips a switchtointer-
connect two pieces of equipment rather than plug in a patch-
cord. The switch-board system has the advantage of speed and
convenience although its flexibility is only as great as the num-
ber of interconnection possibilities wired into its switches.
The patch-panel arrangement has the advantage of a greater
number of interconnection possibilities although it is slower
and more awkward than the switch-board system.

A. First, let us examine the listening apparatus - amplifiers

and loudspeakers - and their method of interconnection. The
following operating suggestions should be observed.

(1) Amplifiers must not be operated without a load (i. e., with-
out being connected to a loudspeaker or earphones). Some
amplifiers are susceptible to extensive damage when operated
without such a load.

(2) The input to an amplifier should be connected only when the

volume, or gain, control is turned down tominimum; otherwise
hum and pops originating in the switch-board or patch-panel
may be transmitted to the loudspeakers at a high enough level
to cause damage.

(3) If there are any controls on the speakers, they must be in

a standarized and agreed-upon position. The entire playback
system should be designed so that, for a given input to an am-
plifier, the output sound will always be the same. The com-
poser should be aware of those variables in the playback sys-
tem which could affect his correct judgment of the volume,
timbre, phasing or equalization of a sound. For most amplifier-
loudspeaker combinations, these variables are: the amplifier
volume (or gain) control, balancing controls on the loudspeaker
itself, phasing controls on the amplifier, speaker placement,
and studio acoustics.

B. Next a student should learn how to connect sound sources

or generators to the listening system. On-off switches, power
 255

indicator lamps, and other controls on the generators should

be demonstrated and their functions explained. The timbres
of various periodic wave-function generators usually found in
electronic music- the sine wave, square wave, sawtoothwave,
and triangular wave - and any other generators found in the
studio (usually a white-noise generator of some description)
should be memorized. Exercises should be formulated in
equipment interconnection, since mastery of minor problems
associated with turning equipment on and off, proper operation
and comprehension of the switch-board or patch-panel, and use
of gain controls and frequency controls will result in the ac-
quisition both of information applicable to more intricate situa-
tions and good working habits. At the beginning, a simple
problem may be posed (making sure that the equipment involved
represents the normal or typical situation), then complicated
by the teacher in every possible way. The student will then be
able to solve any problem associated with that particular situa-
tion and will begin to know which information is fundamental
enough to be used in the solution of other interconnection prob-
lems.

A beginning student should learn to interconnect any generator

with any amplifier-loudspeaker and hear the sound he desires.
Inasmuch as most studios use equipment of diverse manufac-
ture, general concepts of on-off-ness, volume- control-ness,
frequency specification and so forth will be formulated. An
understanding of the abstract properties of machines and fa-
miliarity with cues and signs associated with individual pieces
of equipment will supply the necessary generalizing power for
comprehension of most other equipment.

Although the acquisition of good working habits should never

interfere with profitable experimentation, systematization of
many routine or trivial studio operations will save composing
time and the establishment of operating habits, such as making
equipment fully operable after turning it on, will save time and
embarassment over, for example, turned-off amplifiers which
don't function, unplugged speakers, oscillators set outside the
limits of hearing, and so forth.

Assuming the mastery of the operations associated with gener-

ators, amplifiers, and loudspeakers, investigation should now
be made into modification devices.

C. Modification devices may be divided into the following

categories: (1) Filtering, (2) Envelope control, (3) Combination
devices, (4) Reverberation, (5) Tape editing techniques, and
256

(6) Tape recorder effects.

(1) Filtering is the process of selecting certain portions of the

audible spectrum for attenuation or boosting. Filtering has
the effect of changing to a greater or lesser degree the timbre
of a sound.

(a) The most common type of filter is the band-pass filter

which attenuates the audible spectrum on either or both sides
of the band of sound to be heard (the pass-band). The band-
pass filter is ordinarily used to eliminate hum, low-frequency
noise, high-frequency hiss, or distortion from a signal - or it
is used to modify the timbre of a sound by attenuating upper
and/or lower partials of that sound.

(b) The band-reject filter is inversely analogous to the band-

pass filter. The band-reject filter attenuates the audible spec-
trum between two points on that spectrum, thus permitting the
passage of sound outside the rejected portion of the audible
spectrum.

(c) The notch-filter is similar to the band-pass filter except

that its pass-band is of a fixed width (usually very narrow) and
attenuation of unwanted signals lying outside the pass-band is
very great. This type of filter is used to pick a given signal
out of many (its high selectivity gives it, typically, a band-
width of a minor-second or less) or to filter white-noise into
very narrow, almost-pitched segments.

(d) Equalizers are filters which divide the audible spectrum

into (usually) equal segments (often the octave) permitting
boosting or attenuation of segments separately. Thus a com-
prehensively designed equalizer combines all the attributes of
previously mentioned filters with flexibility and versatility.
Such equalizers are most usually used either to lessen the ef-
fects of distortion or noise, or to correct (or equalize) the
spectrum of a sound.

(e) It is also convenient to think of high-fidelity pre-amplifiers

(found in many studios) as combination equalizers and band-
pass filters. Although these pre-amplifiers are not usually so
flexible nor so effective as professional equalizers and filters,
they are especially handy when several filters are needed si-
multaneously.

Of course, these filtering devices have many varied functions

and produce a greater range of effects than I have mentioned
 257

here when usedwith other devices. For the moment, however,

only their immediate function and effect will be considered,
and more complex equipment interaction will be taken up in the
lessons on composition.

(2) Envelope control is the control both of the growth rate (or
attack) of a sound and the decay rate of that sound. The word
'envelope' is more popular than scientifically apt, and in elec-
tronic music it refers to a curve of signal (or sound) amplitude
plotted against time. Most envelope control devices have push-
buttons to turn the sound on and off and separate knobs to adjust
the growth rate and decay rate of that sound. Other envelope
control devices operate photoelectrically with moving strips of
tape or paper on which are graphically impressed appropriate
signal 'envelopes'. These envelope control devices are all
capable of producing anywhere from a very sharp staccato
sound to sounds with very slow (or long) attacks and decays.
They will control any audible signal, noise, pitches, chords,
etc. Several envelope control devices are often grouped to-
gether to facilitate the simultaneous control of several sounds.
Incidentally, synonyms for envelope control devices are: en-
velope generator, signal gate, envelope shaper, attack gate,
and many other hybrid terms. Synonyms for sound 'growth'
are: attack, onset, risetime.

(3) The term, combination devices, is admittedly vague and

its connotative distinction is probably more electronic than
acoustic. Under the term, combination devices, I group those
which either modify sounds by combining or adding them to-
gether or by producing or generating partials (harmonic or en-
harmonic) of various frequencies and amplitudes. In either
case an additive process is implied.

(a) Modifying sounds by combining them is accomplished by

mixing them in a mixer, perhaps the most indispensable item
in a studio. Most mixers provide for at least four inputs and
one output. The mixer accepts, let us say, four separate sound
sources and combines them or mixes them in variable propor-
tions - the volume of each separate source being controlled by
a knob on the mixer. The over-all volume (or level) of the re-
sulting combination is controlled by a'master' volume control.
The mixerinno way alters the separate input sounds; it simply
mixes them together. In this way, chords may be assembled
from separate sound generators, noise and pitch generators
may be combined, and so forth.

Devices which modify sounds by generating partials are (b)

258

electronic switches and (c) non-linear circuitry.

(b) The electronic switch, found only in older classic studios,

is a device with two inputs and one output. The electronic
switch selects the two inputs alternately (at a variable rate)
and simply passes the selected input through to the output. At
high switching rates, three main signals are present: input 1
alternating with input 2, and the frequency of the switching rate
itself. At the output, though, difference and combination tones
are formed producing a complex sound constituted of many in-
dividual pitches.

(c) In some cases, it may be desirable to supply a strong

enough input signal to a pre-amplifier, tape recorder, or mixer
in order to cause distortion. Such distortion is often rich in
partials and thus useful as a sound source. Distortion is due
to an electronic circuit's operation outside its 'normal' or
linear range. Such operation is termed 'non-linear'. This
phenomenon causes 'clipping' of signals and consequent increase
in the number of partials present. Care must be taken not to
damage equipment by such input overloads. Electron-tube
operated equipment is relatively immune to non-linear opera-
tion but transistorized equipment is more susceptible to dam-
age. If in doubt, experiment. The machines serve the com-
poser, not vice-versa.

(4) Reverberation varies from the 'Grand Canyon Echo' to the

'feel' of acoustics in a large room or auditorium. Mechanical-
ly-coupled transducers or tape recorders may produce the delay
or phase lag necessary for reverberation. There are two popu-
lar types of mechanical transducer: The spring-coupled(or
Hammond-type) reverberation unit, and the large-metal-plate
transducer. The spring variety has the disadvantage of insert-
ing noise into the sound. The large-metal-plate type is expen-
sive but produces a more natural reverberation than other de-
vices. Tape reverberation is inexpensive and easy. Since
tape, as it passes through a tape recorder, traverses the rec-
ord head before the playback head, it is possible to 'pick off'
some of the sound at the playback head (while the machine is
recording) and re-insert it at the record head. This type of
echo is called feedback (some of the sound is literally fed back)
and is accomplished by connecting the output of the tape recorder
to its input through some device (usually a mixer) to vary the
amount of signal fed back, and thus the amount of reverberation.

(5) The category 'tape editing techniques' includes the following

topics: (a) types of tape, (b) tape storage, (c) tape splicing and
 259

marking, (d) miscellaneous techniques.

(a) Magnetic recording tape consists of a 'backing' material,

acetate or polyester film, and an emulsion layer containing the
material (iron oxide) which is impressed by the magnetic field
of the tape recorder's recording head. Variations in both back-
ing and emulsion are important. Polyester and Mylar backings
are the strongest presently made, although these types tend to
stretch physically when placed under strain. The 'pre-tensil-
ized' varieties should always be used (equivalent to 'sanforized'
inclothing). Acetate-backed tape is not so strong as polyester
or Mylar and breaks completely without stretching. This clean-
break feature is desirable, however, in that it allows the broken
ends to be rejoined easily without any tape loss. (The stretched
portion of polyester-backed tape must be completely discarded
before the tape may be respliced, with consequent loss of re-
corded material.)

Emulsion types vary greatly. The most important aspect of

the emulsion is that it be evenly and smoothly deposited on the
backing. Variations in thickness or roughness are audible and
may increase tape recorder head wear. Tape is manufactured
in two usable thicknesses: 1 1/2 mil (.0015") and 1 mil(.001"),
corresponding, respectively, to 1200' and 1800' of tape on a
seven-inch reel. (Thinner tapes are available but their use is
not recommended because of their fragility and susceptibility
to print-through.)

Tape recording is accomplished by magnetizing an iron oxide

compound in the emulsion layer. In a recorded reel of tape,
then, the magnetic field impressed on one layer of tape may
cause its magnetic pattern to be impressed or transferred onto
an adjacent layer of tape. This phenomenon is called print-
through and is most easily avoided by physically separating
tape layers (using recording tape with a thicker backing ma-
terial).

The use of new, brand-name tape is strongly recommended.

Cutrate tape is often computer tape, manufacturers' seconds,
or pre-used broadcasting tape, and will invariably cost in head-
aches what is saved in money. These tapes are often too wide
or too narrow and have splices or drop-outs - areas where the
emulsion is damaged, thin, or absent, causing a dropping-out
of the sound.

Many companies market so-called 'low-noise' or 'high-output'

tapes. These usually combine their advertised advantage with
260

some disadvantage (perhaps only high cost). These tapes should

be investigated, their characteristics learned, and the tape
used accordingly.

Two types of leader tape are manufactured: a plastic and a

paper type. These are made of backing material only with no
emulsion layer. Plastic leader tape is used at the beginnings
of reels and between splices where silence is desired. Paper
leader is wound on the centers of seven-inch reels to give them
an effective core diameter of three inches, the reasons for
which will be explained later.

(b) Tapes should be stored at room temperature (70?F.) and

50% humidity. High temperature will cause tape to stretch and
high humidity will cause paper leader to swelland buckle. Low
humidity and/or temperature causes tape to become brittle,
with accompanying flaking of the emulsion layer or even tape
breakage.

Short lengths of tape which are used frequently in a situation

where perfect sound quality is not necessary may be hung up
'as is' or in the form of 'loops' (to be explained later). Ex-
tended lengths or sections demanding long-term storage with
minimum deterioration should be wound normally on reels,
marked and stored. Tape should not be wound on less than a
three-inch diameter hub, and seven-inch reels should have
paper leader wound on their hubs to make an effective core
diameter of three inches. This larger core diameter prevents
tape stretching and distortion which often occurs when tape is
wound on a small diameter.

Tapes should always be wound onto reels at a uniform tension.

Abnormally high tension causes warping and buckling of the
tape, whereas low tension results in a loose wind and conse-
quent bunching of the tape in either fast-forward or rewind
modes. Valuable or long-term-storage tapes should be wound
onto a reel backwards at normal operating (RUN) speed (not on
either fast mode). This results in a tape wound evenly, at a
constant tension, ready to play. In all modes of tape recorder
operation the tape should never hit the sides of the reel. If it
does, the tape will wind unevenly and 'crimp' at the edges.
This symptom indicates a damaged reel or poor tape recorder
adjustment. Tapes should be stored in a dust-free area and,
if possible, in a steel cabinet (which is non-magnetized) to pre-
vent erasure by stray magnetic fields.

(c) Items required for splicing and marking of tape are: an

 261

editing block, a grease pencil, new single-edge razor blades,

an 18" ruler, and 7/32"-wide splicing tape. Tape is spliced to
tape or leader tape in the following manner: Both sections to
be joined are placed in the editing block over the appropriate
razor-blade slot, (a 90? or 45? angle cut), and a razor blade
is inserted through both layers of tape into the slot in the block.
The razor should be inclined to the block at a 30? angle, and
some very slight sawing motion may be required. The waste
ends are removed and the two ends to be joined are abutted
(over a solid area of the splicing block). A short piece of
splicing tape (1/2" or so) is placed over the splice and pressed
onto both pieces of tape; A back-and-forthmotion with a finger-
nail seals the splice. Excessive handling, pressing, and sliding
of the tape in the vicinity of the splice should be avoided. The
block should also be kept clean and the razor-blade de-mag-
netized.

Tape is marked with cutting and other directions with grease

pencil. Marks to be preserved are covered with a length of
splicing tape, which prevents the grease pencil from being
rubbed off while retaining the legibility of the marking. Felt-
tip pens should not be used. The water-base varieties are non-
permanent and many volatile-base types dissolve tape. The
eighteen-inch ruler is used for measuring time (tape lengths
being equivalent to specific times at specified tape speeds).
Eighteen-inch rulers are handy since twelve-inch rulers are
too short to measure two seconds at 7 1/2 inches-per-second
and, although the work-table itself maybe marked off in inches,
portable rulers are, in any event, handy for use at the tape
recorders.

(d) Splicing blocks have two razor slots: one for 90? splices,
and one for 45? splices. The 45? splice is most often used,
since it provides a gradualtransition from either silence (leader
tape) to sound, or from sound to sound (since the tape head
'reads' only a very narrow strip of tape perpendicular to the
direction of tape travel at any given moment). The 90? splice
is used for abrupt attacks, although these splices sometimes
cause pops or clicks in the recorded sound (a condition which
also may be caused by magnetized razor blades). A little-
known technique for splicing iterations of the same event does
not involve splicing at all. If, for example, a repeated pitch
were needed, ideally the emulsion layer could simply be re-
moved for rests or articulation. (The tape would have been
pre-recorded with a continuous pitch or noise source.) By
using an acetone-moistened cotton swab and very rapid strokes,
the emulsion layer may easily be removed without damaging
262

the backing. Tape and razor blades should be demagnetized

before use, using a commercial bulk demagnetizer or degausser.
Once again, it is extremely important that the tape - especially
the emulsion side - be handled as infrequently as possible.

Terminology about tape 'tracks'may cause some confusion for

the novice. First of all, tape has only one 'side', the emulsion,
dull, or oxide side, which contains magnetic information. When
the operator records on the 'other side'of a tape, he is merely
reversing the direction of the tape travel and recording on an
area of the emulsion not yet recorded upon. A one-quarter
inch width of tape may be divided into either one, two, or four
tracks, depending only on what particular machine is doing the
recording. Full-track (or one-track) tape machines record on
the full 1/4" width of tape: thus they are monaural and not re-
versible. Half-track machines record on one-half of the full
1/4" width of tape: these are monaural and reversible - (i.e.,
the previously unrecorded half is made available for recording
by reversing the direction of tape travel). Two-track stereo-
phonic machines record two tracks simultaneously which are
symmetrically placed around an imaginary line down the center
of the tape. The symmetry of the track placement precludes
reversal of the tape (the tracks would erase each other). Four-
track stereophonic machines record two tracks simultaneously
which are asymmetrically placed on the 1/4" width of tape.
'Four-track' means that there are four available tracks on the
tape - two are used in one direction, and two in the other. (See
Figure 1).

(6) Some special tape recorder effects are variable speed,

multi-track recording, vibrato, and fast playback. Some tape
recorders are specially equipped for variable speed operation.
Speed is usually variable by at least a tritone on either side of
the correct operating speed. Multi-track recorders allow over-
dubbing (listening to one track of a recording while simulta-
neously adding a second track synchronized with the first).
Vibrato may be accomplished by wiggling the tape during re-
cording. Fast playback allows tape to be heard during rewind
or fast-forward - an effective tape speed of up to 250 inches-
per-second. Speed, in this mode, is variable by cautiously
(avoiding scraped knuckles) holding on to the spindle of the pay-
off reel.

D. Tape recorders comprise two basic components: (1) the

tape transport and, (2) the record-playback electronics.

(1) Typical tape transports have the following items: two reel
 203

FIGURE

i- --
4
r--T I
*- ... J
TWOTRACK
FULL-TRACK HALF-TRACK FOUR-TRACK
TWOTRACEO
STEREO STEREO

THE ARROWSINDICATE THE DIRECTION OF TAPE TRAVEL

264

spindles or platforms, tape heads, a capstan-pressure-roller

assembly, a control center, and a series of tape guides, idlers
and automatic shut-off and/or tensioning arms.

(a) The reel spindles carry the tape reels. They are referred
to as the pay-off (or supply) reel and the take-up reel. The
terminology refers to function, not position, so that the supply
reel in the RUN mode becomes the take-up reel in the REWIND
mode.

(b) The tape heads are generally located in the center of the
threading sequence and operate to erase the tape, record sound
on the tape, and play back the recorded sound. These three
functions are usually relegated to three separate heads, al-
though both record and playback facility are sometimes incor-
porated into the same head. The tape is maintained in as close
proximity (or 'wrap') to the head as possible and in careful
physical alignment. Tape heads are very delicate and should
never be touched by anything except tape or alcohol-moistened
cotton swabs. In the normal RUN mode, a given spot on the
tape will contact first the erase head, next the record head,
and last the playback head. Most head assemblies have a cover
or gate which may incorporate 'tape lifters'- small metal or
glass fingers to lift the tape away from the heads. This cover
may also have hum shields which cover the heads when the gate
is closed.

(c) The capstan-pressure-roller assembly consists of a ro-

tating metal rod (the capstan) and a rubber idler or tire (the
pressure or pinch-roller). In the RUN mode, the rubber idler
pinches the tape against the rotating metal rod thus drawing
the tape through the transport at a speed equal to the linear
velocity of the capstan. Changing the rotational speed of the
capstan changes the speed of the tape recorder. Most tape
recorders have two or three possible speeds out of the follow-
ing: 1-7/8 ips, 3-3/4 ips, 7-1/2 ips, 15 ips, or 30 ips.

(d) The control center contains knobs and switches for the fol-
lowing functions: STOP, RUN, RECORD, REWIND, FAST
FORWARD, TAPE SPEED SELECT, REEL SIZE, and EDIT.
(Note: Some of these controls may be located on the electron-
ics assembly.) In the STOP mode, brakes are applied to the
reel spindles and the pressure roller is not pinching the tape
against the capstan. In the RUN mode, brakes are removed
from the reel spindles, low power is applied to the reel spin-
dles to supply take-up and pay-off tension, and the tape is
pinched between the capstan and the pressure roller. The
 265

RECORD mode is identical to RUN except that the recording

circuitry is energized. REWIND supplies full power to the
rewind take-up reel spindle and disengages the pressure rollers
from the capstan. FAST FORWARD is the same as REWIND
inthe opposite direction. TAPE SPEED selects the appropriate
RUN (or RECORD) speed. REEL SIZE switches control pay-off
and take-up tensions and should be placed in a position corres-
ponding to the size of reel being used. The EDIT control on
most machines simply places the tape in contact with the heads
regardless of the mode of operation.

(e) Tape guides maintain mechanical alignment of the tape in

its designated threading path. Idlers are high-inertia rotating
tape guides, which function as low-pass filters, damping out
jerks and bumps in the linear motion of the tape, and preventing
wow and flutter(respectively, low and high frequency variations
in tape speed). Automatic shut-off and/or tensioning arms are
guides on spring-loaded arms which stop the tape recorder
should the tape break or come to the end of a reel, and help
maintain even tape tension in the threading path.

(2) Typical record-playback electronics have the following

items: input jacks, record level control, playback level con-
trol, record level indicator, output jacks, and an A-B switch.
(Other controls which are found on record-playback assemblies
vary greatly from machine to machine, and will be omitted
he re. )

Input jacks are usually supplied for a microphone (or low-level)

input and a 'line' (or high-level) input. The record level con-
trols the gain of the signal being recorded. The record level
indicator is usually a meter calibrated in decibels, although
'magic-eye' indicators may be encountered. With the magic-
eye indicator, maximum recording level is shown by a meeting
of two green or blue shadows in the tube itself. The meter or
'VU Meter' as it is called, is most often calibrated from -20
to +3 decibels. Maximum recording level is usually 0 db., al-
though professional machines may easily record peaks in the
+3 to +4 db. range without distortion. Output jacks are so-
called 'line' outputs suitable for studio purposes. An A-B
switch switches either the recording circuitry (A) or the play-
back circuitry (B) into the output jack. (The VU meter is usu-
ally switched to monitor A or B as well.) This switch may
usually, depending on the machine, be operated when the ma-
chine is in the record mode with no adverse effect on the sound
being recorded. It is useful in determining, both aurally and
visually, the difference between a signal before it is recorded
266

and after it has been impressed on the tape. In the B position,

the output comes directly from the playback head (while the
machine is still recording), and again fed through to the output
jack. In an ideal recording situation, the difference between
A and B should only be the very short time it takes a spot on
the tape to move from the record to the playback head.

III. Preliminary Exercises in Composition

By now fluent in the operation, interconnection, and sounds of

all basic studio equipment, the student will want to understand
tape handling and splicing techniques, so as to prepare himself
for some simple exercises in writing (or realizing) one-line
tunes.

A score of a proposed one-line exercise should be prepared

with complete specification of dynamics, phrasing, articulation,
timbre, durations, and pitch. These parameters will be trans-
lated into their electronic music equivalents, and the results
tape recorded. Doing several such preliminary exercises will
bring skill in studio operation and splicing techniques. A model
for such an exercise could be drawn from the standard literature
or an attempt might be made to duplicate a single line from an
existing electronic piece. Imitation of an electronic piece will
refine the critical ear necessary for the duplication of the most
subtle nuances of timbre, articulation and so forth. This latter
type of exercise will also force the student to develop his own
personal shorthand for notation of electronic music.

Let us first take the example shown in Figure 2 (Mozart: Piano

Sonata in A, K. 331, Variation II) and trace its imitation by
electronic means. This example was chosen because it pre-
sents many different durational values and articulation types,
employing only seven pitches. In this example, pitch and tim-
bre are the only variables the student need take into account
for initial recording. (Dynamic variation will be performed in
the final taping, and phrasing, articulation and durational values
will be determined in the splicing process.) Initial raw ma-
terial should be recorded at as high a level as possible, without
introducing distortion, so that the signal-to-noise ratio is also
as high as possible. Dynamic variation, then, should be per-
formed as late as possible in the realization of a piece so that
normal deterioration of the recorded sound (which would lower
the effective signal-to-noise ratio) will not affect signals re-
corded at a low recording level.
 267

FIGURE

Attw*4 . -- ._
' I
'
_i--
---t f r L- - -

LI
I i
w.
...
~
-eI -- I
,1 '-"-1__.- Ti
'-Fi
m
I.7 I _mw_
I - F
Id
~ ) !etc.
First record four or five seconds of each pitch in the piece:
five seconds of G#, A, B, C#, D, E, F$. The tones must be
recorded with the desired timbre, without distortion or noise,
and at the highest permissible recording level. After recording
the pitches (they willhave to be tuned to a piano or other pitch
standard), each one should be separated from the roll of tape
and stuck to a piece of cardboard hung up near the work area.
Be sure that each piece of tape contains one and only one pitch
and that it contains no silences. The next step is to splice to-
gether bits and pieces of different pitches in a sequence corres-
ponding to the tune. The splicing process moves from left to
right. Keep a roll of plastic leader to the left of the splicing
block, and the ruler in front of the block, as shown in Figure 3.

If J = 30 (or two seconds) and the tape speed is 7 1/2 ips,

then J- also equals fifteen inches of tape. In that case, each
oh corresponds to five inches of tape travel. Ignoring the
staccato indications for the moment, the first phrase would be
5" of C#, another 5" of C#, then four . 833"lengths of alternat-
ing E and D followed by . 833" of C# and . 833" of D, assuming
that the trill is played

The next E would again be five inches, then 5" rest, then four
1 1/4" lengths of F#, E, D, C#, to complete bar 1. (The sym-
bol " here, always means inches.) A student shouldtry several
versions of bar 1 to determine proper modes of articulation.
For the opening staccato C 's, assemble the following: C# with
a 90? splice attack 3"long/ 2" leader tape/ Ct with a 900 splice
attack 3 1/2" long/ 1 1/2" leader/ then E with a 45? splice
.833" long/ D with a 45? splice attack . 833" long/ E, D, C#,
D, simile/ E with a 45? splice attack 5 1/ 2" long/ 4 1/2" leader/
F# with a 90? splice attack 1 1/4" long/ E, D, C#, simile//.
This adds up to 30"-- the tape length duration of bar 1 (assuming
J. = 30 and a tape speed of 7 1/2 ips). The exercise may be
redone at 15 ips, as well. All lengths are then doubled, the
effect of minor cutting errors thereby minimized. The spliced
version may then be re-recorded and appropriate dynamics
added.

Dissatisfaction with the musical results will, most likely, be

traceable to: out-of-tune-ness, note envelope, or phrasing.
Out-of-tune-ness can only be adjusted by re- recordingthe basic
 209

FIGURE

SCOREON -. -
STAND
EDITING BLOCK // PENCILS, RAZORS, ETC.

LEADERTAPE-- TAKE-UPREEL
R IL I ii nl I

RULER
270

pitchmaterial. Note envelope, here, is only determined by the

45? or 90? splice. To obtain more intricate envelopes, the
student may want to record each separate note of the tune using
the envelope control device. Phrasing is adjusted by varying
the amount of leader tape between pitches. Note that phrasing
may also be determined to an extent by correspondences of en-
velope and, in certain situations, this may even over-ride dif-
ferences produced by the insertion of blank leader tape. The
student may also wish to make the example more interesting
by buttressing the phrase structure with timbral variation
brought about by filtering. Satisfactory musical results should
be obtained before abandoning this exercise since it involves
intricate realization problems despite a surface simplicity.

Special equipment associated with semi-automated studio syn-

thesizers facilitate experimentation in these areas. Many of
them provide for real-time keyboard playing of a phrase while
the desired results are obtained by turning knobs and dials -
again in real-time. This short-circuits, to some extent,
a time-consuming tape recording-editing-playback-criticism
route. The student should not, however, become unduly in-
fatuated by the ease with which pretty sounds are obtained from
semi-automated studios. Prettiness enhances a good composi-
tion but, all too often, disguises a lack of substance. The ac-
complishment of several exercises from the literature should
prelude the imitation of an electronic piece. (The Bach French
and English keyboard suites offer an abundance of good tunes
for electronic realization.)

The opening from "Events" by Mel Powell (CRI: S-227) con-

tains an excellent passage for imitation by the student. The
passage runs from 32. 7 seconds to 35.4 seconds in the piece,
and may be approximately transcribed as in Figure 4. Plus
and minus signs indicate sharp and flat pitches. Following the
method used withthe Mozart example, the student should record
each pitch, splice the entire tune together, and compare his
version to the original on the record. To facilitate his work,
the student should record the Powell excerpt on tape and play
it back at half-speed to reveal detail. The piece may also be
imitated at half-speed to assure a more faithful copy. The
primary difficulty to be encountered in effecting an exact dup-
licate of this example will be registral ambiguity of the pitches
due to timbre. Exact filtration of the pitches will be necessary
to place each in its apparent octave.
 271

FIGURE

J=104
(+) (+) (*-)
^ I b(- .?7 (-) SA

.
~t~~~.~ ~i
'IM:4
_~.'~.~ |I.|~i~~- I etc.
272

IV. Compositional Techniques

The notation sample shown in Figure 5 is a somewhat modified

version of an excellent system originally devised by Bilent
Arel. Using this example as a guide, if he wishes, the student
should prepare a very brief contrapuntal piece (not exceeding
about 15 seconds in length) incorporating pitch, noise, differ-
ent timbres, a full scale of short and long durational values,
reverberation, and different envelopes. Realization of Figure
5 will be accomplished by playing different lines of the exercise
on different tape recorders, synchronizing and mixing them,
so the exercise must first be divided into the minimum number
of separate lines necessary for full realization. Fewer lines
means using fewer tape recorders for the final mix, and since
tape recorders running simultaneously do not remain synchro-
nized for very long (30 to 45 seconds, maximum) fewer lines
also means more exact synchronization.

Since the pitch material is a two-part canon (the second voice

in diminution beginning at 4.0"), two lines are necessary at
least. The noise material requires only one line. Note that
the noise material preceding 4.0" may be placed in the same
line with the second voice of the canon. This would allow the
noise material starting at 4.0" to be started (on a separate
tape recorder) synchronously with the second voice of the canon,
thereby increasing the chances of good synchronicity between
the two (as required by the score). On the other hand, con-
sidering the advisability of being able to start both of these
lines at the same time, a decision might be made in favor of
four lines: the first voice of the canon, the second voice of the
canon, noise material before 4.0", and noise material after
4.0". Since reverberation is only to be added to the pitch ma-
terial, the reverberation device will have to be connected so
as not to affect the noise material. The over-all dynamics
should be added when the final mix is copied, not during the
final mix itself. It will be necessary to operate three or four
tape recorders, the reverberation device, and a stop-watch at
the same time. Do not attempt to control the dynamics as well
unless the necessary controls are conveniently placed (although
balance between lines will certainly have to be determined at
this stage). The pitch material of this example may either be
single isolated pitches or chords, and the student should for-
mulate and record what he deems preferable. The envelope
may be supplied by the envelope generator or by splicing. For
fast passages, splicing is usually preferable inasmuch as a
complicated envelope for a signal under 1/4 second or so dur-
ation is not usually relevant (although this certainly depends on
274

context). The signals should be spliced together at their ap-

propriate durational intervals, taking care to measure the
over-all length of the passage from time-to-time (which shows
up the accumulation of small splicing errors). The noise may
be spliced exclusively at 900 (butt-spliced), since noise spliced
at 45? lacks a sharp enough attack to discriminate it from other
material. The noise component itself must be filtered by a
band-pass or notch filter to assure registral discriminability.
The final mix should be made and copied. For this final mix,
the separate lines are placed on separate tape recorders, cued
to be in synchronism, and each started at its appropriate in-
terval. Rehearsal of this mix will be necessary since individual
tape recorder characteristics vary considerably, and several
trys will be necessary to effect optimum synchronicity. Each
rehearsal, though, should be performed and recorded as though
it were the final mix.

Another technique applicable to this and other exercises is the

use of loops. A loop is a section of recorded material whose
end is joined to its beginning by a length of leader tape. It is
threaded normally on the tape recorder, except that idler arms
and pulleys are substituted for the reels. In this way, the loop
will play over and over continuously without constant rewinding
and threading. The only precaution one should observe with
loops is storage and marking. Loops tend to curl and crimp
when hung loose - for which condition there are no remedies.
Loops must be clearly labelled for contents, tape speed, and
direction of travel.

This example (Figure 5) allows for a great deal of experimenta-

tion with different generators, envelopes, filters, and special
effects. The student should exercise his imagination and bring
into play as many variables as possible if only in order to gain
familiarity with the equipment.

Conclusion

Working in large dimensions precludes the use of exact, pre-

realization working scores. A student will begin to learn from
his interaction with the equipment how to maintain an efficient
working level. Occasionally, favorable accidents occur which
change the course of a piece. A student should encourage within
himself a flexible enough outlook to encompass and employ
everything he hears - whether it be accident, or the exact ful-
fillment of a desired compositional aim. This attitude, of
course, requires not only unceasing critical listening but im-
 275

mediate and careful action upon the medium. As in all com-

position, the composer-material interaction is two-way. A
student will therefore come to a notion of electronic music
which combines the best criticism (the ear of the composer)
with the advantage of immediate and private hearability (the
tape) with possibilities for instant correctional response. The
student will most naturally find himself working back and forth
from pre-realization sketches and scores to real-time impro-
visation on tape. It is this mode of studio operation which,
though requiring a great deal of preliminary study and practice,
results ultimately in the most productive and satisfying work
pattern.