Roland System 100M Synthesizer Musical Equipment User Manual

Roland System 100M Synthesizer Musical Equipment User Manual

Below you will find brief information for System 100M Synthesizer. The System 100M is a compact and economical modular synthesizer with all the potential of larger professional systems like the Roland System 700. The System 100M is modular, so it can be built up into configurations to match the particular needs of each individual musician.

advertisement

Assistant Bot

Need help? Our chatbot has already read the manual and is ready to assist you. Feel free to ask any questions about the device, but providing details will make the conversation more productive.

Roland System 100M Synthesizer User Manual | Manualzz
PRACTICAL SYNTHESIS
FOR
ELECTRONIC MUSIC
VOLUME ONE
2ND EDITION
Roland
Contents iii
Contents
List of’ lllustrations and Patch Diagrams «cove «os + 44 00/0/06 a ue as iv
Introduction
Chapter One: The Basic Patches
1-1 OAC ON i fs Sst hau n sn ac ls sana a armen | EOS EN E A MC E aure 1
1-2 The Basic Pateh:for Subtractive Synthesis . . + à à x «5 sts vas o os aro wemsTh os sa LET LT 1
1-3 GCorrtrol er Tone Color E: : 5 2 à a vase ba Ds ts amu ss Eade Mali aan es 7
1-4 The Basic Patch for Additive Synthesis
Chapter Two: Expansion of the Basic Patches
2-1 IMPTOUUENÓN E ei". 2 e a ee Za El mee: AENA A ATA AN RA cd 19
2-2 Fregueney NICdUlaon === == numa qee 34 €: a MANE ES aa EE ai 21
2-3 Tone Golor MOCUIGtION : : == == 25.5 == y € - - e » . - minvene, ee ere e RU ARE] 250 mirado ve 28
2-4 Amplitude Modulation ..........e_—oéeé.emeroredeadronmoeneoerarvoedaceneameraana 37
2-5 Tre MingNOodUlaton > =s = 3 ne. EA E ой ол ues e e mee CUR TIED A a ¡AUN 1 39
Chapter Three: Advanced Synthesis
3-1 Nibxars:and: Fixed" VoltagesouUnoesare: TA. is bbs fra a wen ста 47
3-2 The Sample and Hold ee a Emo EE ace 59
3-3 The Nose: Generator » 50 à ÿ € 28 à tam ne € 8 € 6 à à 8 8 SES à ar 0 € € € à à 4 VAR à à à à « à aNdinve 63
3-4 The Gate Delay and Pulse Shaper.... 1.122220 001 LL LL aa LL LR a LL La AL 71
3-5 CONTO er mum Y ER SUELE TONES Y A UE Y E @ EA à NIE E 3 Y 5 5 & 4 x a CER A VE RE vr: 73
Chapter Four: Accessories Used in Synthesis
4-1 INTrOQUCON: Y puenae » + e € « A € @ 9 taeuate: € 5 a & o в нов 6 Damme E & © 56 5 & 5 § WEEE des €» 5 e % Wk rose 79
4-2 The Grape: Equalizer ; « > 3 a <a zas ya a Es 466 a quae E 6 Ne a Ella a E E AE A 79
4-3 Te VOCOUEF - muro 25 aseos a ETA LE SA IEEE к вов м вов ол RR DUAL LAN E oats Sila 83
4-4 ENase SNITIers. and: PIGRgero:s = « s ve Sa E aan Fol die HE E55 555 BATES 6 5 sus 55 6 & ios Bin 86
4-5 Chorus aNQ:ECRO: iv vi 5 vo 5 3 7 io y o EN € € © 0 ie le 8 REED! € 8 El E © 5 & & 3 4 IEEE: el Ea & 4 Ad $ “ela 16778 99
INGENIO: TONTA ==: a To EEE: в пп мессы Исаково e mua вова в а ООО Ep 103
Mecujes:or the Roland System: T00M ; i cvs visio sv sas sain msm бойко сво во сое: facing p. 104
List of illustrations and Patch Diagrams
1-1
1-2
1-3
1-4
1-5
1-6
1-7
1-8
1-9
1-10
1-11
1-12
1-13
1-14
1-15
2-1
2-2
2-3
2-4
2-5
2-6
2-7
2-8
2-9
2-10
2-11
2-12
2-13
2-14
2-15
2-16
2-17
2-18
2-19
Basic Patch for Subfractive SYNtDESiS “ever € à « à à = coul! INT US LE 2
ENVOIODO ss 1 AT Ss 5s 35 EAE Se CEES IBAA ANT Ra HET USE 3
Basic Patch using other moduleS = : à à anavene oro € à à # & à Gates à 6 SE ZE SN A BURGA E € à à € À 56 4
simple RECORDER susaccasatsiiqumettatsisuiageuss sr :sapnmadoraas suis mad 5
Effects of the envelope on a SOUNd | 0200 124 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 eee 7
HORN (also TRUMPET, TROMBONE, TUBA) ...... ci iii 1e a 2 ee a 4 a a ee a 00 8
RECORDER, CELESTA, XYLOPHONE .......200 222022 04 4 4 4 a 4 4 ee ee ee eee ae 10
CLARINET (aiso BASS GLARINET) + à à & € € à 8 5 saiainad oa vos 2 so 0 ease siete BTE à © 11
ENGLISH HORN (also OBOE) ..... 22121122 LL LL 2 4 4 A 4 4 4 4 A A 4 a 4 4 4 4 4 4 a a a 4 ea 40 12
BASSOON ii EEE EEE KERLE DENT, OR ES a UE 13
Additive Synthesis Block Diagram . . . . oi itt tt et ee ee ee ee ee ee eee 14
GHIME (Adaitive SYnithesis) ; ; y -- <=. as:arisasa aaa: TA as: 15
Variation of CHIME Patch .. 1.220020 0004 44 A 4 4 4 4 4 A 4 A 4 4 4 4 4 4 4 4 4 4 4 4 A A 4 4 a a 0 16
CLARINET (Additive Synthesis) сс с cou nsueuuuacamaesauacssiTaa Poe t esas 17
spectrum for the Clarinet..........024.00000 00000000 m0 ee 0 0 00 00 000000 ce 00 0000 18
Block diagram Tor synihesizar module are à à à à @ à à 0 amena sa ОО с. 19
VIOLIN with Delayed Vibrato (also VIOLA, ‘'CELLO, STRING BASS) .................. 21
Delayen VB =.=.=L= E E e 2 3 75 e seen a ea ey an rn O as 23
MIN p:2a3ranamLs NESIICCIACTADE COREA IASE ICAA NERAL tes EN Zi 24
Frequency Modulation (bY YOO) ... a.=... с « 6 € « à Tee ae a 8 6 a 0000 0 med ms Eu ea ua m0 26
DOG: WHISTLE: ca 5 25 75557 AE E EEE IC ESE RAS UA E Y E na 27
FLUTE (also PICCOLO) .............esmemeearererdemweccefdooQdvroearedorre rara 28
GLARINET WVEF modulation: by МСО) 7: 3 > == 7 5 = 5 4 "amena à © à à € € à à à § #5606 N EN ë 29
Square and Pulse Wave .........oc.eoedodooroneo ro nenonooonoananemoroamoaa aaa 32
BOWED STRING BASS . i ios 5 «6 varia à à à à à à à € à à éarique à à à + ÿ à € à 6 is laine: of 8 as с сана 34
STRING ENSEMBLE . . i tt ee tt te te ee ee tt ee ee ease eee 0 36
Tremolo: EXpaerient= : == = ques 5 6 ratevara! al & 6 & © à 6 à Ta © BEEN of & & & à $ N NEN SG due A e 37
Tremolo: Wavetormi8 . . ..- » - - « ee... e * “2 Bmendid e Ней опа и ee e UNEN 37
Adding tremolo to a patch ........omeredodoade donado dooaradonane naa 4 4 4 14 4 0 38
Fina Modulator VWavero! Ms : 5 5 niece diiisiv 6 85 5ub - 6 A MAS NULA IRA EAN 6 55.5 5 sors 5 5 39
GÉOCRENSPIEL cavas a 50 TZ ER sn arena sr 6 05 a 65 6 8 Soares à à à à « 7 57 а raters hh & 8 00% 24 5 5 12 40
30 25a ET i EE IRE AST En Ee BRE Ae EE LOS EE CEET 42
BELL IGSING SINS WANE à 5 a 7 à à à à sven o 05 8 55 5 6 GHAR & 6 à © 9 € à © OMR SS EEN E UA Ni 44
YAOLIN: BOWED TREMOLO: ; ss sicvis i io assnidnbminn sini ssh mbnoddn ios a s5iaa 46
3-1 Model 132 Audio Signal/Control Voltage Mixer .......e.oxoreeresrereredrseredaoerare, 47
3-2 Adding Waveforms .......Q_.oéceseroecrercoerdorooooanvoa da rare dor arnoroaredreo 48
3-3 ТОО ПОПСА cronies wuss snk ilies CHS SE REUS EEE ELA Ea Es 50
3-4 BANJO (Repeating Trigger) .......r.c..oo2erenreoodroerrreveroeeordoarerdoroaorroaroao 52
3-5 WHISTLING GIA) ov df css ss ss ale EEE GE EEE GN Graeme see es 54
3-6 WHISTLING ©... tt te tt ee et tt tt ett te ee ee 0 8 4 8 0 8 4 10 a 00 55
3-7 Delaved Vibrato (IMproved) : : = à à % à avaveuel & à à € à € à à à à dE YE 6 6 6% % SUERTE 6 WE VE 3 56
3-8 Trill (Improved) ... 1.120200 24 44 4 44 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 0 4 0 4 4 0 4 вона нина a a 0 0 57
3-9 Natural TUU à à à nme ss oo 6 5 5% 6 ©RGIEE 86s E55 5% 5 SBQIGHR © 8 so § 5 55 а Srv § & 6s 6 A 58
3-10 Sample and Hold Patch (Arpeggio-like Patterns) . . .. .. . cc viii iii te et ee ee ee ee ena 59
3-11 Sample and Hold Block Diagralfh :: . == | == 4 % a 4-5 e VUE E dE EA & Glamerd o o 6 o & & % 0 60
3-12 Sample and Hold Waveforms.......o...oesreseroororoeorrnrenodre nono ororo rara 61
3-13: Sample and Hold Patterns ::s - "ue a a € € € à à & 1 Grass’ of € WERE A E ERES EAS 62
3-14 Sample and Hold Output Lag ........c.eceeeeerrorocoreeooeroerorendoorerorarora 63
3-15 WOOD BLOCKS (Noise Generator) «oii à à à + «== n amar Aa Eee dd MENO E AA AA 64
3-16 Random Tone Color Modulation ..........e.oesrereororerorererraoroneoaa aora. 65
3-17 GATE + TRIG for Sample and Hold Clock Input === .= as 63577 ese EEE ele. 66
3-18: WIND. - = e e e lmue: a e e e 8 e Y» ¡e leneelo; e € Y € A | A Y 9 NE EE ET e № mad ES ERA IAS 67
3-19. WIND IN Stergo! : s à « à x à поз TREE à à 6 6 ow à « à à Gael &# & à © © à à à à 6 SUSHEIE #9 66 à = à à à 5 5 8 68
3:20 SURF... 0010000 me ne na 0 m0 mere mana ee mm CSSS Rama SERRES STC RTS 69
321 SURP 1ImStereo à à à à à à © à à à 7 revente, E EUA AN AU E EA EAS DENIA EEE € E e Y Y e de Y SE 70
3-22 Gate Delay portion of Model 172 module ..........o_eoooderssranoaoaorarerareceo 71
3-23. WOLFE WHISTLE (Gate Delay) .......... «5 6 oo 00 6 orsiaiois soso asmsisnioesesessssesesis 71
3-24 Gate Delay/Pulse Shaper WaveToriNs « 5: sc cv vn si aoe s sss 855 558 8% @elsns » ¢ 56% oi as 72
3-25 ¡Render Layer. «La «Ls Avalos ma a a Ma EOS EX YY E 0 0 ANGEL BS 5 KE Bmw e de Veis 73
3-20: JOY SUEK GoNtrollen + : Ina baaa 2 3 4 TMD à à à à à € à à à à SET Ed à el E 6 5 à à à 8 ae 74
3-27 Three-Dimensional Joy Stick Controller . . . . . ii iii tt te te te eee. 74
3-28 Using d Foot Volume Pedal asa: controller : ; « « uu vaso vv vw 5 5 6 5 eaves: & se & 5 508 5 la CANTAN 75
3-29 Foot Control of Dynamics: . . ic ee oo on о в т о» мч 0 esos sssssses ss ses sesssssinicaiiosmss 76
3:30 Analog Sedguenecer Panel Diagram... = === a 4 6 à arenes: of 6 в € à 5 © 6 SUBSE 6 8 € à € © HB se EERIE 77
3-31 Analog Sequencer Simplifiel Block Diagram .....220 020204 44 24 4 4 4 0 0 4 0 4 eee eee 77
3-33 Roland MC-4 MicroComposer (Computer Controller) ...........e:eeeeorooorore roer, 78
4-1 Graphic Equaálizers. . ......... .1s:eneer oi ao oo as (e (een; e y aa aaa e be A 6 RES S 65 5 5 m0 V0 80
4-2 Graphic Equalizer Blockc Diagratfi -. == <= == = eee 0 à à à à à à meet à «à à 6 à à à à ocean 81
4-3 Equalizer COnnectiONS … … . > ave sm eu nm 0 em e e... ESTA KETTE GG ee 81
4-4 Tvpical Equalizer Settings for String SOUNCS « « wa sis € в 6 6 0 wee oo 6 5 eve e ee wise ss 82
4-5 Vocoder Block Diagra . . .-..e..... e. . 5 9 0 0 5 i9 8 wuswmd, 6 8 6 EE FEB IBF SRG 605 806s BEES s 84
4-6 Roland SYC-350.Studio: E'VYpe VOcoder”: + == = ==... "a 6 63666 erez € Y € €» нод анонсов но о 85
4-7 Block Diagram for Phase Shifter ot FIBNGRr . . cv vss sss sem sn dois soos ssss sna enioesds 86
4-8 Roland System 700 Phase Shifter and Audio Delay .........2.042004 44024 0 0 0 0 0 0 0 00000 0 87
4-9 Roland System 100M Phase Shifter and Audio Delay ........ow.eoeereoosdoooorororoerondes 88
4-10 Phase Shifter Frequency Response ......ríX.o.coeroesedorore da reda ae donoonoaarrorao 90
4-11 Flanger (Audio Delay) Frequency Response .......W..eñ.Qeooocoseoosodcdosoodcooreoooorna 91
4-12 SOLO VIOLIN (Audio Delay). ....... r..e..coeserorodoerreroeroorarodoarroarereo 93
4:13 SOLOVIOLIN with “Bowing Effet . . «cc iwiviv eos 65555 swears ess eos ssn emeae ss se 96
4-14 Experimental Violin Envelope . . . . oot tit te ee ee ee ee ee ee eee ee eee a 96
4:15 Complete SOLO VIOLIN Sound ©. : : : y aaa a a ye © 5 & 5 à NTC of в & § à % $ à à eras © À 6 98
4-16 Chorus and Echo Machines ........ir.oÑsseeseroeoroa remo rare moardaara rsarerera 100
4-17 Block Diagram:for Tape Eeno' : = : = = 5 een adan a NEUE LEE TEME EE E E 101
4-18 Using an Echo Machine .........o-.eereerresredra moda rererererdraredsrrdrrara 102
ато 1 ааа оао ото по ВАН
Introduction
The purpose of Volume 1 of Practical Synthesis for Electronic
Music is to provide practical applications of the theory of
synthesis for producing useful sounds on a voltage controlled
synthesizer. The theory itself can be found in A Foundation
for Electronic Music (published by Roland), but this is not
essential for comprehension of the material presented here.
Generally, terms and expressions which are not explained in
the text can be understood from context, and/or from
examples given later,
All voltage controlled synthesizers are basically alike but for
the purpose of presenting experiments in the production of
specific sounds it is necessary to relate to a specific system
to show control settings and their variations. For this
purpose we have chosen the Roland System 100M Synthesizer
because it is small, compact, and economical, yet it is a
modular system with all the potential of much larger
professional systems such as the Roland System 700.
The System 100M, being modular, can be built up into con-
figurations to match the particular needs of each individual
musician. The fold-out at the back of this volume shows
panel diagrams of many of the modules which are available.
In this book, to make it easier to read patch diagrams, we
will show only those portions of the modules which are
actually used in each patch. For example, when we want to
show a VCO, we will probably show only half of a Model
112 Dual VCO Module rather than the whole module or the
Model 110 VCO/VCF/VCA Module. Except for minor
points, these VCO's are the same.
In many places specific values or standards are mentioned. It
should be remembered that these values and standards will be
common to many other synthesizers, but not necessarily to
all synthesizers. Most of the diagrams and settings shown in
this book are, of course, adaptable to other synthesizers. It is
usually only a matter of comparing the System 100M with
the other synthesizer, particularly in regard to how much a
control movement affects the sound.
| wish to express grateful appreciation to my fellow workers
at Roland without whose help and patience this book would
have been impossible.
Robin Donald Graham
Synthesizer Project Manager
Roland Corporation
Osaka, Japan
September, 1979
Introduction vii
Chapter One:
The Basic Patches
1-1
1-2
Introduction
In electronic Music there are three basic approaches to
synthesizing sounds:
1. Subtractive Synthesis
2. Additive Synthesis
3. Direct Synthesis
Subtractive Synthesis is by far the most common form of
synthesis used with the voltage controlled synthesizer. In
subtractive synthesis we take a waveform (sound) which is
rich in harmonics and filter it to produce sound with the
desired harmonic content.
Additive Synthesis involves the controlling and mixing of
sine waves of various frequencies and amounts to produce
sound with the desired harmonic content. Additive synthesis
is used to a certain extent with voltage controlled synthesizers,
but true additive synthesis (the addition of large numbers of
sine waves) is less common due to its complexity.
Direct Synthesis involves the generation of waveforms using
data is stored in a computer memory. This form of synthesis
is highly complex and requires extensive knowledge of higher
mathematics and sophisticated computer programming.
Nevertheless, it is possible, to a certain extent, to use the
basic principle with more highly-developed voltage controlled
synthesizer systems. For example, instead of using a sequencer
for generating a series of voltages representing a melodic
pattern, these voltages could be used for controlling a VCA
to produce complex envelope patterns not possible with
normal envelope generators. At any rate, the subject of direct
synthesis is far beyond the scope of this book,
The Basic Patch for Subtractive Synthesis
Fig. 1-1 (a) shows the block diagram for the basic synthesizer
patch used in subtractive synthesis. To review how this patch
works, the keyboard controller produces two outputs: a
control voltage and a gate pulse. The level of the control
voltage will correspond to the last key pressed. This control
voltage output is most often used to control the frequency of
a Voltage Controlled Oscillator (VCO), thus, when a key is
pressed, the VCO will produce the pitch which corresponds
to that key. The output of the VCO is a waveform (sound)
rich in harmonics. The Voltage Controlled Filter (VCF) is
used to filter and/or accent certain harmonics to give the
final sound the desired tone color. The second output of the
keyboard controller is a gate pulse which is a high level
voltage (+10v) that appears at the GATE output whenever
any key on the keyboard is in the depressed position, The
gate pulse is most often used to trigger the envelope generator
(also called ADSR for the control names) into operation. The
control voltage output of the envelope generator “opens”
the Voltage Controlled Amplifier (VCA) to let the wave-
form out so that when connected to an amplifier/speaker
system, the synthesizer produces sound each time a key is
pressed.
The Basic Patches 2
Fig. 1-1 Basic Patch for Subtractive Synthesis
(@ Block Diagram
vco >| VCF - > > OUT
A
Pitch
Cv ADSR
/
Сате
Keyboard
controller
® Patch Diagram (See alternate patch, Fig. 1-3)
Grad _ Patch cord
me OUT
mu Wem = — Internal connection
Pitch —>
control
input
ADSR E
Ё MANUAL GATE os ou В
2" DEGAY aL RELEASE im
: Pitch CV © TIME | TIME LEVEL TIME и Envelope
® :
В Gate El @-
À — To produce sound, raise this to “10”. |
A сов BE Rolend ||
STINCIA- 10004 81 a _ — PE
Keyboard controller
3 The Basic Patches
If a piano key is struck and held, the sound of the piano will
jump up quickly to maximum loudness, then gradually die
away. This loudness pattern is called the envelope of the (@ Piano envelope
sound (see Fig. 1-2). The time required for the sound to
jump up to maximum is called attack time. The time required
for the sound to die away is called decay time. If the key is
released before the sound has died away, the sound will be
dampened and die away very quickly. This is called release
time. With instruments such as organs, it is possible to hold
a key down and produce sound indefintely. This element of
Fig. 1-2 Envelopes
Loudness
the envelope is called sustain level. Most envelope generators La
have at least four controls, one for each of these four parts of
the envelope, but some are designed with fewer controls for ® Piano envelope (Dampened)
producing simpler envelopes. Key
struck
Pressing a key on the keyboard produces a gate pulse which Key
triggers the beginning of attack time. When the sound reaches released A = Attack time
maximum loudness at the end of attack time, decay time D = Decay time
; ; R = Release time
starts and the sound will fall to the level determined by the
SUSTAIN control. Note that if the SUSTAIN control is at
maximum, the sound cannot fall, thus the DECAY control
will have no effect. The sustain level will be held as long as
the key remains depressed. Releasing the key cuts off the LA D R
gate pulse which starts release time so that the sound dies
away.
A
(©) Organ (Synthesized)
K
ón és S = Sustain level
Fig. 1-1 (b) shows how the basic synthesizer patch is actually
set up on the synthesizer. Fig. 1-3 shows how the same patch
can be set up with different modules. Note that with this
synthesizer, as with many synthesizers, the pitch control
voltage and gate pulse connections are made internally and
patching them is not necessary. As shown, these patches will
Ц, в
(d Synthesizer (Brass sound)
produce a clicking sound. This is because all of the envelope
generator controls are at “0”. Pressing a key triggers attack
time which rises very quickly to maximum, after which
, . ‚ Кеу Кеу
decay is triggered and the sound falls very quickly to ON OFF
minimum. With the controls at “0”, this rise and fall is so \_
quick (on the order of a few milliseconds!) that all we hear is
a click. |
$
Сате
pulse |
ON OFF
1. 1 millisecond = 0.001 second
The Basic Patches 4
Fig. 1-3 Basic Patch Using Other Modules
VCO VCF SA
——"
7
ттт O -
FVCA DUT,
—
INITIAL GAIN
:
E
lev
5
— OUT
e
a
К
BE
=
=
ha
E
E
ML
8
af
Envelope:
HARZ NET
Electronically, this patch is exactly the same as the diagram shown in Fig. 1-1 ®.
коб em =
ati pris ne ne
ii оон
A name A
1
ADSR
НН овен,
| system: 100M 1400
MANUAL ok
ATT ee SUSTAIN HELE
LEVEL
® Hi
tent,
ENV t Out
ME
SZENEN
MANUAL О Е
ATTACH DECAY SUS JAN BELE)
ME TIME LEV
ENV. 2 OUT
ve
re To produce
sound, raise
= i E this to “10”.
al
«св
FREQ FREQ DELAY ще FORM OUTPUT
RANGE LEVEL
KYBD
TRIG МЕ
я {Le e IN
MEDOUTÍ
Roland |
nn oi
To produce sound,
raise the SUSTAIN control to “10”.
Soften the beginning and ending of the envelope by raising
the ATTACK and RELEASE controls to about 1”. Next,
set the VCO RANGE switch at 4’. The resulting sound is
rather like a recorder. With this particular sound, the VCF
controls are set so that the VCF does not affect the sound,
so it is effectively not being used. The same sound can be
produced by merely connecting the output of the VCO
directly to the VCA as shown in Fig. 1
4,
5 The Basic Patches
VCO
Pitch
CV
Fig. 1-4 Simple RECORDER
(a) Block Diagram
ADSR
> —> OUT
Gate À
ñ
ADSR
MANUAL GATE
TIME LEVEL TIME
"© GATE
ENV-1 OUT
A DECAY SUSTAIN RELEASE @
® Patch Diagram (See also Fig. 1-7 for improved version of this sound).
— OUT
Linear
VCA mode
See Fig. 1-5 for
envelope settings
for producing other
sounds.
Try Fig. 1-4 (b) with the ATTACK and SUSTAIN controls at
“0”, and the DECAY and RELEASE controls at about E
This produces a celesta-like sound. Next try the DECAY and
RELEASE at a point between 2" and “3” for a xylophone-
like effect. This demonstrates how merely changing the
envelope can considerably change the character of a sound.
Try playing a scale legatissimo with the celesta-like sound.
Most musicians who are used to play the piano or organ
will probably have a bit of trouble with this because in legato
playing they tend to cling to the keys in order to produce as
smooth a transition as possible between each notes being
played. The result on the synthesizer keyboard controller is
one single, long gate pulse for the entire passage. There is no
one point where there are no keys being depressed. The first
note will be loud with the following notes growing softer due
to the fading decay time. Or, with the xylophone-like sound,
only the first note will sound.
There are two ways to prevent this: The first is to develope
a detached style of playing so that the first key has been
released before the second key is pressed. This may prove
quite difficult in rapid passages which are not intended to be
played staccato. The second method is to use the trigger
output of the keyboard.
Some synthesizer keyboard controllers produce a third
output called a trigger pulse. The trigger pulse is a sharp,
short pulse which occurs any time the keyboard pitch
control voltage changes. To use this to advantage we must
keep in mind that the keyboard (in this case) has a low note
priority, which means that if more than one key is depressed,
the control voltage output will represent the value of the
lowest key pressed. |f we play a scale moving downwards and
hold each key as we press it:
me
"бр. —
fo |
®) |
the result will be a series of trigger pulses which occur each
time a new key is pressed because each new key represents
the lowest key in the group. If we play a scale upwards while
holding all the keys:
the result will be the production of only one trigger pulse
because the lowest note is held. If we release this low note,
the release will produce a trigger pulse because the pitch of
the lowest note has changed. In actual practice, this will
probably not make much difference. Most musicians have a
tendancy to develope a playing technique which produces
the desired effect. When it is desired to have the envelope
re-triggered for each pitch played, it will prove much easier
to play if the GATE + TRIG mode of triggering the envelope
generator is used. The GATE + TRIG mode is actually the
more common mode, with the GATE mode being used in a
special phrasing where it is desired not to re-trigger the
envelope generator in a legato passage.
The Basic Patches 6
7 The Basic Patches
1-3
The sounds made by the above two envelopes may be made
more percussive by using the exponential response mode of
the VCA. In this case, the DECAY and RELEASE controls
will have to be set a little higher than with the linear mode,
as shown in Fig. 1-5 (b).
Control of Tone Color
To demonstrate the effect that the VCF has tone colors,
set the basic patch shown in Fig. 1-1. Change the VCO
waveform output to a sawtooth wave ( \.) and lower the
MOD IN slider which gives the VCO its pitch control voltage
from the keyboard. Raise the VCA INITIAL GAIN control
high enough to produce sound. With the VCF CUTOFF
FREQ at “10” (HIGH), the VCO sawtooth wave passes
through the VCF unchanged. If the VCF CUTOFF FREQ
control is slowly lowered, the upper harmonics of the
sawtooth wave will be gradually shaved off until “0” (LOW)
is reached where all harmonics, including the fundamental,
are filtered out.
With most sounds, the tone color of the sound will change
during its production. Very often, this tone color change is
closely related to the envelope of the sound. As an example,
one of the factors which determines the harmonic content of
the sound produced by a wind instrument is the intensity of
the wind pressure exerted to produce the sound. When air
is blown through a wind instrument, there is a slight delay as
the wind pressure builds up before the air starts flowing
through the instrument. This causes a delay in the entrance
of the upper harmonics. Once the initial blast has left the bell
of the instrument, the air flow and harmonic content settle
down to a relatively steady state. When the air flow is
stopped, there is a slight delay while the air remaining inside
the instrument is expelled. As the pressure dies, the upper
harmonics also die. This effect can be easily imitated by
using the envelope generator to control the cutoff point of
the VCF, as shown in Fig. 1-6 (a).
With slight variations, the settings shown in Fig. 1-6 (b) are
very commonly used in producing brass sounds. The VCF
CUTOFF FREQ control is set at its lowest point so that all
the harmonics are filtered out of the sawtooth wave. When a
key is pressed, the envelope generator is triggered which
opens the VCA. At the same time, with the MOD IN control
of the VCF at 8”, the envelope control voltage from the
envelope generator causes the VCF cutoff point to sweep
quickly upwards (attack) then quickly down (decay) to the
level set by the SUSTAIN control. This action causes a very
quick but large change in the harmonic content of the sound
at the beginning of each note played. When the key is
released, another harmonic change takes place as the
envelope sweeps the VCF cutoff point downwards (release).
Fig. 1-5 Effects of the Envelope on a Sound
(Try these envelopes with the patch in Fig. 1-4 ®)).
(@ With LINEAR VCA mode
Piano-like
Pizzacato-like
(b With EXPONENTIAL VCA mode
Piano-like
Pizzacato-like
MANUAL GATE ? 20UT
ATTACK DECAY SUSTAIN MESA
TIME TIME LEVEL
TRE
ENV-!
MANUAL SA fi OUT
TACK DECAY SUSTAIN RELEASE
ATE TIME
Пей
MANUAL GATE ENV 20UT
ATTACK DECAY SUSTAIN RELEASE
O TIME TIME LEVEL TIME 1
"mu "
>
=
<
MANUAL O ® our
se DECAY SUSTAIN dt
TIME LEVEL
ie
в
Г.
The Basic Patches 8
Fig. 1-6 HORN (also TRUMPET, TROMBONE, TUBA)
(a) Block Diagram
veo > VCF | VCA —> OUT
ADSR
Pitch
CV
Gate \ A
| Trigger
LL LIL]
(b) Patch Diagram
VCO VCF VCA
pee el 0)
p
wing | ТЫ setting | | |
La
e needed for ==
trumpet SESE SE
у
71 *
rm
TRUMPET:
Same as horn but VCF
MOD IN is setat “9” and VCO
waveform switch at [là .
TROMBONE:
Same as horn but VCF
MOD IN is set at 9”.
TUBA:
Same as horn but shift
VCO RANGE to 32 and
change envelope to:
)
A
LLL
TIT
JU
TIT
dE
ЗЕ
=
В
3
7
ТАМ
$ — Envelope
modulation
- input
Ha L
ЧИ
e —
e (©
L
<
o
S
z
e
©
a
a E
3 El
Pitch cp = i aD
4 ar
sa MANUAL GATE Env OUT E e MANUAL GATE ENV Out
e 8 ALE Peas
E A DECAY SUSTAIN RELEASE weg DA Ea ATTACK DEC N sus = IN BELA
E cv O TIME LEVEL — TIME A СОАСеШНЕНЙ Fo O ME TIM ©
А
= EXT GATE {E À E A ” GATE ~ -
fo = e TN A
= @ - = a Er Tu LE в — > ® -
#4 3 — — A —
оо | ея ео = ' + +
— — —
su (Es WEEE en} . - © sr . › 0 W
= о wer | [aro
Kio ate
EF e - | . {
a ‚ Tn, 1 - — <=
«su Ea je Jo; - E y
Fo
© 099 i“ Roland
Ramer rises Tyrie
© NT |
9 The Basic Patches
The level of the SUSTAIN control will determine the
harmonic content of the brass sounds while the notes are
held; however, this sound should be first adjusted so that the
harmonic change which occurs at the beginning of each note
sounds right, then the VCF MOD IN control can be adjusted
for the desired tone color produced with sustained notes.
The higher this control is set, the brassier the sound will be.
This brass sound demonstrates that the importance of such
things as envelope control can be far greater than the actual
harmonic content of a sound. It is the tone color changes
which make this particular patch sound like a brass instru-
ment, more than the tone color of the sawtooth wave. This
can be proven by moving the VCF envelope MOD IN slider
to “0” and moving the VCF CUTOFF FREQ to “10”.
Pressing a key now produces a sawtooth wave with no
harmonic changes, which is very different from the previous
brass sound.
The tuba, being much larger than other brass instruments,
produces lower pitches and its size causes the changes in air
pressure inside it to be much slower because more air is
required. Using the patch of Fig. 1-6 (b), a tuba-like sound
can be produced by changing the VCO RANGE switch to
32’ and changing ATTACK, DECAY, and RELEASE to "2".
Also, try SUSTAIN at about “6”. Fig. 1-6 (b) also shows
how other minor changes will produce other brass sounds.
Try using one of the brass sounds to play a scale passage
legato (pressing the second key before releasing the first)
with the envelope generator switch at GATE + TRIG and
also at GATE and note the difference in the phrasing effect.
With GATE + TRIG, the brass player seems to be taking a
breath for each note, while with GATE, the whole passage
seems to be done in one breath.
The three sounds associated with Fig. 1-4 can be greatly
improved by using envelope control of the VCF to produce
harmonic changes during attack and release, as shown in Fig.
1-7.
The tone color of many instruments will also change with
pitch. Set the patch shown in Fig. 1-7 (recorder). Lower the
envelope MOD IN control of the VCF and raise the VCF
CUTOFF FREQ control to about “4”. Play a few notes at
the very bottom of the keyboard and note that the sound
is highly filtered so as to produce mostly the fundamental
with the upper harmonics being very weak. Next, play a few
notes at the top of the keyboard and note that the filtering
action is so strong that the sound can barely be heard, if at
all, as these higher pitches are above the frequency at the
cutoff point of the VCF.
The Basic Patches 10
Fig. 1-7 RECORDER, CELESTA, XYLOPHONE
Block diagram same as Fig. 1-6 (a).
RECORDER:
— QUT
INITIAL GAIN
DD
Mi
VCA mode switch
USE VCA EXPONENTIAL
MODE FOR FOLLOWING:
Celesta-like:
MANUAL GATE ENV ! OUT
ATTACK OECAY SUSTAIN RELEASE
O TIME LEVEL TIM
EXT GATE
©-
Xylophone-like:
ADSR
(LE
MANUAL GATE
O м Ya DECAY SUSTAIN RELEASE
ME
TIME LEVEL
EXT GATE
Gate |
mes de EZ xt e MANUAL GATE ENV 20UT
В e: O ATTACK DECAY SUSTAIN RELEASE
fo a ia 4 il
ë E : Trigger ©
de ea
Em:
Next raise the VCF MOD IN slider marked KYBD (key-
board) and play a few pitches at both ends of the keyboard.
Now the keyboard pitch control voltage is being used to
control the cutoff point of the filter. This means that when
notes are played, the VCF cutoff point will follow the pitch
so that the cutoff point remains the same in relation to the
pitches. In other words, the harmonic content of the sound
will not change, regardless of the pitch played. When actually
listening to the sounds produced, the upper pitches will seem
brighter, with more harmonic content, but this is because of
the characteristics of the ear. To produce sound which seems
to retain the same harmonic content, set the VCF KYBD
MOD IN slider at about “6” and play a scale from the
bottom of the keyboard to the top.
3
11 The Basic Patches
Fig. 1-8 shows a clarinet-like sound in which the keyboard
control voltage is added to the patch for producing a sound
which seems to get brighter as the pitch moves up.
Fig. 1-8 CLARINET (also BASS CLARINET)
(a) Block Diagram
VCO —>| VCF >
A
Pitch
cv ADSR
Gate
| = Trigger
—
®) Patch Diagram
VCF
co
NTO OUT,
' J
Pitch CV
modulation
input
№
№ MANUAL GATE = 1 out
Pitch ATTACK DECAY SUSTAIN RELEASE
TIME TIME LEVEL
i CV "o GATE 2
si
re
ta Gate y
ve, - Hr
E
И оч он ms me =
«ду
Ë
№
&
&
NUNES NDA
Trigger
INITIAL GAIN
a
——
MOD |
#
—
I
—> OUT
— QUT
CAES EE
ZZ
ESS
VANE
BASS CLARINET:
Shift VCO RANGE to 16’
and make envelope ATTACK
and RELEASE times slightly
longer.
The Basic Patches 12
The VCF RESONANCE control is used for accenting the
frequencies at the cutoff point of the filter. It can be
considered to add a nasal quality to the sound, like that
produced by double reed instruments. Fig. 1-9 shows the use
of VCF resonance to produce English horn- and oboe-like
sounds. The oboe sound adds a new element to the basic
patch: the high pass filter (HPF). In this patch, it is used to
slightly suppress the fundamental and other lower harmonics
to make the sound less full than the English horn sound. A
higher amount of resonance is used in the bassoon patch of
Fig. 1-10.
Fig. 1-9 ENGLISH HORN (also OBOE)
(@ Block Diagram
Y
VCO
|
VCF — HPF — OUT
| |
Pitch CV A A (OBOE)
ADSR
Gate À
Trigger
||
'b) Patch Diagram
o QU
‘ie High pass
filter
| MODO IN OBOE:
- ADSR Set VCF CUTOFF
= rer FREQUENCY at “6”; set
| us mue ue MM CIE mar ee fixed HPF switch at 1”.
E Pitch CV ©, (Fig. 2-7 FLUTE shows a
| Gigi o : — tone color modulation
| q= Em mm Em EE Бин с | Е effect which can be used to
ms ame ds a we a! ЗЕ Е - improve these sounds.)
Trigger
i= Roland Г
13 The Basic Patches
Fig. 1-10 BASSOON
(@ Block Diagram
VCO >| VCF —>| VCA > OUT
A
Pitch ADSR
CV
Gate + Trig À
®) Patch Diagram -
VCO VCF
"a 00 CUBO IN DEAN EAS 8 e O g—
| MANUAL | | | | i
ort OUT
INITIAL GAIN
O
de
= |
‘ © © ©
MOD IN MOD IN 3
а 2557
EE (Fig. 2-7 flute shows a tone
e Pitch CV Yer men re ol Sa color modulation effect
# ver dl fe which can be used to
в - improve this sound.)
Gate + Trig.
E J EEE um Kn BES EE
UA
"
i
æ Roland
Boten) MM une alarm
—
| : | || | ALL E | || |
The tuba sound of Fig. 1-6 is also good for demonstrating
the effects of resonance. Set the synthesizer as shown using
the tuba envelope, and raise the VCF RESONANCE control
о “7”. Pressing a key now produces a “wow” sound, a
sound that is associated with the synthesizer. Resonance
accents the frequencies at the cutoff point of the filter and
due to the envelope control, this cutoff point sweeps up and
down each time a key is pressed, producing the “wow”
sound. The relatively long envelope times used with the tuba
sound makes this effect stronger than with the horn envelope.
Try different positions of the SUSTAIN control.
1-4
The Basic Patch for Additive Synthesis
Fig. 1-11 shows a simplified block diagram for additive
synthesis. The mixer is used for mixing the sine wave outputs
of the VCO's in the proportions desired. Since the harmonic
content of most sounds changes often during production, the
ideal patch would include one ADSR/VCA combination for
each VCO output. Also, most sounds have far more harmonics
than the four sine waves shown.
The Basic Patches 14
Sync
(See text, p. 16) VCO
Y
Fig. 1-11 Additive Synthesis Block Diagram
AD
== == == = == [о === == == ==)
VCO
Ÿ
I
y
VCO
Ген в ве Перечне
|!
e
QE
AU
VCO
\
Сате
ADSR
Pitch CV
||
» OUT
Fig. 1-12 shows the above block diagram patched on the
synthesizer. Set as shown, this patch will produce sounds
very close to the sound of orchestral tubular chimes. The
best sounds are produced in the approximate range between
Ca (middle C) and Cs (one octave above middle C).
15 The Basic Patches
Fig. 1-12 CHIME (Additive Synthesis)
vco-1 | vco-2 |
system-100m 117 О
Ex EDO EXT a», A
PW MOD | PWMOD + т
else
| MANUAL | | MANUAL |
"ЗЕ. ЗЕ ЗЕ.
| | I | LU
. mu . rm
| I |
RANGE SYNC RANGE SYNC
w 8, a oh w 8 4 oh
© О ® 2 2
® o
PITCH PITCH |
AN,
© $
"ее оо
; >
|
-0— qui
Or
0 —
Pitch CV
VCO-3 | VCO-4
svstem-100m ЛО :
.. in, SEE fi
® Oma. 1 ©—@
VCO tuning (See text):
(Optional: VCO-5) VCO-3
©
ADSR
(LA
ENV "OUT
DECAY SUSTAIN RELEASE 223
TIME LEVEL TIME ®)
MANUAL GATE
ATTACK
O TIME
ENT GATE
Yoo > #11 | o
Е f Gate + Trig F
if т
€ oo ned ede
|] Perfect 4th
— OUT
To tune the VCO's, set all the mixer controls at “0”. VCO-1
represents the fundamental pitch of the chime sound. Raise
the VCO-1 SIG IN control of the mixer and tune VCO-1
to the desired pitch relation with the keyboard (A above
middle C should produce 880Hz since the VCO RANGE
switch is at 4’). Tuning will be easier with the envelope
generator SUSTAIN control temporarily at “10”. Next, raise
the VCO-2 slider and tune VCO-2 one octave above VCO-1.
Lower the VCO-1 control; raise the VCO-3 control and tune
VCO-3 to a perfect fourth above VCO-2?. Raise the VCO-4
and VCO-1 controls and tune VCO-4 to a minor sixth below
VCO-13. This sound can be improved by adding a fifth
VCO tuned a perfect fourth below VCO-2.
2. To tune: With only the VCO-2 control raised, press F and note the
pitch; lower the VCO-2 control and raise the VCO-3 control; press
C one fourth below the previous F and tune VCO-3 to the F pitch
noted before; with both VCO-2 and VCO-3 controls raised, fine
tune VCO-3 so as to eliminate the beat.
3. Tune VCO-4 so that it produces the pitch of C when the E a minor
sixth below is pressed.
The Basic Patches 16
Fig. 1-13 shows a variation of the chime patch. The lowest
pitch in the patch is fed to its own separate VCA, controlled
by a second envelope generator (ADSR-2). The ATTACK
| control of this envelope generator is set at about “1”, thus,
| when a key is struck, the lower pitch is slightly late in
entering the total sound. Also, with ADSR-1 DECAY and
RELEASE at 8” instead of “10”, the higher pitches die out
slightly faster than the lower pitch. The ideal patch would
contain an ADSR/VCA combination for each VCO output.
Fig. 1-13 Variation of CHIME patch
See Fig. 1-11
|
VCA
Internal
signal
mixer
INITIAL GAIN
ADSR-1 | en
N
MANUAL ©” Y 10U1 №
ATION DECAY er ir a
TIME №
Et © E
; | NINAS
y
;
}
MANUAL GATE Env 20UT |
ATTACK DECAY SUSTAIN RELEASE
TIME TIME '
“® GATE m | HE =
EEE SE
©-
we
ADSR-2
Regardless of how perfectly the VCO's are tuned, they will
eventually drift off-pitch, causing beat frequencies to
appear. In some sounds, such as the chimes above, this beating
is not undesireable but actually greatly enhances the sound;
in other sounds, the beating would be intolerable. The VCO
sync function can be used to lock the frequency of one or
more VCO's to some musical interval in relation to a con-
trolling VCO. In Fig. 1-11, the controlling VCO is the
bottom one.
17 The Basic Patches
Pitch
CV
(a) Block Diagram
Sync
Fig. 1-14 CLARINET (Additive synthesis)
VCO-1
ad
y
VCO-2
Pitch CV
Gate + Trig
® Patch Diagram
VCO- ane
ADSR
EEE EEE
SEEN DU ¡E
5 a
ss "
| O
№
: y ©
es
a
SE
A
ST
&№
Gate + Trig
NES DEA LINE cn сено нь,
UN AS EN MN BES
yA patch cord
Tune VCO's
to unison
Pitch CV В
ATEO
ENTE
ADSR
TANTA
CDI
MANUAL GATE
O
ATTACK DECAY SUSTAIN RELEASE
TIME IME
TIME LEVEI T
i Roland
mt Yade y в
mun
LA
|
N
ru
OUT
Square wave
NIN. Sawtooth wave
| 11
9...
t 7
— MOD INR-—
— OUT
a
Fig. 1-14 shows an improved clarinet sound which is a better
example of how the principles of additive synthesis are
applied to the voltage controlled synthesizer. The sawtooth
wave ( [X) contains all harmonics, while the square wave
(TL) contains only the odd-numbered harmonics. When
these are mixed together at the VCF input, the odd-numbered
harmonics of the square wave reinforce the odd-numbered
harmonics in the sawtooth wave. In the total sound, the odd-
numbered harmonics are very strong and the even-numbered
harmonics are rather weak, which agrees roughly with the
spectrum of the clarinet shown in Fig. 1-15. This addition of
harmonics takes place because, by means of the sync, the
two VCO'’s are phase locked at unison with each other.
Since the interval used is unison, either VCO could be used
as the controlling VCO. The sync patch cord of Fig. 1-14 (b)
could be reversed so that one end is in the VCO-1 SYNC
OUT jack and the other end in the VCO-2 SYNC IN jack.
The Basic Patches 18
Fig. 1-15 Spectrom for the Clarinet
Pitch = G 3
1 3 | |4 | [|5 |6 |? oi
—40 1
100 300 500 1K 1.3K
Frequency
Harmonic number
The length of the lines indicates the strength of the
harmonics. Note that the second harmonic is missing
from the sound.
Chapter Two:
Expansion of the Basic Patches
2-1
Introduction
As the name implies, mixers are used for mixing individual
signals together. In the studio, the mixing console might be
considered the heart of the studio since this is where all the
musical parts are mixed together to form the finished master
tape. In this book, we shall be concerned primarily with
mixers as used in the process of synthesis.
Fig. 2-1 shows a block diagram for a synthesizer module. As
shown, there are two input mixers, one for audio signal
inputs and the other for control inputs. This diagram could
represent a VCA, a VCF, or, if the audio signal mixer were
omitted, a VCO. These mixers differ from ordinary audio
mixers in two respects: First, they are designed to handle
both audio signals and control voltages; second, and more
important, these mixers are summing mixers. This means that
the output will be the algebraic sum of the inputs at any
given instant. For example, if the control inputs consist of
two control voltages, one +5 volts and the other —2 volts, the
output will be (+5) + (+2) = +3 volts. This is important
because manual control of the module function is derived by
means of varying a voltage source input to the control input
mixer, as can be seen in Fig. 2-1. In the VCF, the control
marked “function control” would be the VCF CUTOFF
FREQ control; in the VCA, the VCA INITIAL GAIN control;
and in the VCO, a combination of the PITCH and RANGE
controls.
Fig. 2-1 Block Diagram for Synthesizer Module
SIG IN
level controls
Signal
input
Signal A A RA e
inputs de
Control
inputs Vv © V »
MOD IN
level controls
+10 volt source
Module
FUNCTION
control
Control
Module
. —— >
Function
A
Control
input
Input
Mixer
Signal
out
1. In the case of the audio signal input mixers, the fact that they pass
control voltages is of little use since the modules themselves will
not pass fixed voltages through their signal inputs.
There is one major limitation to this summing process. The
mixer can produce maximum output levels of slightly over
10 volts (either “+” or “—”), This means that ¡f three inputs
of +10 volts are used, the summed output will be a little over
+10 volts rather than +30 volts. This point should be kept in
mind because it represents a distortion of what the true
output should be. With control inputs, this can usually be
recognized easily. When the module function is driven to its
maximum limit, a further increase of control voltage will
have no effect on the function. With signal mixers, the result
is distortion of the sound, which will usually be indicated by
the red LED's in the VCF and VCA. Distortion can be avoided
by keeping the level controls somewhat lower than maximum
when there are more than one or two high-level inputs.
In the case of the VCF, if the CUTOFF FREQ control is
raised to maximum, the result is a +10 volt control voltage
input to the VCF. Since this represents the maximum high
for the cutoff point, it can be seen that additional control
voltages at the control inputs will have no effect on the
cutoff point of the filter unless they are negative voltages. As
another example, if the VCF produces the desired tone color,
and it is later decided to add a control voltage input from a
source such as an envelope generator, the CUTOFF FREQ
control will have to be lowered an appropriate amount to
compensate because the actual cutoff point of the filter is
decided by the sum of the control inputs.
Expansion of the Basic Patches 20
21 Expansion of the Basic Patches
2-2
Frequency Modulation
Modulation refers to the control of one parameter by
another. Frequency modulation is the control of a frequen-
cy by means of some parameter, such as a control voltage,
as when controlling pitch with the keyboard controller.
Another form of frequency modulation commonly used in
synthesis is the control of frequency by means of a low
frequency waveform generated by a Low Frequency
Oscillator (LFO). Using a low frequency sine wave to
modulate the pitch of a VCO produces a wavering of pitch
called vibrato.
Fig. 2-2 shows the block diagram and patch diagram for a
violin sound which uses vibrato. Also shown are variations
needed to produce other string instrument sounds. Try the
violin sound with the LFO MOD IN to the VCO at “0” and
notice that it does not sound very much like a violin at all.
Next, lower the LFO DELAY control and raise the LFO
MOD IN control to “2” or “3” and note the sound. Last
raise the LFO DELAY control back up to about 3”. With
the delayed entrance of the vibrato effect produced by the
built-in LFO delay, the violin sound becomes very natural.
The delay effect is triggered by the keyboard gate pulse
rather than the keyboard trigger pulse, so the delay effect
occurs only on the first note of any passage played legato.
Fig. 2-2 VIOLIN with Delayed Vibrato (also VIOLA, CELLO, STRING BASS)
(a) Block Diagram
» LFO VCO >| VCF
(with
delay)
Gate triggers delay A
Pitch CV
Gate + Trig
Y
HPF
—
ADSR
VCA
ouT
| (0) Patch Diagram
control
(depth of
vibrato)
FE qu
| input
Gate
SS ES
delay)
LFO
Delay time
( control
(Triggers
* REO CV IN
ADSR
Gata + Trig
MANUAL GATE
O
"©
|
e e
ENV + OUT
ATTACK DECAY SUSTAIN RELEASE Za
TIME LEVEL
Expansion of the Basic Patches 22
— OUT
VIOLA:
Set HPF switch at OFF and
raise VCF RESONANCE
control to 11”.
‘CELLO:
Same as viola, but set VCO
RANGE at 16’ and
= Roland
Rotana Sadas Yy vien
envelope:
MANUAL GATE ? our
ATINA DECAY SUSTAIN BREAN
TIME LEVEL TIM
e
STRING BASS:
Same as viola, but set VCO
RANGE at 32' and envelope:
“© GATE Env 2? e
ATTACK DECAY SUSTAIN HELEASE
TIME TIME LEVEL TIME
HH
(Fig. 44 shows typical
equalizer settings for
improving these sounds.)
23 Expansion of
the Basic Patches
LFO's in some synthesizers do not have a built-in delay
function.
Fig. 2-3 shows how a delay function can be
patched. Even with the built-in LFO delay, this patch is
worth studying because it gives insights into the logic used in
synthesizer patching. The output of the LFO is fed to the
signal input of a VCA. The output level of the LFO wave-
form passed through the VCA will depend on the level of the
control voltage at the VCA control voltage input, which is
derived from a second envelope generator. When a key is
pressed, the second envelope generator is triggered, which
“opens” the VCA to allow the LFO waveform to pass.
Raising the ATTACK control will delay this opening.
INITIAL GAIN
©
Y
rm
qui
——
o
©—
This control should be
set about the same as
the release time which
——
——
—
—— N
———
| —
controls the envelope
Fig. 2-3 Delayed Vibrato To VCO
y modulation
(a) Block Diagram input
LFO >! VCA
TO A
VCO | To ADSR-1 ADSR-2 =
EN
|
EL |
| | | | | |
b) Patch Diagram
LFO
LFO
Depth of vibrato
Pitch CV may be controlled
to VCO by this, or by VCO
modulation input
control.
Gate
to ADSR-1 ADSR-2
This
controls WEEE RGN
delay O TIME bis ve "ue 9
time PE 9 = ЗЕ : ЗЕ Que css BR ARS
Gate all AE 3 ЗЕ © -
Г ро
9000 Roland of the sound.
lo O
To modulation
input of VCO
I Expansion of the Basic Patches 24
Fig. 2-4 (a) shows how trills can be produced by using the
square wave output of the LFO to control the pitch of the
VCO. The KYBD TRIG switch is ON so that the square wave
is phase locked to the keyboard gate pulse2. This means that
whenever the beginning of a keyboard gate pulse appears, the
LFO is forced to start its wave-generating cycle from the
beginning point, as shown in Fig. 2-4 (b). In terms of the
trill, this means that whenever a key is pressed, the trill will
always start with the upper of the two notes and the first
note will be just as long as the other trill notes. Since the gate
pulse triggers the function, it will occur only on the first note
in a legato passage.
Fig. 2-4 Trill
@ Patch Diagram (Block diagram is essentially same as Fig. 2-2 @).
VCO
x! ACTO OUR
A EEE AAA a E TE == roa я
CUTOFF WES
FHE
— OUT
INITIAL GAIN
Controls ; ri - . meo e
. ; ] . . \
trill [| 3 | в - : Y a m
rate =) a A
ao = оне
Controls
€ | upper trill = - ESS
note de = = ЗЕ
x Y DELAY 7 FORM CHE у ADSR e | :
=! | = 11
© o MANUAL GATE e o © ® ® ©
ATTACK DECAY SUSTAIN MEL tas —— MODO IN В MOD IN ——
xYBO O TIME TIME LEVEL TIME a
TRIG i
N 9
0
TTT
ДЕР
ПАРИ $
(0) LFO Phase Lock Waveforms
Sine Wave AAA AN
Square Wave |
Sawtooth
Wave a
!
|
TN ~~"
NE
Sawtooth
Wave
Gate
Pulse
Key Key
| ON OFF
2. KYBD TRIG = keyboard trigger. The name comes from the fact
that this phase-lock function is triggered by the gate pulse.
25 Expansion of the Basic Patches
For tuning the trill, set the LFO at its lowest frequency
(FREQ RANGE at “L", FREQ slider at 07). With the LFO
MOD IN to the VCO at “0”, tune the VCO in the normal
way. To tune the trill to a major third, for example, strike a
key and note the pitch. Next, strike the key which is a major
third below the first key and raise the LFO MOD IN control
of the VCO until the pitch matches the pitch noted before.
Striking a key forces the LFO square wave to start over again
at its highest point and with the FREQ RANGE switch at
“L", this high part of the wave should remain long enough to
set the MOD IN level. (If not, strike the key again). Remove
the MOD IN patch cord and check the pitch by striking the
original key again.
As an experiment, try a trill at a major third, then change the
LFO OUTPUT LEVEL switch to “x1” and try the other
LFO waveforms.
Fig. 2-5 shows a patch in which the frequency of one VCO
is used to modulate (control) the frequency of a second
VCO. The result is a very complex waveform which contains
overtones normally not available from the output of a VCO.
These non-harmonic overtones can be very useful for
producing metallic clanging sounds such as those produced
by bells and metal bars.
To experiment with this type of frequency modulation,
start by lowering to “0” the MOD IN which feeds the output
of VCO-2 to VCO-1 (Fig. 2-5). Tune VCO-1 to the desired
pitch relation with the keyboard. Raise the VCO-2 SIG IN
to the VCA and tune VCO-2 to unison with VCO-1, then
return the control to “0”. Next, while tapping a key near the
center of the keyboard, slowly raise the VCO-1 MOD IN
control and note the changes which take place. Above level
"7", raise the control very slowly because the changes occur
extremely rapidly. The most important point to note is that
the apparent pitch of the sound does not remain stable but
rises as the MOD IN control is raised. Starting with the
MOD IN control at “0” again, play a short five-finger scale
passage over and over while slowly raising the MOD IN
control. Note that although the tuning changes, the overall
pitch relations between the notes remain more or less correct.
Try this experiment again with both VCO's set at 2’ and note
that the pitch relations will sometimes be different, especially
if played in the higher range of the keyboard. This change in
tuning should be kept in mind if it is desired to use this ty pe
of arrangement to play melodies.
Tuning may best be accomplished by first ignoring the
tuning and concentrating on producing the desired sound.
Once the desired sound is found, it is simply a matter of
determining what the tuning discrepancy is and correcting
it. For example, if the desired sound produces pitches a
minor third higher than the correct pitches, tune the VCO's
a minor third lower than normal. VCO-1 may be tuned by
removing the patch cord to its MOD IN jack. In this way,
the delicate setting of the MOD IN slider will not have to be
altered.
a Expansion of the Basic Patches 26
Fig. 2-5 Frequency Modulation (by VCO)
(a) Block Diagram
VCO-1
— = >| VCA OUT
VCO-2
Pitch CV À
NUM
| (Cb) Patch Diagram
ADSR
Gate A
y
Vv C 0-1 = VCO-2. == ue a a o ES e a | НЫ a EE me EER
SYSTEM-100M 112 LO) E Ы Е Temporary patch cord for tuning VCO-2
“7
со
М
a
Ext
Py MOD W MO
=
@ ® e ==
ATTACK OFCAY SUSTAIN
O TIME TIME LEVEL
©
| ® e — OUT
— E 8% BE
| i= Roland
IrIEM- 00M Di = — —_ оной Ме Зав,
_
The most obvious variations on this particular sound would
be to try different VCO waveforms and tuning the VCO's
to intervals other than unison. Another variation would be
to leave the VCO-2 SIG IN to the VCA raised. A VCF
could also be used to alter the tone color.
27 Expansion of the Basic Patches
One “practical” application of the synthesizer is to use it to
whistle for the dog, using the envelope generator to control
the pitch of the VCO (Fig. 2-6). Only the ATTACK control
is raised so that when a key is struck, the output rises
quickly to maximum, then “instantly” drops back to zero.
The VCO pitch follows this pattern to produce the whistle.
An interesting variation of this is the “wolf whistle”. Press a
key near the center of the keyboard. Next quickly raise
the DECAY TIME control to about “2” and press the same
key again, holding it down until the sound stops.
Fig. 2-6 DOG WHISTLE
(a) Block Diagram
VCO
Pitch CV A A
—>| VCA
ADSR
Gate + Trig
nino
(® Patch Diagram
“® тр тт
2
— MOD in —
ats
2% 3 MANUAL О WA
# Pitch CV ATTACK DECAY SUSTAIN чужая т É
т ME TIME w
E:
E Ex e es
e . ar
ES -
© Gate + Trig o.
и он Шин син > i
ES A pa
[o oe
“e “а
Ш 8
© 5 © 9 = Roland
LT ya
OUT
Expansion of the Basic Patches 28
2-3 Tone Color Modulation
Two of the most common forms of tone color modulation
were discussed in Chapter One: Pitch control of tone color
and envelope control of tone color. In many types of sound,
the tone color will waver at the vibrato rate of the sound.
This effect can be easily produced by using the output of the
LFO to control the cutoff point of the VCF, as shown with
the flute and piccolo sounds of Fig. 2-7.
Fig. 2-7 FLUTE (also PICCOLO)
(a) Block Diagram
VCO >| VCF “Y OUT
A A A 1
LFO
ADSR
A
|
() Patch Diagram
VCO VCF VCA
„ «=» ' >
— OUT
Try adding tone
color modulation
AA
O: [| effects to Figs. 1-9
ma and 1-10.
{== | |
. o cosmos
A LFO
=a Ex = = = == a
Е «св ADSR
Gate + Trig
FREQ FREQ DELAY WAVE FORM OUT
RANGE
o y,
e
a
o? FREO CV IN
LLFO QUE
MANUAL GATE Ñ
ATTACK DECAY SUSTAIN LEN
О “*
ExT GATE
|
Es |
ENV-1
TIME LEVEL
i
PICCOLO:
I= Roland
ROA Mudo rue
— bes
Set VCO RANGE at 2‘
and envelope:
MANUAL GATE
HT
ЕМУ 2 ОМ!
ATTACH DECAY #5 3 pn PELE
O TIME
EXT GATE mi
© -
29 Expansion of the Basic Patches
LFO modulation of the VCF cutoff point is called growl,
although this term is more often associated with the effect
produced by exaggerated LFO control of tone color. This
effect can be heard with the flute/piccolo patch of Fig. 2-7
by raising the LFO FREQ slider to “10” and changing the
LFO OUTPUT LEVEL switch to “x1”.
Fig. 2-8 shows a clarinet sound which uses a second VCO to
modulate the cutoff point of a VCF. Tune VCO-1 to the
desired pitch relation with the keyboard. Raise the VCO-2
SIG IN to VCF-1 and tune VCO-2 two octaves and a perfect
fifth above VCO-1 before inserting the sync patch cord
between the VCO's. It is easier to set the VCO-2 RANGE
switch to 8', tune to a perfect fifth above VCO-1, then return
the RANGE switch to 8'. After tuning, insert the sync patch
cord. When a key is pressed, the VCF-1 cutoff point rises to
a level determined by the envelope control of the cutoff
point. The VCO-2 MOD IN to VCF-1 causes the cutoff point
to waver very rapidly above and below this level, adding
harmonics which were not present in the original VCO-1
square wave.
Fig. 2-8 CLARINET (VCF Modulation by VCO)
(@ Block Diagram
Sync
> VCO-2 VCO-1 >| VCF-1 » VCF-2 VCA
|
A À A A A
Pitch
CV LFO ADSR
| Gate + Trig
| | | | Tuning: VCO-1
(Tune VCO-2 before inserting
sync patch cord)
OUT
® Patch Diagram
(For tuning only)
VCO-1
I
system-100m 112
® @ -
Expansion of the Basic Patches 30
МСЕ-1 VCF-2
ny
Gate + Trig
75 — — — — а
E Pitch
в СУ
Е
: LFO
ep
| REO DELAY — WaV ver ORM |8
Но
É Fais
5 = =, N
E
:
E
Е
|
УРУМ ме
0 0)
——
1D O:
le]
Museum Ма vada Wat
31 Expansion of the Basic Patches
Another form of tone color modulation can be produced
with the VCO pulse wave ([|__|) output. Fig. 2-9 shows
how the VCO square wave and pulse wave are generated. In
(a), the sawtooth wave of the oscillator is passed through
the square/pulse wave shaper. The shape of the output wave
will depend on the level of the control voltage input to the
wave shaper. When the level is 0 volts, the wave is square
as shown in (b). When the control input is +10 volts, the
wave is a pulse wave with a 10% duty cycle, which means
that the waveform is high (“on”) 10% of the time and low
(“off”) 90% of the time. If the total control voltage input is
more than slightly over +10 volts as shown in (e), the pulse
wave will be over-modulated and there will be no square/
pulse wave output.
The English horn sound of Fig. 1-9 uses the MANUAL con-
trol to generate a pulse wave with a 10% duty cycle, produc-
ing its nasal double reed sound.
Г Expansion of the Basic Patches 32
Fig. 2-9 Square and Pulse Waves
(a) Square/Pulse Wave Block Diagram
VCO = = = == = = = = = = >= [< OUT
+10 volts
A
Signal Square/
I (pitch input
MANUAL | Pulse +;
control control Wave [i | out
mixer Shaper
not
h
Control e A Pulse width
input MIX control input
EXT P.W. MOD
control,
(b) 50% Duty Cycle (Square Wave)
«—— (100%) ——
vCo
| -4— 50% — > «< 50% —>
m QUT
ON
Duty cycle = “ON” time
OFF de |
(©) 30% (approximately) Duty Cycle
30%
<4+——> «<—?os/0% —
OUT
ON
VCO
EXT VCO QUT,
PW MOD 1 2
OFF
| MANUAL |
© 10% Duty Cycle =
mi
vCO “il
EXT vCO OUT,
PWMOD 1 2 10% — >| |-«—— 90%—_
mee >> OUT RANGE SYNC
ON — в Ва 2
©
OFF PITCH |
0
(e) Over-modulation ||
# +10 volts in с
VCO =
OUT |
ON 0%
| 1
©
——MODIN—
OFF
33 Expansion of the Basic Patches
Fig. 2-10 shows a bowed string bass sound which uses the
envelope generator to modulate (change) the pulse wave
during the progression of each note played. While holding
down a key, try raising the VCO EXT P.W. MOD control
to “10” and note that the pulse width becomes so narrow
that the sound almost completely ceases. If the MANUAL
control is now raised to only about “1” or “2”, the pulse
wave will be over-modulated and sound will cease. Compare
the differences in the string bass sound with the EXT P.W.
MOD control at “0” and at 4”.
Expansion of the Basic Patches 34
Fig. 2-10 BOWED STRING BASS
(a) Block Diagram
>| VCF-1
Y
© PWM
LFO + > VCO Le, VCA OUT
NA >| VCF-2
Pitch CV
ADSR
Gate + Trig A
(b) Patch Diagram
— OUT
PSS
LFO
LFO
PS
T
FREQ FREQ DELAY WAVE FORM OUTPUT
RANGE LEVEL
TEO. Т |
| 1 1
‚© ®
+ MOD IN—
Pitch CV | Е
=m mde mf
“ ADSR
ENV-1
MANUAL CATE ENV four
ATTACK DECAY SUSTAIN RELEASE 2
e im rm LEVEL TIME \
TIME E LEVE
| Gate + Trig . fs - o.
8
35 Expansion of the Basic Patches
Fig. 2-11 shows a string ensemble sound which uses LFO
modulation of pulse width to produce an ensemble or
chorus (“group”) sound. Try this sound first with only
VCO-1. When trying both VCO's, tune the VCO's with the
VCO-1 EXT P.W. MOD control at “0”, The sound is greatly
enhanced if the two VCO's are just out of tune enough to
produce a beat with about the same frequency as is often
used for vibrato effects.
Expansion of the Basic Patches 36
Fig. 2-11 STRING ENSEMBLE
VCF OUT
(a) Block Diagram
PWM
>| VCO-1 =
LFO A =
Vco-2 >
Pitch
CV
Gate + Trig
ADSR
(b Patch Diagram
LFO
7
KYBD
TRIG
|
2.
1 z
FO OUT! RANGE
в во
=
LP
...
j
— MOD IN —
Roland
a шв
Pitch CV
e na es vs ZA e Vx Men WO
Gate + Trig
Zu Un Vu o o e Ue 0 UU"
+
i
i
x
Ê
&
(-)
+ es
o):
—
fr——:
|
|
— OUT
l=
(@) © o
—mob N—
ADSR
LE
MANUAL GATE ENV TOUT
ATTACK DECAY SUSTAIN RELEASE =
() ME LEVEL TIME
TIME Li E
EXT GATE
BEY:
Cu
= Roland
Retard cho tres
37 Expansion of the Basic Patches
2-4
Amplitude Modulation
The control of a VCA by an envelope generator is by far the
most common form of amplitude modulation found in
electronic music. Less common is the use of a low frequency
sine wave to control a VCA to produce a wavering of loud-
ness called tremolo. Tremolo, growl, and vibrato are all
closely related; indeed, tremolo is sometimes considered
a special form of vibrato.
Fig. 2-12 shows block and patch diagrams for demonstrating
how tremolo effects are produced. Start with the VCA
INITIAL GAIN and the LFO MOD IN to the VCA at “0”.
Slowly raise the LFO MOD IN control and notice the effect
on the sound. Technically, VCA’s are known as 2-quadrant
multipliers which means that they are sensitive to control in-
puts above zero, but not below. Fig. 2-13 (a) shows that the
LFO sine wave output varies above and below the zero line.
Since the VCA is sensitive only to levels above zero, the
VCA will “open” only half of the time. Lower the MOD IN
to the VCA and raise the VCA INITIAL GAIN control to
about “5”. This has the effect of inputting a fixed control
voltage of approximately +5 volts to the VCA control input
mixer, and the VCA may be thought of as half “open”, In
this condition, if the MOD IN control is raised, the upward
and downward swings of the LFO sine wave will be added to
the +5 volt level so that the swings no longer go into the
negative control region. The result is tremolo, a wavering in
the loudness of the output sound. Using this arrangement
requires the use of two VCA's: one for the tremolo, and
second one for the envelope control. Fig. 2-14 shows how
tremolo can be added to a patch.
Fig. 2-12
®
VCO
Tremolo experiment
Block diagram
VCO
>| VCA
Pitch A
as
Patch diagram
=
4
+
1
FE
FE
F
MT
L$
т
ci}
—
RACE
Ц
ЦРО
Controls
tremolo
depth Ng
ouT
OUT
Fig. 2-13 Tremolo Waveforms
VCA Control
Input
(a) LFO Only O
+
(b) Initial
Gain Only
0
+
© LFO
+
Initial
Gain 0
VCA Audio
Output
LLAMA
Tremolo
Hh
Г Expansion of the Basic Patches 38
Fig. 2-14 Adding Tremolo to a Patch
(a) Block Diagram
VCO —» VCF —>»|VCA-1 VCA-2 OUT
LFO
Pitch CV
ADSR
Gate + Trig A
|
(b) Patch Diagram VCA-1 VCA-2
VCO VCF SYSTEM-100M systEm-100m 130' О
ie = Do Lo
® (0Ps|i===== == ===
i
í
RANGE SYNC VCF
| =
0 Hi
+ T
| ee |
© ©
— мод IN—
LFO
MANE FREO DELAY TR FORM OUTPUT
A H O.
we |
On
AN | FREQ CV IN
ADSR
MANUAL GATE
O ATTACK DECAY SUSTAIN RELEASE
TIME TIME
EXT GATE E
Gate + Trig = | o.
Sm mm == mm mm E
= Roland
— Roland 40a Sywern
39 Expansion of the Basic Patches
The Ring Modulator
The ring modulator is a circuit which falls in the same class as
пе УСА. Technically, it is known as a balanced modulator
or a 4-quadrant multiplier which means that it is sensitive to
both positive and negative swings of the control input. Fig.
2-15 shows the type of waveform produced by the ring
modulator. Notice that for negative swings of the control
input, the output signal reverses in phase so that it is inverted.
This is one of the reasons for the metallic clanging sounds
usually associated with the ring modulator. For all practical
purposes, the “X” and “Y” inputs may be considered the
same. Reversing the input connections to the ring modulator
will make no difference in the sound.
Fig. 2-16 shows how to use the ring modulator to produce
metallic glockenspiel-like sounds. The VCO's may be tuned
by first temporarily connecting their outputs directly to the
VCA inputs. Note that the sound produced by the ring
modulator in this patch is one octave below the sound
produced when VCO-1 is connected directly to the VCA.
The reason for this is that the output of the ring modulator is
a compound of the sum and difference of the input frequen-
cies; the input frequencies themselves do not appear at the
output. As an example, assume that when a key is struck,
VCO-1 produces the frequency of 880Hz. If VCO-2 is tuned
to a beat-free perfect fifth above VCO-1, its frequency will
be 1320Hz (the ratio of pitches in a perfect fifth is 1:1.5; 1.5
x 880 = 1320). The output of the ring modulator will be a
compound of 2200Hz (sum: 880 + 1320) and 440Hz
(difference: 1320 — 880). The lower of these two frequencies
we hear as the fundamental pitch of the sound. The upper
frequency becomes a strong overtone, in this case the fifth
harmonic (2200 + 440 = 5), which would be equivalent to
C# two octaves and a major third above the fundamental
pitch.
Tap on a key and slowly turn the VCO-2 PITCH control in
one direction or the other. Note that both the tone color and
pitch change. This is because changing the frequency of one
VCO will change both the sum and difference frequencies
which appear at the output of the ring modulator.
It can easily be seen that it would be possible to tune the
VCO's such that pressing A above middle C would produce a
difference frequency of 440Hz, for example, and a sum
frequency which could be any desired overtone, either
harmonic or non-harmonic. The possibilities become even
wider if the original VCO pitches are fed directly to the VCA
in addition to the ring modulator output.
Fig. 2-15 Ring Modulator Waveforms
E
Input —| Ri
ing
—_— 3
se modulator Output
Input ——»
"XX" Input
Y” Input
Output
Phase Reversal
T Expansion of the Basic Patches 40
| Fig. 2-16 GLOCKENSPIEL
@ Block Diagram
VCO-1
Y y
RM >| VCA OUT
VCO-2
Pitch A
CV ADSR
Gate + Trig
[ Л A VCO tuning:
МММ
® Patch Diagram
e TERI
tan,
ve:
ямоот (Y
RANGE SYNC RANGE SYNC
o 8 o o RING
® ® MODULATOR
— OUT
' = D : 3 E = of
© ©. | © © |
Г В M ‘о - т ; - Е
4 . N ` . rn 00 =
. m . mu =
L | L | ;
Peer vs
ADSR
am
MANUAL О’ ЕМУ ! OUT
ATTACK DECAY SUS na RELEASE we
TIME TIME LEVE =
© A
| Pitch
+ Tri
cv e Ц
и
EE I EES
Mm
ms
Bord aus er
| 3/0). ju
| ol .
#1)! ВНЕ
slo):
41 Expansion of the Basic Patches
Try the patch of Fig. 2-16 with VCO-2 tuned a minor third
above VCO-1*. In this case, the apparent pitch of the sound
has moved down more than two octaves. If VCO-1 produces
880Hz when a key is struck, VCO-2 will produce 1046.5Hz
(the ratio of frequencies in a minor third is 1:1.18921;
1.18921 x 880 = 1046.5). The sum frequency is 1926.5Hz
(1046.5 + 880) and the difference frequency is 166.5Hz
(1046.5 — 880). Neither of these frequencies fall within the
same scale system as the original input pitches. The upper of
these frequencies is not an even multiple of the lower
frequency, thus it is a non-harmonic overtone of the funda-
mental pitch of the sound and lies somewhere between the
eleventh and twelfth harmonic of the fundamental pitch.
At this point, if the output of VCO-1 is fed to the VCA in
addition to the ring modulator output, the sound will regain
the correct pitch relation with the keyboard because the level
of the ring modulator output is slightly less than that of
the VCO. The difference frequency becomes a non-
harmonic frequency which lies slightly over two octaves and
a major third below the fundamental, and the sum frequency
becomes a non-harmonic overtone somewhere between the
second and third harmonic of the VCO-1 pitch.
In the above example, if it is desired to use only the ring
modulator output, it is possible to tune the VCO's so that
the ring modulator produces the correct pitch relations by
first setting up the sound as shown and then finding what the
tuning discrepancy is and correcting it. First, tune VCO-1 in
the normal manner so it produces the desired pitches in
relation to the keyboard; then tune VCO-2 to a minor third
above. Next, using only the ring modulator output connec-
tions to the VCA, determine which key on the keyboard
produces the pitch of A. In this example, the nearest key will
be the D key. Disregarding pitch shifts of more than one
octave, we can think of the ring modulator pitch as being
approximatly a fifth too low. Next, with only the VCO-1
output connected to the VCA, tune VCO-1 so that it
produces pitches approximately a fifth too high. Do this by
striking an £ and tuning VCO-1 to the pitch of A (A isa
fifth above E). Re-tune VCO-2 so that it is a minor third
above the new VCO-1 pitch. Re-connect the ring modulator
to the VCA and check the pitch produced when A is struck.
If the original discrepancy was not an exact musical interval,
it will be necessary to touch up the tuning slightly. Try
changing either one or the other PITCH control slightly to
bring the sound into perfect tune.
Fig. 2-17 shows a patch in which the outputs of both VCO's
are mixed with the ring modulator output to produce a bell-
like sound. In this patch, VCO-1 is first tuned to the desired
pitch relation with the keyboard. Set the VCO-2 RANGE
switch at 8” and tune VCO-2 to a minor third above VCO-1,
then return the RANGE switch to 4’. Try the sound of the
ring modulator alone first, then add VCO-1, then VCO-2.
Also try different settings of the VCO RANGE switches.
3. First press A and tune VCO-1 to unison with a test pitch or tuning
fork. Next, press F# a minor third below and tune VCO-2 to
unison with the test pitch. Now when A is pressed, VCO-1 will be
at unison with the test pitch and VCO-2 will be a minor third
above.
CALCULATING RING MODULATOR INPUT FREQUENCIES
Musicians with access to a frequency counter and an inclination
to experiment will find the following useful.
The output of the ring modulator consists of the sum and
difference of the two input frequencies. The pitch of the sound
produced will seem to be that of the lower or difference
frequency while the upper or sum frequency will seem to be
that of a dominating overtone. If we assume the difference or
pitch frequency to be the 440Hz of A above middle C, the
input frequencies necessary to produce any given sum or
overtone frequency may be calculated with:
In = sum + 440 “and In, = sum =
EXAMPLE:
We want the sum frequency to be the ninth harmonic which,
for 440Hz, would be 3960Hz (9 x 440 = 3960): therefore:
In, - S900 + 440 = 2200Hz;
and:
In, = 3860 — 480 = 1760Hz.
When the A above middle C on the keyboard is struck, one
VCO should produce a frequency of 2200Hz and the second
VCO should produce a frequency of 1760Hz. The apparent
pitch of the sound will be 440Hz with a harmonic of 3960Hz.
Expansion of the Basic Patches 42
Fig. 2-17 BELL
(a) Block Diagram
VCO-1
ouT
VCF —> VCA
VCO-2 A
Y Y Y
Pitch A — >
CV RM
Gate + Trig
|
| on
(b) Patch Seen 6 — O VCO-1
E
voor} § A | In
Ext TCO OUT)
pwwon ff Ш
o
|
]
4
оооед В || 4 ооо A
a: ep 1e
1
i
#2
) ——— +
ug ge
Ш
ето — @
: que
| MANUAL | MANUAL
НЕ E
RANGE SYNC RANGE SYN
oo:
o ?
%
> ОЧТ
Nm mr EE
AM OUT ==="
RING
MODULATOR
| MICH ICH INITIAL GAIN
E
En
ADSR |
=
NE à
MANUAL GATE ЕМУ 1 007
ATTACK DECAY SUSTAIN RELEASE er и” =
() ME TIME LEVEL | =
: A M
oe a
NA y =
ña O E %
Eom == E”
= EXT GATE 3
i N DE : à EN AA
= : я = КЛ
ыы E.
A - Ma on pn ния mt ate]
= {SS = E | ==8 5 5 VS a | [uma]
E SS ПЫЛИ
HI
| © o o o @ Roland
VV 100 ef Nore? oat yer
43 Expansion of the Basic Patches
If waveforms other than sine waves or triangle waves are
used, the spectrum of the ring modulator output becomes
extremely complex. For example, if we use sawtooth waves,
each of the harmonics of one wave will interact with each
harmonic of the other sawtooth wave to produce sum and
difference frequencies. In other words, the fundamental of
VCO-1 would interact and produce sum and difference
frequencies for each of the harmonics of the VCO-2 saw-
tooth wave. The second harmonic would do the same, and so
on. The result would be a conglomeration of muddled
harmonics which may or may not prove rather useless
musically. The square and pulse waves would produce similar
results because they are also rich in harmonics. The harmonics
in the triangle wave are low enough that the effect is similar
to using sine waves. The presence of these low intensity
harmonics often adds to the quality of the sound. Fig. 2-18
shows how to use VCF's to filter triangle waves to obtain
sine waves. The sound is very similar to that of Fig. 2-17.
Expansion of the Basic Patches 44
Fig. 2-18 BELL Using Sine Waves
Block Diagram
VCO-1 NY VCF-1 oe
A > — OUT
RM —>| VCA
> >LA
VCO-2 LA » VCF-2 dl
Pitch CV
ADSR
Gate + Trig
| A VCO tuning:
© VCO-2
A —
(b) Patch Diagram
— —” OUT
MODULATOR
; ADSR
Pitch CV
Ë MANUAL GATE ЕМУ 1 ОБТ
Ви A ES e KE uE d. es ATTACK DECAY SUSTAIN RELEASE /=
O TIME TIME LEVEL TIME ©) '
EXT GATE
8 Gate + Trig В |
45 Expansion of the Basic Patches
Fig. 2-19 shows how to use the ring modulator to produce a
very effective bowed tremolo string sound. The LFO
frequency determines the tremolo rate. The LFO KYBD
TRIG switch insures that the tremolo will start each time a
key is pressed.
To analyze this sound, assume that the VCO frequency is
880Hz and the LFO frequency is 6Hz. The output of the ring
modulator will be a compound of 886Hz (880 + 6) and
874Hz (880 — 6). Since the LFO frequency is low, the sum
and difference frequencies are very close to each other and
we may think of them as being equally pitched above and
below the 880Hz VCO output. The apparent pitch of the
sound, then, seems to be centered between the sum and
difference frequencies at 880Hz. If the sum and difference
frequencies happened to be farther apart, 500Hz above and
below, for example, the pitch intervals above and below
would not be equal. The sum frequency of 1380Hz (880 +
500) would be approximately a minor sixth above and the
difference frequency of 380Hz (880 — 500) would be
slightly over one octave below.
Expansion of the Basic Patches 46
Fig. 2-19 VIOLIN, Bowed Tremolo
(@ Block Diagram
усо |-> >| VCF me RING VCA OUT
= <a e _
A hal |
G
ate > LEO uy
Pitch
CV
ADSR
Gate + Trig À
® Patch Diagram
VCO
————» OUT
E Pitch CV
MANUAL GATE
В O ATTACK DECAY SUSTAIN RELEASE a
TIME TIME LEVEL TIME (Y
5 Е o "
В Gate + Trig wide :
0009 Roland i
SYSTEM 100m 1 Resor Miedo Wyvern
— = — — = =
6 ol
—
Chapter Three:
Advanced Synthesis
3-1
Mixers and Fixed Voltage Sources
In Fig. 3-1, part (a) shows an example of a mixer as used in a
synthesizer and (b) shows multiple jacks. The functions of
these two are exactly opposite. Mixers are used for combining
various audio signals and/or control voltages, whereas multiple
jacks are used for dividing a source signal or control voltage
so that it can feed several inputs simultaneously. The multiple
Jacks cannot be used in place of a mixer. |f two outputs are
connected to the same multiple jack set, the result will be
what is effectively a short circuit to both outputs which
could cause damage.
The outputs of the mixers shown in Fig. 3-1 (a) will represent
the algebraic sum of the inputs at any given instant. These
mixers provide two outputs: normal and inverted. The
inverted output merely changes the sign of the summed
output voltage. For example, if at a given instant the normal
output is +3 volts, the inverted output will be —3 volts. Or,
if the normal output is —5 volts, the inverted output will be
+5 volts. In many situations, this inverting function can be
guite convenient and important.
Fig. 3-1 Mixers and Multiple Jacks
(@ Model 132 Audio Signal/Control Voltage Mixer
|
MIXER-1
MIXER-2
SYSTEM-100M 4 320)
г 15 NA
MIXER-2
> NA
i
VOLTAGE
PROCESSOR
0
+ VOLTAGE SOURCE
(@ =
OUT
OL
Roland |
(® Multiple Jacks
— VOLTAGE SOURCE
KEYBOARD INPUT OUTPUT
v @—©
Ge
CAE@v—®
6 @v—@
rr Roland |) .6—e—e—oe
Roland Studio System ?® — © — © — ©
SYSTEM-100M 191-J
C'@—@—@—@
MULTIPLE JACK
be
-9—0—0—0O
0—0—0—0O
b
'0—0—0—0
-0—0—0—0
"o—o—o—e
O
)
OUTPUT JACK
O
1 ®—@ |
2 ©— ® = ON
ЛА OFF
Often, when mixing audio signals, it will not matter whether
the normal or inverted output is used, but in some cases it
may be important since signals which are similar but of op-
posite phase (inverted) will tend to cancel each other. Fig. 3-2
demonstrates this. In this case the mixer in (a) is merely
acting as a source of both inverted and normal versions of the
same waveform and the mixing is done at the VCA input.
Start with the VCA SIG IN controls at “0”. First raise one
of them to “5”. Next, slowly raise the other level and note
the decrease in loudness. At some point near “5”, the two
input levels will be the same and the signals will cancel each
other. If the level is further raised, the sound will reappear
because the levels are no longer the same.
Try the above experiment using the same mixer output,
either inverted or normal, as shown in Fig. 3-2 (b). Raise one
of the VCA SIG IN sliders to “5”. Raise the other slider and
notice how the sound becomes louder. This is because the
two signals are in phase with each other so that the voltages
at any given instant are added together.
Advanced Synthesis 48
Fig. 3-2 Adding Waveforms
(a) Cancellation of Out of Phase Waveforms
SLIDER 1—— — >
49 Advanced Synthesis
The mixers shown in Fig. 3-1 incorporate built-in voltage
sources which are useful when it is necessary to bias some
signal or control voltage. Bias refers to the addition of a
fixed value to a signal or control voltage. The “function
control” of Fig. 2-1 can be thought of as merely biasing the
module function at some level after which control inputs are
used to alter that function level. In the tremolo example of
Fig. 2-12, for example, the VCA INITIAL GAIN control
biases the VCA so that when added to the LFO input, the
result is tremolo.
Fig. 3-3 shows how to use a voltage source for biasing the
output of an LFO so that tremolo can be added to a sound
without the need for a second VCA. With the “—10v”
mixer level control at “0”, set the VCA MOD IN level for the
LFO to the approximate depth of tremolo desired while
holding a key on the keyboard down. Next, raise the *—10v”
mixer level until the tremolo effect disappears when a key is
not being held down. This should happen near “5”. With the
level at “5”, a voltage of about —5 volts is added to the СРО
waveform. This pushes the zero line of the LFO sine wave
down to —5 volts, a level low enough to keep the LFO sine
wave from “opening” the VCA when a key is not depressed.
When a key is depressed, the output of the envelope generator
is added to the VCA's control inputs so that the LFO sine
wave is now above the zero. The output of the VCA is
shaped by both the envelope and the LFO wave simulta-
neously.
Advanced Synthesis 50
Fig. 3-3 Tremolo (One VCA)
(a) Block Diagram
Vco >| VCF Na os OUT
Pitch CV A À
Gate + Trig
LFO >
MIX
r
>| ADSR
LT]
( Patch Diagram
VCO
EXT SIG INT
PW MOD 3 Srv: |
HEN Hi sE.
x
| MANUAL |
I~
.
tu 107% В Tu MENTES Ta
ll,
ПИ
Ц
|
OUT
\
> 1
К:
‘Св @—©® :
e
Controls depth
= Le 08 tremolo
УПИ
ALLI
®—
5 @ — rar
o
= mr
à
“EN ES
Pitch CV la
%
ADSR E
MANUAL GATE ENV IOUT
= ATTACK DECAY SUSTAIN RELEASE
= O TIME TIME LEVEL TIME
FREQ FREQ DELAY WAVE FORM OUTPUT
RAN LEV
Е ru EL
0 N я
~ ~
«1
`
1
* 710
EXT GATE
E
=
G
„Ш.
ul |e
.
Es o ]9 KYBD =
= SEES EE тю, on . y =
= or . =
FREQCVIN 1 2 =
Gate Ns 2
ERT
51 Advanced Synthesis
Fig. 3-4 shows how to use this principle of baising to provide
a rapidly repeating trigger to imitate a playing technique
sometimes used with the banjo or xylophone. As an
experiment, start with all the mixer controls at “0”. Raise
the LFO input to “5” and note that the LFO square wave
now passes through the mixer to trigger the envelope
generator. Next, raise the “—10v” level to “5”. This biases
the LFO square wave into the negative region, thus the enve-
lope generator is no longer triggered. Last, raise the gate pulse
input level to “5”. Now when a key is pressed, the envelope
generator is triggered and continues to be triggered as long as
the key is depressed. With the “—10v” slider at “5”, this
represents an input of approximately —5 volts to the mixer.
With the gate pulse input slider at “5”, the +10 volt gate
pulse becomes approximately +5 volts, and when the gate
pulse appears, it cancels the fixed —5 volt level. In this
configuration, the mixer is being used as a gate which turns
some function on or off (in this case, the LFO triggering of
the envelope generator) as the result of an external pulse or
voltage (in this case, the gate pulse).
With the LFO KYBD TRIG switch ON, the square wave
output of the LFO is phase locked to the keyboard gate
pulse to ensure that sound will be produced as soon as a key
is pressed. To demonstrate this, Move the LFO FREQ
control down to “0” and the FREQ RANGE switch to “M”
or “L”, Tap a few keys. Try this again with the LFO KYBD
TRIG switch OFF.
Advanced Synthesis 52
Fig. 3-4 BANJO (Repeating Trigger)
(@ Block Diagram
VCO > VCF »| VCA OUT
Pitch CV A A
ADSR
MILL Á
G
ate > LFO ==.
> MIX
Gate 1 L Г A
Key Key
| ON OFF
Patch Diagram
®
VCO
EXT Tí,
PW MOD
D (0—
_—
| MANUAL |
u
ми
— OUT
AA
© rd
@ —
sd
| К | Pitch CV
| = Жила EOS ER ET GE ES YEE am m“ ` ocre
LFO
LFO ] LEE MIXER —10V
/
FREQ FREO DELAY WAVE FORM OUTPUT
HANGE ru LEVEL
a
fi "| dom
Gate van
TRIG
4
3
:
3
UI > = |
I
3
;
ADSR
ENV=1
MANUAL GATE ENVIO
ATTACK DECAY SUSTAIN RELEASE 2
O TIME TIME LEVEL TIME y
1" каво Су м
1 2
SFO OUT
(It may be necessary to move this down
to about '8”' to cause repeated triggering.)
i
9 E Roland EN
оба! Уля Syvem
53 Advanced Synthesis
The remainder of this section on mixers will be concerned
with the invert function. Fig. 3-5 shows how to produce an
effect sometimes called glide where each note played is
preceded by a ‘slide’ upwards from some predetermined
lower pitch. This is particularly effective with human
generated sounds such as imitations of the human voice or
whistling. The output of ADSR-2 is a short pulse which
jumps up to +10 volts when a key is pressed. Inverting this
makes it a —10 volts pulse which, when controlling a VCO,
causes the VCO pitch to jump quickly down, then slide back
up to its normal pitch. The DECAY TIME of ADSR-2
determines the amount of time required for the glide, while
the VCO MOD IN level from the inverter determines the
starting pitch of the glide. The patch of Fig. 3-6 is exactly
the same as Fig. 3-5 with the exception that the internal
inverter of the envelope generator is used instead of an
external inverter.
Advanced Synthesis 54
Fig. 3-5 WHISTLING (Glide)
(a) Block Diagram
VCO —> VCF >| VCA OUT
Pitch CV N AAA \
ADSR-2 ADSR-1
|A wo HW |
Gate
Trig
|
(b) Patch Diagram
—— OUT
LFO - ADSR-1
= MANUAL GATE ENVIROUT
FREQ FREQ DELAY ~~ WAVEFORM OUTPUT Ei O ATTACK DECAY SUSTAIN RELEASE @
o т
о
RANGE EVE
; „ “о 2 E TIME LEVEL TIME
J . ~ ~ aa т EXT GATE M ”
м . s J = @
1 y
' . * 0 :
KYBD Eje
TRIG | a ate
onl Te = Le ?
. ML" FREO CIN ! ?
Pitch tro OUTÍ
rre
e
cb
INVERTER
O
<
MANUAL GATE
ATTACK DECAY SUSTAIN R
TIME TIME LEVEL
Gate + Trig
qm Em mE EE Ee.
z
=
>»
=
<
6:
En
55 Advanced Synthesis
Fig. 3-6 WHISTLING
Pitch ;
CV |
Ë
Es
This patch is essentially the same as Fig. 3-5.
LFO
FREQ FREQ DELAY WAVE FORM DUTPUT
RANGE LE
in O 1
Gate + Trig
SEED es Roe BEE he de He E E
ee NN do O
>
ADSR-1
EXT GATE
O
®
а
м
MANUAL GATE
ATTACK DECAY SUSTAIN RELEASE ñ
TIME TIME LEVEL TIME
MT
ENVIE OUT
—0UT
1
~
Pr
©-
O
EXT GATE
$
=
ADSR-2
MANUAL GATE
ATTACK DECAY SUSTAIN RELEASE
TIME TIME LEVEL ПМЕ
НД
ENV 20UT
[O}
Pot
]
fol ns mam
Is
la
Ez
| E t
20!
010
Keo!
3 Ea Ï
~ E
Advanced Synthesis 56
\ Fig. 3-7 shows how to use the above principle as a better way
to produce delayed vibrato effects in those synthesizers with
LFOs which do not include the delayed output. Since the
VCA-2 INITIAL GAIN control is at “10”, the VCO receives
the LFO sine wave continuously. Pressing a key triggers the
envelope generator. Since the inverted output of ADSR-2 is
being used, a negative pulse occurs which counteracts the
VCA INITIAL GAIN control and “closes” the VCA. The
DECAY control determines the time required for the ADSR-2
voltage to rise to zero again, “opening” the VCA after a
delay.
Fig. 3-7 Delayed Vibrato (Improved)
To experiment with this patch, use the violin sound of Fig. 2-2. To VCO
(The three elements of this diagram would replace the LFO of Fig. 2-2). vibrato input
LFO VCA-2
- „В
a + . à
em
AACR FREQ DELAY ba FORM by df
Pitch CV Te 2. о-@® @=
to VCO row
——
Controls delay time
\ ADSR-2
ОН ся
Ва
с
2
=}; .
MOD м
Gate + Trig AS . | : Durs EEE me &
J fe chon sien
o:
3
Y b
Blob
$0):
= Roland
Aor? Ym Sy vier
||
|
1+0 ! O
> 10
nm
'
57 Advanced Synthesis
In the patch of Fig. 2-4 where the square wave output of the
LFO is used to generate trills, the use of the LFO KYBD
TRIG function causes the trills to start on the upper of the
two pitches. If the square wave is inverted as shown in Fig.
3-8, the result will be a trill which starts on the lower pitch.
Fig. 3-9 shows how to make the trill sound a little more
natural by using the inverted envelope generator output to
control the frequency of the LFO. When the inverted
envelope jumps down to its negative value, the frequency of
the LFO falls. As the inverted envelope rises, the frequency
of the trills will increase. This imitates the style of playing
where the first two or three notes of a long trill are slower
than the remaining notes.
Fig. 3-8 Trill (Improved)
(a) Block Diagram
VCO —» To VCF
Pitch ПЛ. Ш
a — >| ЦРО
Gate
—» To ADSR
(В) Patch Diagram RANGE sr
Controls 4 * a
trill O
rate .
LFO ®
д ,
FREO FARO DELAY WAVE ess UT - +
RANGE ku m ||
He
Ma) FREO CV IN
E
INVERTER |
interval
control
1 2
tLFO OUT
Advanced Synthesis 58
Fig. 3-9 Natural Trill
(@ Block Diagram
VCO (——— To VCF
Pitch CV
№
ADSR-2
A
Ш
(b) Patch Diagram
Gate + Trig табов.
>» lo -
ENV-2
MANUAL GATE ENV 20UT
ATTACK DECAY SUSTAIN RELEASE On
O TIME TIME LEVEL TIME
EXT GATE
ne Cor trols
= | amount of
rate
frequency | © > @_ тонн
chage
Controls MICH |
trillN LFO AS a
Trill
” interval
INVERTER Te ml
|
| asi
000
| efe
.
. = я
- po E a =
- pe | =
3 Es bs a
a o io o e a e ME E
Piteh CV
«Su st E 533 ES EE НН О БШ
BE _
6000 i= Roland
Sym ОМ 181 Конка? Мн Зум
—]——— — =
МММ
59 Advanced Synthesis
The Sample and Hold
The sample and hold (S&H, or S/H) can be used to produce
random or patterned sequences of control voltage. Fig. 3-10
shows a patch for producing musical patterns.
Clock Out ||]
LFO
(b) Patch Diagram
Y
Fig. 3-10 Sample and Hold Patch (Arpeggio-like Patterns)
(a) Block Diagram
VCO
S/H
>» VCF
Input
Selector-7 я мол ©
Switch
NOISE
FREQ FREQ DELAY
NGE
ro || B И
* FREO Cv IN
®
EXT CLOCK IN
va FORM = LA
VEL
LLFO OUT
5 Roland |
|
(Internal
Connection)
MANUAL GATE
ATs DECAY SUSTAIN ELL.
TIME LEVEL
DI
Tn
|
®® €
OUT
Fig. 3-11 shows how the sample and hold works. At the left,
the clock circuit generates short, sharp pulses at regular
intervals which are used to control the gate. Each clock pulse
opens the gate for an instant, thus allowing the hold circuit
to “see” what the voltage level of the waveform is at that
particular instant. The hold circuit “sees” this voltage and
‘’remembers’’ it until the next time the gate opens. The clock
circuit also generates a square wave of the same frequency as
the gate pulses so that other synthesizer elements can be
triggered in synchronization with the timed voltage samples.
Advanced Synthesis 60
From
LFO
From
Noise
Generator
Fig. 3-11 Sample and Hold Block Diagram
ИИ
alll
= |
GATE
EXT
input
Input Selector
Switch
— OZ — — ——
o
A
> — Ш
CLOCK
Clock
Rate
Control
Y
Clock
Output
JUUL
HOLD
eV
Output
Lag Time
Control
61 Advanced Synthesis
Fig. 3-12 shows more clearly the relations between the
different waveforms in the sample and hold circuit. Also
shown is the musical representation of the pattern produced
if the sample and hold output is used to control the pitch of
a VCO. The actual frequencies of these pitches will depend
on the voltage levels at the time each sample is taken and,
more often than not, these will not fall within conventional
musical scale systems. This is not a disadvantage, however, as
the resulting sound pattern can add beautiful tonal effects
and coloring to music built on a conventional scale. By
changing the frequency of the input waveform and/or the
rate at which the samples are taken (CLOCK RATE control),
the patterns may be varied. Fig. 3-13 shows a few possibilities.
Almost all keyboard controllers used with synthesizers
contain a built-in sample and hold circuit. The keyboard gate
pulse is used as the clock and the sample input is the voltage
generated by pressing a key. Each time a key is pressed, the
sample and hold gate opens so that the hold circuit will
“see” the voltage level of the key being pressed. When the
key is released, the voltage level returns to zero, but the hold
circuit holds the level it previously “saw” until a new key is
pressed and the gate opens again.
Fig. 3-12 Sample and Hold Waveforms
LFO
| | | | | | | | | | | | | | In this example, the fre-
Clock quency of the clock out-
Out put is a multiple of the
LFO sawtooth.
Gate
Out
(Sample)
| | | | | | Musical Pattern:
T Advanced Synthesis 62
| Fig. 3-13 Sample and Hold Patterns
Below are a few of the patterns possible when sampling the LFO sawtooth wave.
© ©
Sampled Sampled
Waveform Waveform
s/HClock | | 11111 S/H Clock | | |
e > Y
PA
Gate L | E | > 1 Gate a ———
Output a 1 1 зе | Output -r—— TT |
| cv A ar CV
Output Output ea.
Melodic | Melodic - a Г
Pattern 1 1] Pattern — .
® @
Sampled Sampled
| Waveform Waveform
неее ИИ нее ИЦ
| Сате Сате
SE i … Ouen
Cv CV
Output Output
Melodic |
Pattern
Melodic
Pattern
63 Advanced Synthesis
Portamento acts like a low pass filter on the keyboard
voltages changes. When it is used, the keyboard control
voltage output will slide from one level to the next when
different pitches are played. The sample and hold LAG TIME
control has a similar effect on the sample and hold control
voltage output, as shown in Fig. 3-14.
Fig. 3-14 Sample and Hold Output Lag
— |
“OR! LAG
"2" LAG
“8" LAG
“10” LAG
The Noise Generator
The noise generator is generally used for effect sounds such
as wind, surf, and whistling, or non-pitched sounds such as
the wood block sound of Fig. 3-15. Since VCF resonance
accents the band of frequencies around the cutoff point of
the filter, it is possible to play melodies by using the keyboard
pitch control voltage to control the cutoff point of the filter.
This may be done by raising the VCF KYBD MOD IN
control to “10” in Fig. 3-15. Adding this voltage to the VCF
cutoff point raises the cutoff point too much, so this must be
corrected by lowering the CUTOFF FREQ control to 2’ or
3”, Now melodies can be played.
Advanced Synthesis 64
Fig. 3-15 WOOD BLOCKS (Noise Generator)
(@ Block Diagram
VCF OUT
Y
NZ
(Optional Pitch Control)
>
о
O
D
ei OUT
| =
IE SELF
ze — ‚=
| el
To produce “pitched”’
white noise, raise this
control to 10” and lower
the cutoff freq. to about
2’ ог 3”.
= ®
|
т
E a
т
—-— Canon?
ADSR
ED
MANUAL GATE ENV ‘OUT
ATTACK DECAY SUSTAIN RELEASE i
O TIME TIME a
EXT GATE Fi
© PARA EE mm
va. °
— o Ши Ш Ms pai
sta E Er rs phe, o ee .
pes её : a a ait E тео
В |
e
VW
0 000 == Roland
je re Mentes dy vio”
65 Advanced Synthesis
Since noise is a random combination of all frequencies, the
sample and hold may be used to sample the noise waveform
for producing random voltage patterns. In Fig. 3-10, random
pitches may be produced by simply changing the sample and
hold input selector switch from the LFO position to the
NOISE position. In Fig. 3-16, noise is used to produce
random and rapid changes in tone color when sustained
pitches are played. If the optional patch cord is used, each
note played will be of a different tone color because the key-
board gate pulse is used as the clock to trigger the sample and
hold function. Fig. 3-17 shows how to use the output of an
envelope generator so that the sample and hold clock can be
triggered with the keyboard GATE + TRIG output. With all
the control at “0”, the output will be a quick sharp pulse
whenever a gate or trigger pulse appears at the input. This
short pulse is enough to act as a clock pulse for the sample
and hold.
Fig. 3-16 Random Tone Color Modulation
(a) Block Diagram
VCO > VCF OUT
| AA
Pitch CV
NZ S/H ADSR |
Optional) A A
quo i a om; em ome om ty ii. À
Gate + Trig
Advanced Synthesis 66
(b Patch Diagram
NOISE S/H
EXT SIG CLOCK SAM
q | CLOCK HATE | AG TIME
e |”
2 + |3 ;
— OUT
E
©
EXT CLOCK IN
Pitch CV
чин ся вып Бин БЕ En mm Ems EP Em mm mm mm шт Ÿ
Optional patch cord
to S/H EXT clock in ADSR
jack.
"О САТЕ ENV ! OUT
ATTACK DECAY SUSTAIN e
TIME LEVEL
Gate + Tri
|| Oo oo 0 E Roland
. 8 SAONE
Fig. 3-17 Gate + Trig for Sample and Hold Clock Input.
To sample and hold
EXT clock in jack.
MANUAL GATE ENV OUT
ATTACK DECAY SUSTAIN RELEASE
O TIME TIME LEVEL TIME
EXT GATE
e ol}
Шин We MS Em Es => E
Gate + Trig
69 Advanced Synthesis
Surf may be produced by using the rushing sound of pink
noise, as shown in Fig. 3-20. With surf, the roll of the waves
should be a little more regular than the changing patterns of
wind; therefore, the LFO sine wave output is used as the
basic source of control. The frequency of the LFO, however,
is controlled by the random output of the sample and hold
so that the pattern of waves does not become too regular.
Fig. 3-21 shows a stereo surf patch.
Fig. 3-20 SURF
(@ Block Diagram
Pink
NZ | White
S/H
Y
Y
VCF — OUT
LFO
(b) Patch Diagram
NOISE
SE
a
LFO
FREQ FREQ DELAY WAVE FORM OUTPUT
RANGE mu LEVEL
= 3
= _
=> OUT
Advanced Synthesis 70
Fig. 3-21 SURF in Stereo
(a) Block Diagram
NZ-1
NZ-2
VCF-1 » LEFT OUT
® Patch Diagram
LFO
» \/CF-2 — RIGHT OUT
INV
NOISE-1
NOISE
GEN
1 O
WHITE PINK
==
__WVCF-1 | VCF2
svstem-00m 124 OM
дя EES Em END EN A EN y
ace FREQ DELAY
: ii ©.
«vec
"TRIG
ve FORM
* rREÉNEV 1
71 Advanced Synthesis
3-4
The Gate Delay and Pulse Shaper
In the System 100M, the gate delay portion of the Model
172 module (Fig. 3-22) provides both gate delay and pulse
shaper functions. The gate delay function is used when it is
desired to delay a pulse and/or to change the length of the
pulse “on” time. The pulse shaper is used where it is necessary
to take a rough pulse such as might be recorded on tape and
reshape it into a form more useful in the synthesizer.
Fig. 3-23 shows how the gate delay function can be used to
help produce a wolf whistle. The gate pulse from the
Fig. 3-22 Gate Delay Portion of the Model 172 Module
(Complete module shown in Fig. 4-9)
GATE
IN
Le
keyboard triggers the first half of the whistle and the delayed >
output from the gate delay triggers the second half. Fig. 3-24 GATE DELAY
(a) shows the waveforms for the gate delay. The DELAY THRESHOLD RIDE ÉATETIME
TIME control determines how much time will elapse between 6 5 6
the start of the input pulse and the start of the output
pulse. The GATE TIME control determines the length of the
output pulse.
0 10 10 ( 10
Fig. 3-23 WOLF WHISTLE (Gate Delay)
(a) Block Diagram
PALIN De
VCO —» VCA > »> OUT
A
Pitch CV
ADSR-1 ADSR-2
A A
— Gate SL _|GATE |JL
DELAY
UE
MANE
Advanced Synthesis 72
VCO VCA
(b) Patch Diagram
win
— OUT
Pitch CV
E has SE SE DES ES ES ES a m?
ADSR-1 ADSR-2
ENV-2
(VE
MANUAL GATE
y MANUAL GATE ENV 2 OUT
AY TACK OECAY SUSTAIN REL puse 3 ‘
||
О ATTACK DECAY SUSTAIN Pe AA
ME ME
5 8 ` =
A : > LEVEL
"© GATE UE Es . on .
® De Sn se 2 © @ |<
ne = >
= 6 DE LAY ВЕ 3
= THRESHOLD DELAY TIME GATE TIME ll le
а | A .
ms e UE Sp
5
= rT
EES e
i ооо Roland
MEM я Дожей Ма Ц
E
Sometimes it is desirable to synchronize some device such FIG: 3.28: Gate Delay Pulss Shaper Wavstoriis
as an analog sequencer (discussed later) by triggering it from
(@ Gate Delay
pulses recorded on tape. Because of the nature of the
recording medium, it is possible to record only very short na
clicks on tape to act as trigger pulses. The pulse shaper —
function lengthens these pulses so that they are more useful. eyo]
Also, the output level of a tape recorder is usually quite low ve
in comparison with normal synthesizer levels. The input of Output
the gate delay has an amplifier stage which brings this level а Po
up to where it will trigger the gate delay. The THRESHOLD | Time |
control determines the level of input which will trigger the ® Pulse Shaper
gate delay. Normally, set the THREHOLD control at “10”, E. | |
then turn it slowly down towards “0” while inputting pulses to Tape
from the tape. There will be a place where the gate delay is
triggered for each input pulse. Above this setting, the gate Tape
delay will not be triggered and below this the gate delay may Output
be triggered by random tape noise. The GATE TIME control Pulse
should be set somewhere above “0”, since at “0” the gate
pulse may be too short to be noticed. The DELAY TIME Gate
, ; Delay
control should, of course, be set at “0” unless a delay is Output |
desired. The pulse shaper waveforms are shown in Fig. Se
3-24 (b). Time
73 Advanced Synthesis
3-5
Controllers
In electronic music the term controller can be applied to any
device which is used to control another device. The envelope
generator is a controller because it is commonly used to
control other devices, such as a VCA. When a VCO is used to
modulate a second VCO, the first VCO becomes a controller.
The sample and hold is a controller. The keyboard controller
could perhaps be considered the most common controller
used with the synthesizer. In this section we shall briefly
consider a few other types of controllers.
Many synthesizer keyboards include an auxiliary controller
which is usually called a bender. Fig. 3-25 shows a bender.
With the bender at its normal center-off position, the output
at the BENDER output jack is O volts. If the bender is moved
to the right, the output voltage will increase to a maximum
of +5 volts. Movement to the left produces a negative voltage
which can go as low as —5 volts. One of the more common
uses of the bender is to connect the BENDER output to a
VCO control input to create manually controlled pitch
bending effects. The bender could also be connected to a
VCF for manually controlling tone color, or to a VCA for
creating accents. There is a similar type of controller called a
ribbon controller which consists of a length of metallic film
which outputs a control voltage when touched with a finger.
The level of the voltage will be proportional to the distance
between the finger and the left end of the ribbon. If the
finger is placed at the left, then moved towards the right,
the control voltage output will slide from a low value to a
high value,
Fig. 3-25 Bender Lever
VCO
®
AE EE SRR,
} |
dug
mir
the maximum pitch
The setting of this
4 control determines
| deviation when bending.
tes |
© © ®
= ME D IN —®
САТЕ IG BEM BENDER
CIC
SYSTEM-100M 181
fa OUTPUT —— J
| 4” yk
~ ~ Ne
(E 1)
A A A A
~\ ~~
0 10
TUNING PORTAMENTO
С 05
MH ON
THANSPOSE
——— (| — — = +
BENDER
В | !
Bender Lever
The joy stick controller can be thought of as a two-
dimensional bender lever which can be moved to the left
and right (“x" axis), or backward and forward (”*y” axis),
or in any combination of these directions, as shown in Fig.
3-26. Fig. 3-27 shows a three-dimensional joy stick which can
also be moved up and down (“z" axis). Each axis can be
assigned to control a specific parameter. With the three-
dimensional joy stick, it is possible to simultaneously control
the pitch, tone color, and loudness of a sound.
Advanced Synthesis 74
Fig. 3-26 Joy Stick Controller
==
PS
U
Fig. 3-27 Three-Dimensional Joy Stick Controller
Z
Zz
<
ZZ
U
75 Advanced Synthesis
The foot volume control pedal (such as the Roland FV-2)
can also be used as a synthesizer controller. Fig. 3-28 shows
how to connect such a pedal to form a foot-controlled
variable voltage source which can be used to control pitch,
tone color, or loudness. Fig. 3-29 shows how loudness
(dynamics) can be thus controlled. Tone color can be
controlled by substituting the sample and hold input to the
VCF of Fig. 3-16 with the voltage output from the foot
pedal.
Fig. 3-28 Using a Foot Volume Pedal as a Controller
Foot controlled
variable voltage
source output
Evo
Output
Jack
—
CAUTION: Do not connect
voltage supply or battery « В
to output jack. BOSS Г É я
FV-200
Input Jack
EE ° ll 10 | or—
Altenate voltage supply:
©
vOLTAG
PROCESSOR
;@ Q ¿ei -8
®
9 Volts Transistor— — >
Radio Battery
9 volts
Disconnect battery when
not using.
Advanced Synthesis
76
Fig. 3-29 Foot Control of Dynamics
(a) Using Two VCA's
(b) Using One VCA
VOLTAGE
SOURCE
sour
- Ом!
From
VCF
From
ADSR
© 70—0—0—0
-© :0—0—0—0
© *0—0—0—0
INITIAL GAIN
SYSTEM-100M 1308)
ep |
— OUT
Output Jack
—— OUT
Set as
needed
Control Voltage
from
Foot Pedal
Instrument
(Input)
Jack
FV-2
77 Advanced Synthesis
The analog sequencer is another common type of controller
used with the synthesizer. Fig. 3-30 shows a two-channel Fig. 330 ‚Analog Sequencer Pans| Diagram
sequencer. The voltage registers determine the voltage output
for each step in the sequence. When the series output is used, system 100m 1820) |
it is possible to produce a sequence of from one to sixteen TEMPO
steps. When the parallel output is used, it is possible to Y SD a
produce two voltage outputs for each step (for two-note WA A
chords, for example) with from one to eight steps. Fig. 3-31 PORTAMENTO
shows a simplified block diagram for the sequencer. The YA rN O:
clock is an LFO which generates a square wave. The series of ©) ©) ~~
blocks labeled “GATE” form a circuit called a ring counter. de aly —
Only one of the gates will be open at any given instant. Each Prot (5
time a clock pulse appears, the ring counter will shift one _ von was
place to the right. For example, if GATE 2 is open, the Ka IA Ш |
voltage source will pass through the gate and through the 0 ¿
STEP 2 voltage register, then to the output. When a clock STEP NUMBER
pulse appears, GATE 2 will close and GATE 3 will open, thus > < 5 Ze не С Jomo
the voltage at the output will change so that it corresponds 0 sun Zo
to the setting of the STEP 3 voltage register. If the clock M O
oscillator is allowed to run continuously, the ring counter Dr Rance
will run in circles so that the sequence repeats itself. The Or 3 h В |
most obvious use for the sequencer would be to run a 7 “ CH1 CH2
melodic pattern such as an arpeggio, in which case the CV 0 ©
output would be used to control a VCO and the clock = re o o
output would be used to trigger an envelope generator. PA, AL GATE TEMPO
DW e
о e Roland |
Voltage
Registers
Fig. 3-31 Analog Sequencer Simplified Block Diagram
+ Voltage Source
Y Y \ Y
GLOCK NI = GARE = a = и ET, — 2 pe a.
OSC.
Step Step Step Step Voltage
1 2 3 8 Registers
y | | Lo
CLOCK cv
OUTPUT OUTPUT
The control of a synthesizer by means of a computer gives
the musician complete and absolute control over the music
being produced. There are two basic approaches to computer
control. The first is to use a computer especially designed for
the control of a synthesizer, such as the Roland MC-4 Micro-
Composer (Fig. 3-32). The second method is to adapt an
ordinary computer to this type of control. This requires
intimate knowledge of computer programming, and usually
involes the use of custom-made interface devices for con-
verting the output of the computer into control voltages
and pulses for controlling the synthesizer.
Advanced Synthesis 78
Fig. 3-32 Roland MC-4 MicroComposer (Computer Controller)
9 Г
>>
>
gy
">
Chapter Four:
Accessories Used in Synthesis
4-1 Introduction
The voltage controlled filter, being a dynamic filter, is most
often used for controlling changes which occur in tone color
during the production of notes. It must be remembered that
most sound sources also contain tone color elements which
remain relatively constant, such as those introduced into a
sound by the body of a musical instrument. This is why, in
trying to reproduce many kinds of sounds, many synthesized
sounds seem inadequate unless one is striving for purely
electronic sounds. Imitation of acoustic instruments, whether
real or imaginary, often demands more than just the syn-
thesizer elements discussed so far.
Most of the devices to be discussed in this chapter are used in
conventional music studios as effects devices, or as correc-
tive devices to compensate for some fault in the recording
chain. In electronic music, of course, these devices can serve a
similar purpose, but in this chapter we shall look at these
devices primarily as aids to synthesis.
4-2 The Graphic Equalizer
In the conventional music studio the graphic equalizer is
used, more often than not, as a corrective device. Most
literature on the subject seems to take a justifiably conserva-
tive attitude on their use, the idea being that if we have to
use equalization to make a flute sound like a flute, something
must be wrong somewhere in the recording chain. In some
cases, equalizers can be very useful. As one example, equali-
zation can help to clarify instrumental voices in a “muddy”
passage, but still it is very easy to contract “equalization
fever” in which equalization is applied to everything abun-
dantly and with abandon so that the recording no longer
resembles what was originally desired.
In electronic music we may, more or less, ignore the
conservative attitude of conventional music recording studios
because the graphic equalizer is a very important tool in the
process of synthesis. It is so important, in fact, that the
graphic equalizer can be considered second only to a rever-
beration unit on the list of priorities for accessories for
the new electronic music studio. Nevertheless, “equalization
fever” is still easy to contract and care should be taken not
to overdo it. Too much of a good thing is still too much.
Accessories Used in Synthesis 80
Fig. 4-1 Graphic Equalizers
(@ Boss GE-10 (10 Bands)
31Hz 62H: 17477 | 250H: S00He 1KHz 2KHz 4KHz B KH: 16 KHz
Graphic
tay SE ag MT] din os ni аа LT
uN 10 0 0 “00 “IN “e UN 2% y 3 * e r TC CE LOC 2.4% AK BD OK E.
81 Accessories Used in Synthesis
Fig. 4-1 shows some graphic equalizers. The name graphic
equalizer comes from the fact that when the sliders on the Fig. 4-2 Graphic Equalizer Block Diagram
front panel are set, they give a graphic representation of the
frequency response produced by those settings. The block > BAND
diagram of Fig. 4-2 shows that this variable frequency 1
response is obtained by means of a group of band pass/reject
filters (pass or reject, depending on whether the slider is up
or down) in parallel. BAND
y
Fig. 4-3 shows how a graphic equalizer can be connected to
become a part of the synthesis chain. Often it makes little
difference whether the equalizer comes before or after the BAND
final VCA, but if the higher frequency bands are set for high a 3 >| MIX
levels, the equalizer will have a tendency to generate more `
noise than normal, thus the arrangement in (a) may prove
better. Fig. 4-4 shows four sample settings. These settings can BAND
be used for improving the quality of the string sounds of Fig. 4
2-2. |
Input
Output
Y
Y
e
y
A parametric equalizer is an equalizer in which the center Other
frequency of each band and the band width are adjustable, > BAND Bands
as well as the frequency levels. With the graphic equalizer we
must be satisfied with adjusting the level of the frequency у
closest to the desired point, whereas with a parametric To Other
equalizer, it is possible to fine tune each band to exact Bands
Y
frequencies.
Fig. 4-3 Equalizer Connections
(a) Before the VCA
(Switch set
for OdB)
From |
VCO — > VCF
In this case, the input level to the НН +
VCF should be reduced to prevent |
over-driving the equalizer. With the 9
GE-810 or GE-820, this is not ces mM] From
necessary if the +20dB inputs are Envel
used. BOSS GE-10 nvelope
Generator
(b After the VCA
Use Low (Switch set
/ Output for OdB)
From y OUT
VCF VCA | | (To Amplifier)
Prom [| == a
Shyslope BOSS GE-10
Generator
i.
Accessories Used in Synthesis 82
Fig. 4-4 Typical Equalizer Settings for String Sounds
(@ Violin
b) Viola
(Change VCF CUTOFF
FREQUENCY to “6”.)
(© ‘Cello
©) Bass
2 BKH: 16KHz
EQUALIZER
| NORMAL
HKHz 16KHz
| 55055
i EQUALIZER
NORMAL
GE-10
В EQUALIZER
NORMAL
EQUALIZER
NORMAL
83 Accessories Used in Synthesis
The Vocoder
In 1939, H. Dudlay announced a band width compression
device for use in telecommunication systems. The present
day vocoder is based on the principle of this device.
The vocoder requires two inputs: the carrier input (sometimes
called the replacement or excitation input signal) and the
program input. The program input is sometimes called the
speech input because it often consists of spoken or sung
words. This input is broken down and analyzed by the
vocoder circuits, then reconstructed using the carrier input
for raw material. The result is that the carrier will take on
the qualities of human speech or singing while retaining the
original pitch and tone color character of the carrier. For
example, if the carrier input is a recording of a dog barking,
the dog will seem to talk. If the carrier input consists of
recorded street noises, the traffic will seem to talk.
Fig. 4-5 shows how the vocoder works. The vocoder circuits
consist of two major sections: the analyzing section and the
synthesizing section. The program or voice input is analyzed
by passing it through a series of band-pass filters, as shown in
the analyzer section on the left side of the diagram. The
envelope follower is a circuit which outputs a control voltage
whose level is proportional to the level of the audio signal at
its input. The output of the analyzer section is a set of
control voltages whose levels are proportional to the levels
of the signals being passed by the band-pass filters. As an
example, the sound [ee] consists primarily of higher frequen-
cies, thus the control voltage levels will be higher for the higher
frequency bands than they will be for the lower bands. On
the right side of the diagram, the synthesizer section of the
vocoder also contains a series of band-pass filters which are
used to divide the carrier input into separate bands. The
outputs of these filters are passed through VCA's, then
mixed for the vocoder output. The control voltages from the
analyzing section are used to control the VCA's in the
synthesizer section. With the [ee] sound, the control voltages
from the analyzing section that represent higher frequencies
will be higher than those control voltages representing lower
frequencies thus opening partially or completely those
synthesizer section filter VCA's associated with the higher
frequencies. By this means, the original voice is reconstructed
using the tone color material from the carrier input. The
sound of the instrument used as the carrier input will seem
to speak or pronounce the [ee] sound. It can immediately be
seen that except for special effects, the carrier input must be
a sound rich in harmonics, otherwise there will not be
enough raw material from which to completely reconstruct
the program input. Also, the pitch of the carrier must be low
enough to provide the low frequencies which are needed to
build the new sound.
Accessories Used in Synthesis 84
Fig. 4-5 Vocoder Block Diagram
PROGRAM
(MICROPHONE) CARRIER
INPUT INPUT
! |
Г | == ee НЗ БО 1 a A US RS E
| Eo
| |
| | |
| | |
| | > BPF vos
|
| o
| > BPF > EF —
| | |
| ||
| | | > BPF |
| No
Ot BP > EF — | |
| l |
| oc] >| BPF
| | |
| | | —
| > BPF > EF >
| | >
| | = ва a VOCODER
| | From § == —> po OUTPUT
| | | other ¢ _ _ — |
| | bands | — — — > |
| —
|] |] |
|! oa] |
o] Noa |
To other | | To other |
| bands | | bands |
| | | |
| ANALYZER SECTION | | SYNTHESIZER SECTION |
ee EN a a ey Se J = aps a Ea EE EEE
HPF = High Pass Filter (See text)
BPF = Band Pass Filter
EF = Envelope Follower
VCA = Voltage Controlled Amplifier
85 Accessories Used in Synthesis
Most carrier inputs will not contain enough of the higher fre-
guencies to reproduce accurate consonant sounds such as the
hissing of an s sound. The high-pass filter at the top of Fig.
4-5 is set so that it passes these higher hissing sounds directly
to the vocoder output, thus producing speech effects in
which the words are more easily understood.
Phase Shifters and Flangers
Phase shifters and flangers (audio delay lines) are more often
thought of as effects devices, but they can sometimes be
used in the synthesis of sounds with a complicated spectrum.
To better understand them, however, we should first look at
them as effects devices.
The first such effects produced in recording studios were
produced by feeding an audio signal to two tape recorders
and mixing the outputs from the playback heads. When a
light finger pressure is applied to the supply reel on one of
the machines, it will run slightly slower, thus giving a slight
delay to one of the outputs. When this slightly delayed sound
is mixed with the non-delayed sound, the result is that
certain frequencies will be cancelled. The effect is similar
to the effect which can be heard when standing near a
runway as a jet plane takes off. Certain frequencies are
cancelled as the direct sound from the plane is combined
with the delayed sound reflected from the runway. The
amount of delay changes as the plane moves off into the
distance. In the studio, this effect is known as flanging
because it was first produced by finger pressure applied to
the flage of a tape reel.
In earlier attempts to electronically imitate flanging effects,
electronic phasing systems were used. In these, the sound is
passed through a series of phase shift networks, then com-
bined with the original sound. The number of cancellations
which occur due to this combination depends on the number
of phase shift networks used. The frequency ratios between
the cancellations are a result of circuit constants and retain a
constant harmonic relation when the cancellations are shifted
up and down. The result, then, is not the same as flanging,
but in more recent times phasing has become accepted as a
very useful effect in its own right.
Modern digital technology has made it possible to design
solid state delay lines which more closely resemble the
original tape flanging effects.
The block diagram in Fig. 4-7 can represent either a phase
shifter or a flanger, depending on whether the block in the
center of the diagram is a phase shift network or an audio
delay line. The shift/delay control is usually a low frequency
sine wave oscillator. In the electronic music studio, the most
useful units for synthesis of sound are those which allow
optional external control of the function by means of an
external control voltage. Fig. 4-8 shows such units. The three
jacks near the bottom of the panels are for the external
control input. Fig. 4-9 shows the phase shifter and audio
delay designed for the System 100M. At a quick glance,
phase shifters and audio delays like these seem to be exactly
alike, but they are different externally in that one has a
SHIFT FREQUENCY control while the other has a DELAY
TIME control. In synthesis, they would be patched into the
system in exactly the same manner.
Accessories Used in Synthesis 86
Fig. 4-7 Block Diagram for Phase Shifter or Flanger
Input
“ Wi 0
“añ фо
Resonance
Control
Phase Shift
Network
or —_
Audio Delay
Line
A
Shift
or
Delay
Control
OUT
87 Accessories Used in Synthesis
(a).
Fig. 4-8 Roland System 700 Phase Shifter and Audio Delay
Model 720B
2-Channel
Phase Shifter
(b). Model 721A
2-Channel
Audio Delay
“ CH-1
BYPASS
SHIFT FREQUENCY
/ \
300 6KH
MOD INTENSITY
on (О) он
INVERTER
out
2CH PHASE SHIFTER
CH-2 out
BYPASS
SHIFT FREQUENCY
IK
SWEEP SPEED
\ ,?
05.
© su
017
0037 NOR
Roland 7208
2CH AUDIO DELAY
CH-2 wr
I |
BYPASS O
„=. ON
JM Off
RESONANCE RESONANCE
5 5
> 1 ‘ de : + | 4 у,
” ”
(O): O)
/ \ y
0 10 10
DELAY TIME DELAY TIME
7 ?
\ \
7 >! =}
NS = A / \
5 (би 5 0 5!
MOD INTENSITY
ON (O) ="!
INVERTER
77 ~~ #7 - N #7 =N
O—O—O
NS — Ду Bee A WN — A
SWEEP SPEED
©
Roland 721A
Accessories Used in Synthesis 88
Fig. 4-9 System 100M Phase Shifter and Audio Delay
Sweep
oscillator
for phase
and delay
effects
srren-00m 475:
IGN
EXT CV EFFECT pe
1 06
an
MOD INTENSITY SHIFT FREQ RESONANCE
[AUDIO DELAY | [AUDIO DELAY | DELAY
SIGN
EXT CV EFFECT Je аЬ
re Jon ® @
MOD INTENSITY DELAY TIME RESONANCE
5 5 5
0 10 0 10 о 10
LF
FREQUENCY «св САТЕ
IN
© OUT
GATE DELAY
THRESHOLD DELAY TIME GATE TIME
5 5 5
0 10 o 10 0 10
- SO
Roland
=
89 Accessories Used in Synthesis
Fig. 4-10 (a) shows the frequency response of a typical phase
shifter, The solid line shows the response with no control
voltage input and the SHIFT FREQUENCY control set at
TkHz (note that the highest portion of the center hump
occurs at 1kHz). The dotted line shows the response when an
external control voltage of +1 volt is applied to the control
input. The entire response curve is shifted up approximately
one octave. The important point here is that the shape of the
curve changes very little, if at all, and the distance in pitch
between each of the peaks and dips remains the same. When
using the phase shift as an effect, it is not uncommon for the
center hump to sweep between 30Hz and 10kHz. Figs. 4-10
(b) and (c) show what happens to the response curve when
the RESONANCE control is used. The result is that the
humps and dips become more rounded. At maximum
resonance, the response curve almost becomes a inversion
of the normal response curve. To the ear, the use if resonance
enhances the phase shift effect.
Accessories Used in Synthesis 90
Fig. 4-10 Phase Shifter Frequency Response
(a) With Shift Frequency Control set at 1kHz.
+10
de —10
30 100 300 1k 3k 10k
FREQUENCY
®) With Resonance Control at “5” (Shift Frequency = 1kHz)
30 100 300 1k 3k 10k
FREQUENCY
(© With Resonance Control at Maximum (Shift Frequency = 1kHz)
| < AN]
DN NE \ /
æ N MA N /
= u \/ IN
Nr
91 Accessories Used in Synthesis
Fig. 4-11 shows the frequency response of a typical flanger.
Note that the dips (cancellations) in the response curve occur
at regular frequency intervals. In (a), the distance between
any two adjacent dips is approximately 50Hz. In (b), where
the delay time is different, the dips are separated by
approximately 470Hz. In (c) and (d) are shown the effects of
resonance, which enhances the effect as heard by the ear.
Fig. 4-11 Flanger (Audio Delay) Frequency Response
(@ With Delay Time = 16ms* (No Resonance)
30 100 300 1k 3k 10k
FREQUENCY
(6) With Delay Time = 2ms (No Resonance)
30 100 300 1k 3k 10k
FREQUENCY
(© With Resonance at “5” (Delay Time = 2ms)
Accessories Used in Synthesis 92
* ms = millisecond; 1 millisecond = 0.001 second
30 100 300 1k 3k 10k
FREQUENCY
(d) With Resonance at Maximum (Delay Time = 2ms)
+10
TN |
| В ва
= \ i ~
ue | | x
\ V N
—10
|
-20
30 100 300 1k 3k 10k
FREQUENCY
93 Accessories Used in Synthesis
Fig. 4-12 shows an example of the use of an audio delay unit
for the synthesis of a solo violin sound.
Fig. 4-12 SOLO VIOLIN (Audio Delay)
(@ Block Diagram
—>| VCF-1— HPF
VCO | MIX >
PL VCA ST
AA >| VCF-2 À
A
Pitch
cv _|aubio|
LFO ADSR > DELAY
Gate + Trig \ /
Accessories Used in Synthesis 94
(b Patch Diagram
system-100m 121 O
1
OUT
AUDIO
DELAY
FIXED
EXTCV EFFECT Fa i
; == A „COM @ | pT
MOD INTENSITY DELAYTIME RESONANCE
3
50
0
: LFO
«св
FREQ FRED DELAY WAVE FORM OUTPUT
RAN LEVEL
Pitch Cel SA I
«01.
E
© eo
Pitch CV | ADSR
fr : GD
MANUAL GATE ENV 1 OUT
CK DECAY SUSTAIN RELEASE
АТТА!
O TIME TIME LEVEL TIME
Gate + Trig i | - | | ;
«= |
= 5 1
G
" “
CV ul |
4 .
3 KYeD
TRIG
“a
3 OR" REO Cv IN
Сосо = Roland
IYIEM= Cm — 1 == = Neo y
— — —
O: O
С ]
|| Ш
95 Accessories Used in Synthesis
One of the factors which determines the tone color produced
by a violin is the speed and pressure of the bow as it passes
across the string. It follows, then, that the tone color will
change slightly as the bow changes direction at the top and
bottom of its stroke. One way this can be imitated is shown
in Fig. 4-13 where the output of the envelope generator
is used to alter the amount of delay produced by the audio
delay. As an experiment, try this patch with the envelope
settings shown in Fig. 4-14.
97 Accessories Used in Synthesis
Fig. 4-15 shows what might be termed a “complete” violin
patch. It uses two VCF's and an audio delay to help produce
the correct tone color. A foot pedal is used to control the
loudness of the sound. The loudness, pitch, and envelope all
affect the tone color.
Accessories Used in Synthesis 98
Fig. 4-15 Complete SOLO VIOLIN Sound
1729)
—
SYSTEM-100M
Roland
JO
OUTPUT JACK
ra»
DD.
Швы ЕМУ
с d
r0—0—0—O
b
KEYBOARD
Roland
INPUT O pu
CV o 4 )
98% GATE (@) O
O
Roland Studio System , :
SYSTEM-100M 1944)
To Keyboard
“Output”
Jack
o
v O
EN
3>
ou
> wn
Ч
о
и ©
99 Accessories Used in Synthesis
4-5
Chorus and Echo
Normally with audio delay effects, the time delay is slowly
varied over a wide range (typically from 0.5 milliseconds to 5
milliseconds) and a high amount of resonance is often used
to enhance the effect. If, however, the time delay is varied
over a small range and no resonance is used, the result is what
is known as a chorus effect. The effect causes the sound of a
solo instrument to sound like a group of the same instru-
ments. In a group of instruments such as the string section
of an orchestra, it is highly unlikely that each and every
member is playing the exact same pitch: some instruments
will be slightly flat and others slightly sharp. The overall
sound, then, extends slightly above and below some average
pitch to give the sound of the string section a very broad
feeling. A fixed delay time would merely produce a sound of
the same pitch, delayed. Constantly but slightly varying the
time delay, however, causes slight variations in the pitch of
the delayed sound. These slight variations, when mixed with
the original sound, produce the chorus effect. Most flangers
and audio delay units can generate chorus effects, but there
are also machines designed particularly for the production of
these effects.
Echo can also be used very effectively for broadening group
type sounds, most notably sustained string sounds. When
used in sustaining legato passages, an echo which follows
closely behind the original sound tends to double the
apparent number of instruments in the sound. When the
pitch changes, the fact that the original pitch remains for a
second or so in the background further increases the feeling
of great depth and broadness. The effect is lost, however, in
quickly moving melodic lines because often the pitches do
not remain long enough for the echo to “catch up” to them,
therefore, the original sounds are rarely reinforced by their
echoes. This in itself, though, can be a desireable effect. For
example, the complexity of arpeggio passages can be
increased by using echo; or, as another example, imagine a
short quick phrase played using a flute-like sound which is
followed with a long rest filled with decaying repetitions of
that phrase.
The principles of echo and audio delay are exactly the same,
but echo usually involves very long delays measured in
seconds rather than milliseconds. There are two basic
approaches to generating echo effects electronically: the tape
echo and the solid state delay line. Fig. 4-16 shows an
example of each type. Both of these machines also include
provisions for generation of chorus effects.
Accessories Used in Synthesis 100
Fig. 4-16 Chorus and Echo Machines
(a) Roland SRE-555 (Tape Echo; with Chorus)
Erase Head,
Record Head,
3-Playback Head,
S on $ Head.
Tape loop
(6) Roland SDE-3000 (Digital Delay)
101 Accessories Used in Synthesis
Fig. 4-17 shows how tape echo is generated. The delay is
produced by the time required for the recorded sound to
travel between the record head and the playback head. In
most tape echo machines the tape speed is variable, or the
position of the playback head can be changed so that the
delay time can be set as desired. Many such machines also
include more than one playback head so that many types of
echo patterns are possible. An ordinary tape recorder with
three heads can be used as an echo machine if it is connected
as shown in Fig. 4-17. This is the method by which electronic
echo was first generated in recording studios. The main
disadvantage of using an ordinary tape recorder is that the
amount of delay is usually limited by the number of playing
speeds available. Also, unless a special tape loop is made for
echo purposes, the tape will have to be rewound frequently.
Solid state echo machines work on exactly the same princi-
ples, the only difference being that the record-tape-play
portion of the chain is replaced with a solid state delay line.
The main advantage to such a system is that there are no tape
heads or tape transport to wear out.
Fig. 4-17 Block Diagram of Tape Echo
TAPE LOOP
The level of this input
Feedback loop controls the amount of
echo contained in the
The level of this input determines the intensity output sound,
of the echos, or number of repetitions. At “0”,
there will be only one echoed repetition.
MIX OUT
Fig. 4-18 shows how an echo machine can be patched
between the synthesizer and the amplifier (or mixer) input.
In experimenting with echo effects, try varying the echo
repeat rate (tape speed in tape machines) while the machine
is actually producing echoes. Varying the rate will cause the
pitch of the returning echo to vary, which may be effective
in certain cases. If it were possible to vary the echo rate
relatively slowly and smoothly up and down at a rate
comparable to vibrato, the result would be the chorus effect.
In discussing chorus and echo machines, we have crossed
the hazy border between synthesis methods and effects.
Accessories Used in Synthesis 102
Fig. 4-18 Using an Echo Machine
From
VCF
To Input
From
ADSR
1
—— MOD IN ——
Roland SDE-3000
From Output
(MIXED)
To Amplifier
/
The output at this jack
is the original sound
combined with the echo.
103 Index to Terms
Index to Terms
Additive Synthesis, 1
ADSR, 1
Amplitude modulation, 37
Analog sequencer, 77
Attack time, 3
Audio delay, echo, 99
flanger, 86
Balanced modulator, 39
Bender, 73
Bias, 49
Carrier input, vocoder, 83
Chorus, 99
Clock, sample and hold, 60
sequencer, 77
Computer controller, 78
Controller, 73
Controller, analog sequencer, 77
bender, 73
computer, 78
digital sequencer, 78
foot volume control pedal, 75
joy stick, 74
MicroComposer, 78
ribbon, 73
Control voltage, 1
Decay time, 3
Direct synthesis, 1
Echo, 99
audio delay, 99
tape, 99
Envelope, 3
Envelope generator, 1
Equalizer, graphic, 79
parametric, 81
Excitation signal, vocoder, 83
Flanger, 86
Foot volume control pedal, 75
Frequency modulation, 21
Gate delay, 71
Gate pulse, 1
Glide, 53
Graphic equalizer, 79
Growl, 29
High pass filter, 12
HPF, 12
Invert function, mixers, 53
Joy stick controller, 74
two-dimensional, 74
three-dimensional, 74
LFO, 23
Low frequency oscillator, 21
Low note priority, 6
MicroComposer, 78
Millisecond, 3
Mixers, 19,47
Modulation, 21
amplitude, 37
frequency, 21
tone color, 28
Multiplier, 37
2-quadrant, 37
4-quadrant, 39
Noise, 63
pink, 69
Noise generator, 63
Parametric equalizer, 81
Phase lock, 18, 24, 51
Phase shifter, 86
Pink noise, 69
Portamento, 63
Program input, vocoder, 83
Pulse shaper, 71
Release time, 3
Replacement input, vocoder, 83
Ribbon controller, 73
Ring modulator, 39
Sample and hold, 59, 69
keyboard, 61
Sequencer, analog, 77
digital, 78
S&H or S/H, 59
Speech input, vocoder, 83
Subtractive synthesis, 1
Sustain level, 3
Sync, VCO, 16
with pulse shaper, 72
Tremolo, 37
Trigger pulse, 6
VCA, 1
VCE: 1
VCO, 1
Vibrato, 21
Vocoder, 83
Voltage controlled amplifier, 1
Voltage controlled filter, 1
Voltage controlled oscillator, 1
Voltage registers, sequencer, 77

advertisement

Key Features

  • Modular design
  • Compact and economical
  • Subtractive and Additive synthesis
  • Voltage controlled oscillators (VCOs)
  • Voltage controlled filters (VCFs)
  • Voltage controlled amplifiers (VCAs)
  • Envelope generators
  • Mixer
  • Trigger and gate outputs

Frequently Answers and Questions

What is the difference between subtractive and additive synthesis?
Subtractive synthesis involves filtering a waveform rich in harmonics to produce sound with the desired harmonic content. Additive synthesis involves controlling and mixing sine waves to produce sound with the desired harmonic content.
What is a voltage controlled oscillator (VCO)?
A voltage controlled oscillator is an oscillator whose frequency is controlled by a voltage.
What is a voltage controlled filter (VCF)?
A voltage controlled filter is a filter whose cutoff frequency is controlled by a voltage.
What is a voltage controlled amplifier (VCA)?
A voltage controlled amplifier is an amplifier whose gain is controlled by a voltage.
What is an envelope generator?
An envelope generator is a circuit that produces a voltage that changes over time, which can be used to control the amplitude, filter cutoff frequency, or other parameters of a sound.
What is a mixer?
A mixer is a circuit that combines multiple audio signals together.

Related manuals

Download PDF

advertisement