- Entertainment & hobby
- Musical instruments
- Synthesizer
- Roland
- System 100M Synthesizer
- User manual
- 105 Pages
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.
▼
Scroll to page 2
of
105
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.