US2274370A - Method and apparatus for producing musical sounds - Google Patents

Description

Feb. 24, 1942. E, L, KEN1' 2,274,370

METHOD AND APPARATUS FOR VPRODUCING MUSICAL SOUNDS T T I i@ hlllllllllrf.

' INV ENTOR ATT'oR Y Feb. 24, 1942. 5.1.. KEN-r 2,274,370

METHOD AND APPARATUS FOR PRODUCING MUSICAL SGUNDS Filed Oct. 21, 1939 7 Sheets-Sheet. 3

Camp/ex Wal/e INV ENTOR -ATTORNEY Feb. 24, 1942. E, L, KENT 2,274,370

METHOD AND APPARATUS PQR PRoDucING MUSICAL sounns Filed oct. 21, 1939 7 sheets-sheet 4 INVENTOR ATTORNEY Feb. 24, 1942. E. L. KENT METHOD AND APPARATUS` FOR PRODUCING MUSICAL SOUNDS 7 Sheets-Sheet 5 Filed Oct. 2l, 1939 www:

msm .wma ww.

dii?? wmw QS N..

- ATTQRN EY INVENTOR v /L/e/f BY Z Z Z f Nl bmw. KQ\ Wh.

Febfzl, V1942. E, L, KENT 2,274,370

METHOD AND APPARATUS FOR PRODUCING MUSICAL SOUNDS .41 L Vw E l E. L. KEN T Feb. 24, v1942.

METHOD AND APPARATUS FOR\ PRODUCING MUSICAL SOUNDS Filed Oct. 21, 1939 'T Sheets-Sheet 7 INVENT'OR an@ ,Ye/7% ATTORN Y.

Plantea Feb. 24, 1942 METHOD AND' APrnATUs Fon rnonucnm MUSICAL soUNDs Earle L. Kent, Chicago, lll.

ApplicationOctober 2l, 1939, Serial No. 300,683

(Cl. Sdi-1.28)

' 19 Claims.

My invention relates to a method and apparatus for producing musical sounds, and more particularly to a new electronic method and al)-l paratus whereby organ-like musical instruments may be achieved.

Heretofore lin the electronic musical instruk`ments oi' the prior art, it has been necessary-to produce tones by means oi' individual generatorsin order to obtain the desired timbre at each diilerent pi h. Every musical note has two distinctive qua ties, namely, its timbre, depending upon the instrument which produced it, and its pitch. Any means for producing a harmonic vibration of a given pitch possessing the characteristic wave shape of a violin, will produce a violin tone. It is possible to produce any deaired wave form by synthesisfbut it is necessary to have a plurality of synthesizing means to obtain the desired pitch.

Electrical engineers have for some time been cognizant of .means o! producing multiples or submultiples of a given wave by means of the relaxation oscillator, rectification and the like. I have found that I may change the frequency o! a given wave, simple or complex, as for examkple a wave of one `liitch having a predetermined desired timbre to a new frequency, and retain the same wave form as the original wave.

-Itwillbeapparenttothoseskilledinthe art, for example, ii diapason tones are desired I may,

according to my invention, provide a single diapason generator and furnish tones at-any audi,- blepitch or any combination of pitches. The

musician may sound diapason C," CL D andv the like, independently or together in any combination, 'from the same diapason generator.

Similarly, I may by my method and apparatus, provide an instrument having any desired number yof generators for generating musical timbres of known instruments or of new qualities, and convert them into any desired pitch or combination o! notes having a plurality ot desired pitches.

'I'he advantages of such an instrument will be apparent. The complexity of an instrument made according to my invention is reduced, and its nexibility .tremendously increased. For example, i! it were desired to produce'an organ having 1000 pipes (individual tone producers), to geierateaodierenttonequalitiesforszpedone moet of my invention au provide a.'

-novel methodotproducingmusicalemds.

Another object of my invention is to provide a novel apparatus for producing musical sounds of any desired pitch and timbre.

Another object oi my -invention is to provide. 5 a novel electronic musical instrument.

Another object of my invention is to provide a novel electronic musical instrument in which standard and easily obtainable radio parts are employed. f 4

Another object of my invention is to provide any electronic musical instrument in which the circuits are not critical and their adjustments vcan be made by any person having knowledge of radio repair, without special `apparatus or skill. Another object of myv invention is to provide an organ-like electronic musical instrument in which standard stop controls may be used.

Another object of my invention is to provide an organ-like electronic musical instrument in which the harmonic content may be varied or synthesized to suit the taste of an individual and in which a large number of diilerent stops may be easily obtained. Another object of myl invention is to provide a musical instrument in which -a tone of any given timbre and pitch may be changed to another tone of the same timbre and of a diii'erent pitch.

Fig. l is aschematic Another object of my invention-is to provide an electronic musical instrument in which every tone produced has true harmonics and not tempered harmonics. j

Another object lof my invention is to provide an electronicmusicalinstrument'in which every tone of the`scale may have exactly the same quality or harmonic content, or may be made to vary somewhat from octave to octave in order to obtain the same quality variations as are present in other musical instruments.

Another object oi' my invention is toprovide a `musical'instrument in which the entire instrument may be given a vibrato or combined vibrato and tremolo `with a Asingle vibrating means.

Anotherr obiect ci my invention is to provide a novel musical instrument in which the vibrato may be applied to individual manuals vor to the pedals Other objects oi'my invention will appear from the following description.

In the accompanying drawings which form part of the instant specincation and which are to be read in conjunction therewith, and in which like reference numerals are used to indicate like parts in the various views:

viewed' one form of fn'e- Y quency changer which enables me to apply a complex wave at a particular fundamental frequency and sine waves of other frequencies to obtain waves of dierent frequencies whose wave shapes conform with the shape of the original applied complex wave.

Fig. 2 is an end view of the cathode ray tube shown in Fig. 1.

Fig. 3 is an oscillogram taken with no complex wave applied to the cathode ray tube of Fig. l, representing a curve of voltage against time.

Fig. 4 is a diagrammatic view showing a complex wave'plotted against time with relative positions of the probe.

Fig. 5 is a schematic view of a frequency changer circuit capable of producing the results produced by the circuit of the apparatus shown in Fig. 1.

Fig. 6 is an. oscillogram similar to Fig. 3, showing unidirectional pulses produced by the circuit of Fig. 5 with no complex wave applied.

Fig. 7 is a schematic view showing a stop control for producing predetermined timbres.

Fig. 8 is a curve showing voltages applied to the grid of the frequency changing tube, and the resultant curve of plate current.

Fig. 9 is a schematic view showing the vibrato and tremolo producing means.

Fig. 10 is a schematic View of a portion of an organ showing the relative location and connection of parts.

Fig. 1l is a schematic view showing the method of keying pulses employing a single frequency changer tube with a plurality of pulse generators.

Fig. 12 is a fragmentary detail view showing one method of keying.

Fig. 13 is a sectional view taken on the line I3-I3, Fig. 12. l

Fig. 14 is a sectional view taken on the line I4I4, Fig. 13. I

Fig. 15 is a schematic view showing an instrument in which the human voice may be employed to initiate the complex wave.

In general, my invention contemplates the provision of a musical instrument in which a complex wave of any desired wave form is continuously generated and this complex wave changed in its frequency to obtain the desired pitch.

More particularly, referring to Fig. l, a cathode ray tube I is provided with the usual cathode 2, a grid 3, a first anode 4 and an accelerating anode 5. There are two sets of deflecting plates i and 1 disposed at right angles to each other. 'I'he end of the tube is provided with a. collectingring 8 and a probe 9, the end of the tube I0 being prol vided with the usual fluorescent screen. cathode is heated by battery II.

Negative potential from battery I2 is impressed by conductor I3 upon the grid 3. The anode 4 is connected by conductor Il to the positive'terminal of battery I5. Battery I6 is connected in series with battery I5, and its positive terminal is connected by conductor Il to the accelerating anode 5. A vacuum tube oscillator IB, of any suitable type, is adapted to generate a high frequency potential in the form of a sine wave. The output of the vacuum tube oscillator Il is impressed across deecting plate 6 through an in- The ductance I8, and across deflecting plate l through a capacity 20. The inductance and capacity are such that the voltage impresed on defiecting plate 6 is 90 out of phase with the voltage impressed across defiecting plate 1. Considering now the action of the cathode ray tube, let us assume that the complex wave generator 2I is generating q a sine wave. Then the voltage applied to the grid e. may be represented by the expression eg--A sin 21u-ft.

Disregarding the D. C. bias applied to the grid for the sake of simplicity, the intensity of the electron beamib may be' represented If the cathode ray beam is focused on the end of the tube by the usual procedure, and the beam moved by the application of voltage :I:

l 1:8 cos 2rf1t l applied to the vertical plates, and by the voltage y 1V=B sin Zrfit the beam, the generated voltage between` the" secondary emitter and the collector may be expressed by the expression where k is a constant, ib is the intensity of the electron beam, and the expression Fmt) is a function of fit.

With the probe 9 in the position shown in Fig. 2, ep will have a value only when y equals zero, and :r is positive. So, F(fif) might be defined as cos 21m where 11:0, l, 2, and also n=f1t then,

ep=kit cos 2nr=A sin 21rft cos 112x.

From the above equation we obtain aseries of discrete values every time /it=n=0, 1, 2, 3, and of course the usual voltage would be the average. of these discrete values. It would be equally correct on the other hand to define FUit) so that since the last term is always zero when the equation is evaluated. Since n=f1t we may write Y l l e=A[sin 2gb cos 21a-cos 211i' sin 21m] f1 f I and by -trigonometric identity,

It will be apparent from the foregoing equation that the voltage generator will be a series of discrete values whose average can be seen to be a 'sine Wave whose frequency is equal to f-fi,

and Whose peak value will be proportional to thepeak value of the voltage applied to the grid, and independent of the peak value of the voltage applied to the deflecting plates. It must be remembered that the voltage is unidirectional and can have no reversal of' direction as the above equation indicates, but if' the D. C. component had been carried along through this derivation the physical significance of the equation would be at once apparent.

If no signal was applied to the modulating lgrid 3, the electron beam would be deflected in a circle. Every time the beam of electrons hit the secondary emitter S, it would cause the probe to emit electrons which are collected by the collecting ring Il. The potential difference between the probe and the collecting ring may be ampliiied by the vacuum tube 22, the output of which tube is loaded so that any rapidclianges in potential of the plate 23 are practically prohibited `by the condenser 2l. High frequency signals are blocked by the choke coil from going to the grid 2i of the vacuum tube 21. It will be seen that with no signal applied to the modulating grid 3, that the pulses of the voltage between the probe and the collector will be of uniform height proportional to the beam current at the time it strikes the probe.

With the amplifying arrangement shown, the occurrence of the pulses in rapid succession and at the same magnitude will result with no signal output across points 29 and 29 of the amplifier.

If we 'apply' a simple alternating potential to the grid eg=A" sin 21rft, as pointed out above. the

intensity of the electron beam ib may be expressed by ib=A sin Zrf't-i-K Let us consider that the frequency of the signal applied to grid 3 (,f) is equal to the frequency of the signal applied to the defiecting plates 6 and 1 (fr). in a circle `on the end of the cathode ray tube I,

the circle of light would be a stationary pattern of sinusoidal distribution such as shown in Fig. 2,I

in which the portion would be of maximum brightness, the portion 3| of half brightness, and the portion 32 of minimum brightness.

The equation of the circle on the end of the4 tube 9 is as follows 1206,11) :B cos- Zrfit-HB sin 2rfit ever values of t will make iB sin 21rfit=0. Furthermore iB sin 21rj1t will equal zero if fit=n where n is any positive integer.

From this relation we can determine the value In such case when the beam sweeps made one revolution the variation in voltage of.

the pulses will have gone through one cycle.

' To be more specific take as an example a case where f'=101,000 cycles per second f1=100,000 cycles per second where the beam would make one vrevolution on the end of the tube I in .00001' sec. but ib goes through a cycle of values in M6100@ sec. and therefore the pattern will advance .0l rev. in the clockwise direction every time the beam goes around. Since the beam goes around 100,000 times in one sec., the pattern will make 100,000 .01=1000 revolutions per second.

With these values for f and l1 we will get 100,000 vpulses at the probe in onesecond but they will not all be of the same magnitude because ib is not always at the same value when it hits the probe. We can see in fact that since the pattern is moving with respect to the probe as it is. that the height of the pulses will go through sinusoidal variations 1000 times per sec.

While condenser 24 offers a low impedance to a pulse frequency of 100,000 cycles per second, it

will be a high impedance to the variation in average pulse' height which occurs at the rate of 1000 cycles per second, so that the 1000 cycle per second variations will appear at the grid 26 of the tube 21 and hence across the output\28 and 29.

When the diierence-between f. and fr is small compared to f1 the equation o=velocity of the beam variations V=velocity of the beam :wave length ,y f=frequency at output 29 and 29-4 vThat the above equation does not completely interpret ourlphysical conditions can be seen if of me intensity of the beam at the time it mts the probe, thus where K is the beam intensity due to the battery l2 and is the steady' component above and below which the alternating component operates.

If in=K every time the cathode ray hits the maximum value before -it reaches the top of the circle (Fig. 2) after one revolution of the bea'm and as this process continues .the sinusoidal pattern of light appears to move in the clockwise direction. n,

When there is relative motion betwe'nwthe probe and the pattern the value of in will not always'be the same but after the pattern has we make,

' f'=200,000-C. P. S. ,f1=100,000 C. P. S.

in which case the i would undergo exactly two V sinusoidal variations while the beam made one revolution on the end of the tube and therefore would give a standing pattern as before except that now our circle of light would have two maximum points and two minimum points. There wquld be no signal 'at output 28 and 29 since there is no relative motion between the pattern and the probe. Yet the above equation indicates f=100,000 C. P. S. So we might express the equation as follows, f=f'-f1m where m=integer representing the number of waves in the rotating pattern, then f='200,000100,000 2=0.

Suppose now that welapply to grid), Fig. 1,

which will modulate the intensity of the electron beam amil' sin 21rf't+A2' sin ff't and for example let us assign the following values: o

f'=100,000 C. P. S. fi=100,000 C. P. S.

which will give us a stationary pattern on the end of the tube and the pattern will consist of a fundamental wave and a second-harmonic. Since the pattern does not move with respect t9 the probe there is no signal at output 28 and 29.

Then consider,

f'=101,000 C. P. S. 1==100,000 C. P. S.

which will cause the fundamental pattern to rotate as pointed out above at 1000 revolutions per second but the second harmonic will pass the probe twice as often or 2000 R. P. S. The rotation of the pattern will cause variationsA in pulse heights that will-give a signal at output 23 and 23 made up of a fundamental whose frequency is 1000 C. P. S. and whose amplitude is proportional to A1 plus a second harmonic whose -frequency is 2000 C. P. S. and whose amplitude is proportional to A2', In other words, the signal at output 28 and 29 is e=A1 sin 21rft+Az sin 21rft It will be understood, of course, that the circle of light does not exist as such, but is rather the result of the persistence of vision and the persistence of the screen itself. The electric circuit, however, makes the instantaneous values appear as average values across the output 28 and 29, and therefore evaluates them much the same as our eye does the moving spot of light caused by the uorescence of the screen under the impact of the cathode ray. l

It will be understood that my frequency changing principle makes use of an electrical effect analogous to the Doppler effect in sound. The relative motion of the probe to the revolving electron beam modifies the frequency of voltage variations in the electron beam, and may be expressed as follows:

in which v=t`ne velocity of the electric wave V=the relative velocity of the probe A=the wave length of the wave, and

f=the frequency of the wave experienced by the probe It will be observed that in my method for producing an electrical effect analogous to the Doppler effect I make use of a cathode ray tube, but my method is not to be confused with systems using cathode ray tubes for the generation of signals in which the output is dependent upon the.

shape or material of the object being scanned by the electron beam for the wave shape and frequency.

=A, sin 1),@ A magg-1) It will be clear from the foregoing equation that we will obtain a generated voltage whose fundamental has a peak value dependent only upon the peak value of the fundamental applied to the grid, and further, a second harmonic whose peak depends upon the peak of the second harmonic applied to the grid. It will be readily understood by those skilled in the art that a complex wave with any number of harmonics can be treated in the same manner as the simpler waves. Thus far we have considered only cases in which sine waves were applied to the deflecting plates 8 and l. 1f we are to produce a musical instrument, we must be able to produce more than one note at one time from the same tube.

I can apply a multiplicity of sine waves to the deflecting plates and obtain a corresponding number of complex waves. The physical concept of the mechanism of this operation is much more difficult to visualize than the simple case, since in this instance the electron beam no longer travels in a true circle at the end of the tube.

The fundamental basis of my invention is not limited to any specific piece of apparatus. It will bel observed that the cathode ray tube is used in the embodiment shown in Fig. l to generate a plurality of pulses at predetermined frequency, the pulses being sharp, that is, the width of each pulse is narrow as compared to the distance between pulses. When the height of the pulses is the same, no output will be obtained from the amplifier. Upon this pulsing current I impose a modulating current having any desired wave form. The result of the superimposition of the modulating current will be a series of pulses whose envelope'` will describe a wave whose frequency is a function of the difference between the frequency of the modulating wave and the frequency of the pulses. The form of the envelope, furthermore, will be the form of the modulating wave.

Any means, therefore, of producing a plurality of pulses of uniform height and at a given frequency may be used as the basis upon which to impose the modulating current to obtain a current having a form identical to the modulating current form, but at a new frequency depending upon the frequencies of the pulses and the modulating current. The result lwill be the equivaient of that obtained with the cathode ray tube arrangement just described.

Referring now to Fig. 4, the curve 33 represents the voltage plotted against time t. If the wave 33 is probed or scanned at a frequency which is the same as the frequency of the wave 33, the probe, which is represented by the reference numeral 3l, will remain stationary at full line position, and will experience no voltage variations. If, however, the probe is moved in the a: direction, within a *time t-i-At, to a position shown by the dotted line 35 in Fig. 4, and within a time t+2At to a position shown by the dotted line 36 in Fig. 4, and so on at the same rate, then the probe will experience voltage variation that is the same as the wave shape of the voltage distribution but at a different frequency than if it had-no relative motion.v It will be clear that the effect is equivalent to the Doppler effect in sound, which may be observed by the sudden change in pitch of the whistle or bell of a passing train. The change in pitch is due to the relative motion of the sound source with respect to the ear.

Another embodiment of my invention is shown in Fig'. 5, in which I employ a feed-back oscillator and a mixer tube.

Referring now to Fig. 5, the tube 31 is a double triode acting as a feed-back oscillator to create sharp positive pulses at the grid 33 of tube 33. The tube 33 is a pentagrid mixer amplifier tube.

The grid 33 is normally biased negatively so that the tube 33 is at cut-olf and no plate current flows in it. However, the positive pulse applied tothe 'grid 33 causes the plate current to flow in tube 33 in sharp pulses. e wave whose frequency is to be changed is pplied to the grid 43 of the tube 33. Since all pulses applied to the grid 33 are of equal magnitude the plate current pulses will also be of equal magnitude if the voltage on grid 40 is not varied. Then, too, if the complex wave applied to the grid43 is of the same frequency as the pulses applied to the grid.

pulses obtained from the cathode ray tube in the' arrangement in Fig. 1.

The sampling or scanning action is shown graphically in Fig. 8. When the tube 31 has a B voltage of, for example, 100 volts, applied to the plate 4I, a large plate current will ilowu since the grid 42 is connected to a positive point 43 on the B battery 44. The flow of current from plate 4i to cathode 45, to the negative terminal 45 of the battery 44 is through a resistance 41. The voltage drop across the resistance 41 biases the tube 31 and causes a decrease in the plate current flow. The drop in the plate current occurs rapidly, causing a sharp rise in voltage across the plate resistor43. This positive pulse is impressed upon'the grid 43, where it undergoes a 180 degree phase change with amplification and is sent as a negative pulse to the grid applied across points 53` and 54 is double the frequency at which the circuit would normally operate, then every other pulse will reach the grid 42 at a time when it is notI near enough to cut- 5 oi! value to 'f tripf the circuit, but the other pulses would be effective.

Generally feed-back oscillators of theA type shown are not sufficiently stable to operate dependably at a given frequency unless they are held ,at that frequency by a stable signal whose frequency is the multiple of the frequency kdesired from'the oscillator. An oscillator of this type, however, is desirable in my invention because of its simplicity and because it gives the l5 properly shaped pulses employed in my frequency changer and frequency changing method.

Referring again to Fig. 5, tube 5,5 is apentode used as an electron-coupled oscillator with high frequency stability: It will be understood, of

course, that an oscillator of any design could 'be used, provided it was sufliciently stable at the desired frequency. Furthermore, if desired, a crystal controlled oscillator could be employed sincev the oscillator operates at. radio frequency.

This is a point of great advantage over the elecmultiple thereof tronic instruments of the prior art whose oscillators operate at audio frequency. 'The oscillator may operate at any desired frequency,- say. forexample, 100,000 cycles per second, or some In this case the twin triode tube 31 acting' as a feed-back oscillator would be tuned to oscillate at slightly less than 100,000

cycles per second. The output of the'rxtube 55 is impressed across points 53 and 54 by means of transformer 55. The impression' o f the output of,the oscillator 55 across points 53 and 54 will lock the tube 31 to the frequency of oscillation of the tube .55, since the signal of the tube 55 will be impressed across points 53 and 54.

40 'I'he arrangement is such, therefore, that the output of tube 31 will be narrow pulses at th discharge through condenser 53 since the time 42, where it drives the grid to cut-off" and plate current ceases to ilow from plate 4I to cathode cycles per second. This output will be applied, 45. The voltage across resistor 41 will quickly as pointed ont above, to the grid 33 of tube 33. 45 -A complex wave whose frequency is to be frequency of the control tube 55, that is, 100,000

constant of resistor 41 Aand the condenser 53 is very small. 0

When the voltage acrossresistor 41 falls farenough to permit plate current to ow again lin the plate. circuit of plate 4I, the cycle is repeated. The values of the,resistors and the condensers are such that very narrow positive pulses are created across the resistance 4l, which pulses are impressed/upon the grid '33 of tube 33 through conductor 5|. The resistor 52 has considerable iniiuence upon the frequency at which the circuit will oscillate, and this will'be more fully dis'- cussed hereinafter. The other resistors and conkdensers have some effect on frequency, but they should be chosenso' as io'give'the sharpest pulse near the desired frequency. 'Ihe resistor l2 can then be relied upon to adjust the frequeny'over relatively-wide range. with a` minimum eect on the width of the pulse. The exactfrequency 'at which the circuit oscillates is controlled by the introduction of pulses at point 53, as will'be pointed out hereinafter.

s0f hereinafter.

changed is applied to 'the grid 43 at point 51, the other side of the ,output beinggrounded. 'I'he complex wave may be generated in any suitable manner, one of which will be pointed out Let us assume that the complex wave has a frequency of 101,046,502 cycles per second, and this is applied to the grid 4I of the tube 33- through point 51. 'nibe 33 acting as a frequency 55,-changer will create in its plate circuit 100,000.

pulses per second-and these pulses will vary periodically in magnitude 1046.502 cycles .per second with an envelope who wave form is identical with the complex wave applied to the i0 grid 43. Awave whose frequencyris 10462502 `If the resistance 52 is set so that the circuit will oscillate slightly slow'- A er than is desired, then upon the introductionof positive pulses, of exactly the correct frequency upon the grid 43 across P0ints'53 and 14, these pulses will be amplified lby the left hand section of tube 31 and arrive at,the right hand'section ofr tube 31 to force it to cut-off if it is near the4 cut-oi! value. If the frequency of the pulses cycles'per second is Ca in the musical scale.

The tube 33 will function if it `has the proper "bias The bias ofthis controned yin the diagrammaticv view of Fig-.5 by the key 53. Key 05 '53 is normally against the stop 53 and makes 'contact with 50 if the cam 5I is in the position shown. When the key 53 isin the up position against Stop 55, th grid l0 0f the tube 39 is biased to "cut-Off by the battery 52. It ls biased .T0 to cut-off regardless of the position of the cam 5I. When the Vkey is pressed down toward stop 53 it vmaires contact with contact points 54 and 55 simultaneously if the cam 35 is in the position shown.` If the switch 51 is in contact with`con- 5 tact'point 33, there is enough positive potential from battery 69 to bring the grid Il of tube 39 to a low enough negative potential for the tube to be properly biased when the pulses occur. When the key 58 is released and comes into contact with contact point i0, the voltage across condenser 1I is lowered to bias the tube again to cut-off. The condenser 1I is in this case charged through resistor 1| and the time constant of resistor 1I and condenser 10 is such that the condenser charges quickly but not abruptly. It is fast enough for a standard release of tone, but not fast enough for a click or thump effect. If the cam 6| is turned so that the key 58 cannot contact contact point 60 when it is released then the condenser 10 must take its charging current through the resistor 1I and resistor 12 in series, thus increasing the time constant. The resistor 12 is large enough to cause the tone to decayl suiliciently slowly to make string tones sound like plucked strings, chime tones to sound like chimes, and bell tones to sound likebells.

Cam 6l is operated by a pedal and the shaft 13 upon which it is mounted may run the full length of the key manual. Cam 65 is mounted on shaft 14 and it may be moved to three positions. In one position contact points 64 and 65 are normally in contact so that when the key 58 is depressed it makes contact with both these contact points as a unit, and the resistor 15 aly Ilows charging current to build up a less negative voltage across the lcondenser 10 quickly but not fast enough for key thumps. The second positionv to which cam 66 may be moved separates contact points 65 and 6I so that they will close in sequence when the key 58 is depressed, giving second touch which will be explained more fully hereinafter. The third position to which cam 55 may be nioved is to depress contact point 65 downwardly so that the key 58 will contact only contact point 54 when pressed. In the third case the charging current for condenser 10 must required .to push the key down; The switch I1 may be arranged at the end of the shaft 1l in such a position that the switch I1 will contact contact point 11 when the cam is in the first position, and contact contact point Il when the cam is in the third position making it possible to change the attack time while keeping the operating bias the same in either position.

The operation of the arrangement can be readily understood by reference to Fig. 8, in which ip represents the plate current output of the frequency changer tube and 'e represents the grid voltage impressed upon the control grids of this tube. 'I'he curve 1l is a conventionalized characteristic curve of the tube. 'I'he curve 1I represents the pulses generated by the pulse generatcr at a predetermined frequency. The curve I0 represents the complex wave whose frequency is to be changed. It will be noted that the positive pulses alone operate below the point of cutoi. It will be likewise noted that the complex wave is below the point of cut-ofi'. The pulse wave scans or samples the complex wave at a predetermined frequency, and due to the diiference of the frequency between the complex wave and the scanning wave, the output of the irequency changer will, by an eii'ect analogous w the Doppler eiect described above, produce a series of pulses 8| which will vary riodlcally in magnitude at a frequency depe ding upon the relative motion of the scanning or pulse wave with respect to the complex wave, and whose envelope l2 will comprise a wave form exactly like the complex wave Il.

The timber of a musical` note, that is, its wave shape, is governed by its harmonic content, and

flow through resistances 16 and 15 in series.

with a small number of parts and contacts.

When cam 66 is in the second position, that is, with contact points 64 and 65 separated but in such a position that contact will be made with contact points -64 and 55 in succession,rif the key 58 is pressed only part of the way down, it

makes contact with contact point N and brings the bias of tube I! to an operating region at a relatively slow rate of attack. It the key, however, is pressed further, so as to make contact with both contact points N and i5, the bias is made less negative quickly and the resulting tone will be louder because of this change in the operating pointof the tube 39. The cam 66 can be so shaped and located that when it pushes contact point 65 into the second position, it may still rest against the bottom of contact point El, serving as an index to the musician because extra pressure will be required to make the further contact with contact point 65. In this manner the musician may play lightly and put ernphasis on some notes by more pressure, making them sound louder and brighten Quickly pushing the key down as far as it will go, gives a faster attack time as well as making the note soun'd louder than those not pressed quite all of the way. When the cam 65 is in the rst or third positions, it has no effect upon the pressure the relative amount of the various harmonics present., In order to obtain a complex wave equivalent to a musical instrument or a plurality of musical instruments, I provide the arrangement shown in Fig. 7.

A radio oscillator of any suitable type capable of generating waves at radio frequency and with a high harmonic content, is shown diagrammatically by the reference numeral 83. The oscillator 83 can be a feed-back oscillator, such as described above, to generate the pulses. The voltage pulses which comprise the output of 'the complex wave generator 83 are applied at terminals 84 and 85 by conductors I6 and I1. Between terminals and l5 are a series of tuned parallel circuits 88, 89, 0U, Il, l2, 93, 94, and 95. If, for example, the complex wave generator were tuned to oscillate at 101,046.502 cycles per second, the parallel tuned circuit Il would be tuned to this frequency, namely, 101,046.502 cycles per second. Circuit Il would be tuned to the second harmonic or 202,093.004 cycles per second. Circuit would be tuned to the third harmonic, or 303,139.506 cycles per second. Similarly, parallel tuned circuit ll would be tuned to the 'fourth harmonic; circuit 92 to the fifth harmonic; circuit 93 to the sixth harmonic; circuit v94 to the seventh harmonic; and circuit 95 to the eighth harmonic, and so on for every harmonic it is desired to control. The parallel tuned,r circuit I8 is short-circuited by a resistor 86. Parallel tuned circuit l is short-circuited by a resistor 91. Similarly, resistors SI, I9, III. lill, |02 and |03 short-circuit tuned circuits II, 9|, 92, 9J, 94, and 95,' respectively.I

It is understood, of course. that a tuned circuit oilers a high impedance to waves of a frequency to which the circuit is tuned, while permitting the passage of waves of other frequen- I ing on the respective resistances.

cies. With tuned circuit 09. across terminals I4 and 05 all of the fundamental will appear at point |01. If, however, tuned circuit Il is shortcircuited by the resistor 96, then none of the fundamental will appear across terminals |01 and |00. i

The panel |09 represents the front of the iny strument board of an instrument built in accordance withy my invention. A plurality of control shafts H0, HI, H2, IIJ, III, III and Ill',

are provided. Each of the shafts represents/a stop, preferably arrangedin accord with the standards of the American Guild of Organists. f

Each stop produces a complex wave of the desired timbre to represent theinstrument. Tak' shunted except for the portion embraced be-- tween points ||9 and |20.' That portion of resistance 99 embraced between points |2| and |22 represents the fourth harmonic component of the instrument represented by stop H3. None of the fifth harmonic-,is present in the complex wave of stop H3. That portion of the resistance |0| embraced between points |23 and |24 represents thesixth harmonic component present in the stop |13. 'None of the seventh harmonic is present in the complex wave of stop H3, while that portion of the resistance |03 embraced between points |25 and |26'represents the component of the eighth harmonic present in the stop H3.

It will be seen that the arrangement is such that none of the resistance. is normally shunted with the stop controls pulled out. The pushing in of any stop produces a complex wave having a desired wave form. Its harmonic content can' Stop |||y may produce the characteristic wave form of a diapason. Stop ||2 may produce a complex wave which has the wave form of a violin' tone. Any desired complex wave may be synthesized by the proper adjustment of the harmonic content.

If it is desired to synthesize tones the stop |0 is pulled out. This shunts the resistances to amounts depending `upon contact armsy |21, |29, |29, |30, |9I, |92, |32, |94. These contact arms may be contact arms of potentiometersoperat- In this manner harmonic values may be added or subtracted timbre.

Itis understood, of course, that any suitable place by the operation of the shaft through the stop IIB. y l

Itwill be seen from the arrangement in Fig. 7

to synthesize a complex wave having any desired A' that 'we may obtain a complex wave of desired wave form provided the complex wave generator is one which will produce a wave rich in harimonic content. Inasmuch asy the complex wave generator 03 is a feed-back oscillator, such as shownin Fig. 5, the pulses produced will be of the shape shown in Fig. 6. The harmonic analysisof a wave of the shape shown in Fig. 6 show that it may be expressed as follows:

, This expression shows that when the width of the pulse is small compared to the length o f the cycle the wave contains essentially an equal amount of harmonics. As an example if 2a=.0lf the harmonics in the wave would-compare as follows: 'a

s Fundamental =l.0 2nd harmonic=l.0

It is possible, therefore, to produce any desired wave format terminals |01, |08 by having av tuned circuit and a voltage'divider for every harmonic that it is desired to control. Of course if it is not desired to have absolute control over each harmonic contained in the wave, a less complicated iilter may produce a variety of tones that may be desired. However, since only a few of such filters is needed for the entire instrument it seems desirable to be able to synthesize any desired tone. v y

It is desirable too, toy arrange the controls of the instrument in accord with the standards of the American Guild of Organists since so many of the musicians who would be apt to play this instrument have spent manywhours practicing with organs that meet these'standards. s

Referring again to Fig. 5. it was vassumed that pulses of a frequencyof 100,000 .cycles per second were applied to the gridr 30 and the complex wave applied to the grid 40 was at a frequency of 101,046.502 cycles per second, thereby producing in the output circuit a wave whose frequency was 1046.502, which is Cs'in the musical scale. Another feed-back oscillator such as shown by tube -31 in Fig- 5,v and associated frequency changer produces the note C4 in the musical scale. Its feed-back oscillator may operate at 50,000 cycles per second and is synchronized or locked at this frequency by the `100,000 cycles per second pulses across the terminal |39 and ground of tube '31. 1

In an instrument accordingv to my invention there would be another fed-back oscillator lgenerating the pulse wave like that generated by the n complex wave generator 03, but` at half its f-requency. For example, if the complex wave gen-4 across point |31 and ground of Fig. 7. This secondcomplex wave generator feeds a complex wave to the grid of another frequency changer tube producing Ca. This process is repeated for Cz, C1, and Co. It will be seen by Virtue Of this locking feature of the feed-back oscillators, all of the Cs are exact octaves of each other. In this manner a complete instrument could be built up using 12 stabilized oscillators such as shown in Fig. 5, each controlling its series of feed-back oscillators. There are 6 feed-back oscillators for producing complex waves such as indicated diagrammatically by in Fig. 7. These 6 feed-back oscillators would all be staoilized by the ilrst oscillator.

Another way of building an instrument embodying my invention is shown diagrammatically in Fig. 10, in which a plurality of stabilized impulse generators are shown, one for each note of the scale. Three octaves are shown by way of example, namely, the zero, first and second octaves.

is of such value as to shunt frequencies above the audio range but permit audio frequencies of the octave produced by frequency changer |39 to be unaffected.

In an instrument having six octaves there will be six primary windings such as |69, |64 and |65. In each case the primary winding will be provided with a suitable by-pass condenser to shunt 'unwanted frequencies of the pulses and its harwhich actuates loud speaker |69. If desired, the

Each octave employs a single frequency changer tube. The zero octave employs a frequency changer tube |39. The rst octave employs a frequency changer tube |39. The second octave lemploys a frequency changer tube |40. A single stabilized complex wave generator |4| is employed. Each of the impulse generators is electrically connected to a key, a few of which are shown by way of example in Fig. 10.

Key Key Key Key Key Key Key Key Key Key Key |42- controls the impulse generator Bz. |43 controls theimpuise generator A#z. |44 controls the impulse generator A2. |45 controls the impulse generator G#z. |46 controls the impulse generator Gz. |41 controls theimpulse generator Fitz. controls the impulse generator F2. |49 controls the impulse generator En. |50 controls the impulse generator Dah. |5| controls the impulse generator Dz. |52 controls the impulse generator C#z.

Key |53 controls the impulse generator Cz.

The complex wave generator is provided with the stop control shown in Fig. '1, and produces a complex wave at 4a'predeterminedradio -frequency as, for example, 50,000 cycles per second. The complex Wave of this frequency is impressed at all times upon the grid |54 of the tube |40, the grid |55 of the tube |39 and the grid |56 of the tube |39.

Since no scanning or pulse waves have been impressed upon the frequency changer tubes and the frequency changer tubes are all biased to operate below cut-off, no note will sound as long as none of the impulse generators may impress the output upon the grid |51 of tube |40, the grid |50 of tube |39, or grid |59 of tube |30.

When, however, a key is pressed the result of this action is such as -to permit the complex wave to be scanned by the wave of the predetermined radio frequency, differing from the complex wave frequency by a number of cycles equal to the frequency of the pitch of the desired note. If two keys are pressed, the effect of course is that of two notes sounded in unison. Similarly, any number of notes can be sounded by the actuation of the corresponding number of keys.

All of the plate outputs of the frequency changer tubes are in parallel and feed into a transformer |60. The condenser |6| is of such value as to shunt frequencies above the audio range but permit the audio frequencies of the octave produced by frequency changer tube |40 to be unaffected. Similarly, thecondenser |62 higher three octaves may be fitted with a transformer similar to that shown for the lower three octaves and fltted with an individual amplifier and volume control and loud speaker. Then, too, each octave may have its ownl individual loud speaker especially resonant to audio tones of that octave, so that truer sound effects may be achieved. The wealth and flexibility of an instrument according to my invention becomes apparent.

In Fig. 10 I have shown diagrammatically at |69 a vibrato. 'I'his is any means for producing a voltage at a low frequency which is applied in varying amounts to the grids |51, |50, and |59 of the frequency changer tubes |40, |39 and |00 through variable resistances |10, |1| and |12, which may be operated by a common pedal, if desired;

Referring now to Fig. 9, I have shown an arrangement employing a gaseous discharge tube to produce a vibrato. The amplitude of the vibrato may be controlled by the variable resistance arm |10, or by the variable resistor |14 of Fig. 9'. The switch |10 cuts the vibrato on and off. 'Ihe frequency of the vibrato is controlled by the resistor |16.

' quency'of the oscillator tube |11 to shift periodically slightly from its normal frequency. The value of resistance |10 controls the frequency of the change. The arrangement will be readily understood by those skilled in the art by reference to Fig. 9. It will be understood that any suitable vibrating means, vibrating'at a low frequency maybe employed to introduce changes of voltage upon the screen grid |19 of the oscillator tube |11 to produce a vibrato effect.

If the tube |11 represents the complex wave generator 03 of Fig. '1, it will produce a vibrato effect to all of the tones produced by the instrur ment.

Tuned circuits 09, 09, 90. 9| and the like, are sharply resonant and the changing frequency therefore will also bring about a change in amplitude of the tones and their harmonics, giving a combination vibrato and tremolo effect. Furthermore, by controlling the shapes of the resonant curves of the tuned circuits 00, 99, and the like, it is possible that the tremolo be greater on high tones than on low ones, and greater on the high harmonics of the tones than son the fundamentals.

If it is desired to give each of the twelve intervals of the scale a\`slightly different vibrato than others, the tube |11\may represent the a multi-manual instrument.

pulse generator, in which case obviously a difierential vibrato may be achieved.

It 'will be observed that the Iresistance |80 upon which resistance arm |13 acts as a potential divider in the screen grid .circuit of the tube |11, is so adjusted that theoscillator is not changed in frequency with changes in voltage supplied at B+ helping the frequency stabilization. Such an oscillator can, with temperature compensatedtuned circuits, maintain commercially a frequency stability of 20 vto 100 parts in a million, which is ample for tone fidelity. I can very easily arrange a pedal or other control, for moving the voltage arm |13 either way from its optimum operating point. For example, if a rocking pedal is arranged so that it changes this tap, it being situated in a stabilizing oscillator of the complex wave generator, it makes it possible for the musician to slur up or slur down the tones which are produced since the pedal thus changes the frequency up or down. A catch on the pedal would tell the musician where to leave it for proper pitch and for the best voltage stability of the oscillator. If the instrument should be used with another instrument not tunedto International pitch, this pedal,v

obviously, could be used lightly to raise or lower the pitch of an instrument according to my invention. It will be obvious, furthermore, that since in the form shown in Fig. the complex wave generator controls all of the notes of the instrument 'and all of the tones of an octave would be raised by the same amount, the intervals of the octave would not be as accurate as' they were before the raise was made, yet the octave relations would still be maintained and considerable change in pitch could be made before the intervals were noticeably oi."

It is to be understood further, that each octave may have its own complex wave generator or each manual its own complex wave generator in rangement I may easily give a stop a slightly different quality in the different octaves. We know, for example, that the quality of the tones produced by many instruments, as, for example, a clarinet, changes greatly throughout their ranges. Accordingly, if we wish to reproduce clarinet tones faithfully on an instrument ac cording to my invention over its entire range, we simply alter the harmonic content of the clarinet stop in the different octaves.

It will be readily understood by those skilled f in the art that it is possible with my invention to build an organ-like instrument which will cover the entire range of pitch and timbre with 25 pulse generators or less, and one complex wave generator employing the principle involved in automatic telephone switching. 'An organist has ten fingers and two feet, so that there never are more than ten keys and two pedals down at the same time. Accordingly, the number of pulse generators in use at a single time is never more than twelve. Using an automatic telephone switching arrangement, when a key or pedal is pressed, it can be made vto do three things in In such an ar-- six octaves of tones.

follow any synchronizing frequency which follows within the audio range of its natural period. For example, if the natural frequency of a pulse generator is 20,000 cycles per second, it will follow any synchronizing frequency between 20,000 and 22,000 cycles per second. This will cover I would also like to point out; thatI can key to permit the different harmonics in a tone to have different attack and release times or to cause some harmonics to exist only at the time the tone is initiated. This enables me to obtain effects such as the scrape of the bow of a violin, enabling me to obtain the effect of an actual instrument being played.

Referring now to Fig. l1, I have shown diagrammatically the keying arrangement for the instrument assembly shown in Fig. 10. Tubes `|8| |82 nd |83 are pulse generators connected by conductors |84, |85 and |86 to their corresponding frequency locking oscillators. The vacuum tube |81 is the frequency changer tube, the output ofA which is connected to the amplifier by conductor |88. The key |89 is in the off position. The keys |90 and |9| are in the sounding position. It will be seen that the sounding tube |81 vis subjected to the output of the complex wave generator through conductor |94.

A keying arrangement is illustrated in Figs. 12, 13 and 14, though it is to be understood that -any suitable keying means may be employed. The keying arrangement shown is of particular advantage since it will eliminate any clicks or thumps. The coupling is capacitative.

vReferring now to Fig. 12, secured to a suitable support |95 adjacent the key board are a plu- 4 rality of hanger members |96 comprising a pair sequence, namely, first, pick a pulse generator of arms |91 and |98 terminating in inwardly directed pintles |99 and 200, as can readily be seen by reference to Fig. 13. A housing 20| made of any suitable dielectric material is mounted for rotation on thel pintles |99 and 200. Pins 202, 203 and 204 pass through the housing and terminate in flattened heads 205, 206 and 201. The interior periphery of the housing is covered with A body of The key proper 2|0 is pivoted in any suitable manner on a support 2|| and carries an extension '2|2 which is normally slidably mounted in the end of the key. The end of the extension 2 I2 is pivotally secured to the housing atan eccentric point thereon. The action is such that when the key is depressed the housing is rotated around pivots |99 and 200.

In thel position shown in Fig. 14, the key is in the off or non-sounding position It will be noted that in this position the body of mercury 209 capacitatively couples pins 204 and 203. In this position, by reference to Fig. 14, it will be noted that the output of the pulse generator yis not fed to the grid |92 but is grounded. As the key is depressed, the housing is rotated and the body of mercury, due -to its weight, `wi1l remain stationary. In the depressed position, pins 202 and 203 are bridged by the mercury, though separated therefrom by a thin layer of dielectric material 208. This capacitatively couples the two pins 282 and 203 since they terminate in attened portions which act as the plates of a condemerandatthesametimeuncouplestheoutput of the pulse generator from ground. In this position the output of the pulse generator is coupled to the grid |02 d the changer tube |01. It will be further noted that.

pin 2li/is connected by conductor III to ground.r The pin 203 is connected by conductor 2M tol the plate output of the pulse generator. The

pin 202 is connected by conductor 2li to the nid of the frequency achanger tube.

Prom the foregoing it will be readily appreciated that my invention makes it possible to change the Afundamental frequency` of a complex wave aidat the same time maintain the original harmonic content of the complex wave. I have disclosed one means for generating a complex wave harmonically. It is understood, of course, that any suitable electronic means for generating a complex wave may be employed for impression upon the grid of the frequency changerI tube.

The human voice is the most versatile and flexible complex wave generator known. Words spoken are made up largely of frequencies harmonically related. Some of the frequencies contained in speech are made with the teeth, throat and parts other than the vocal chords. The vocal chords generate waves rich in harmonics. This harmonic content is controlled by the use of the mouth, the throat, the tongue, the teeth and the lips. A person speaking in a monotone, say with a fundamental frequency of 200 cycles per second, will emit largely frequencies harmonically related to a wave whose frequency is 200 cycles per second. In other words, the vocal chords vibrate in such case at 200 cycles per second and by changing the shape of the oral cavity and the throat resonance to the dierent harmonics of 200 cycles per second is controlled, imparting different character to the sounds.

Considering further the example of the person speaking in a monotone at 200 cycles per second, the frequencies generated will be 200 cycles persecond, 400 cycles per second, 600 cycles per second,.800 cycles per second, and so on. The y magnitudes of these harmonics will, of course, vary as the person makes different sounds, but the frequencies will always be 200 cycles per second and multiples thereof as long as the voice continues in the same monotone and disregarding the extra sounds which are not necessarilyharmonically related tones.

In carrying out my invention, I may employ the human voice as the source of origin or complex waves whose shape is imparted to an electrical wave whose frequency may be changed. An instrument of this character makes it possible for a musician to play it as an organ having a large variety of tone qualities. It makes it possible for the musician to exercise control over the attack and release of tones. He can make such an instrument, according to my invention, sing. and hum as a chorus of voices in which the sounds are words. shading and slurring to the music. He can make the voices have any tone quality he may desire over the entire pitch range, which range incidentally will be much greater than that of any human voice. He can make such an instrument He can impart delicate..

whistle. An organ in which the human voice is 75 a,an,s7o

`the origin of the-complex wave, and built in accordance with my invention, is shown diagram-` matically iny Fig. 15. The person whose voice is the origin of the, complexwavespeaks into a microphone lll in a monotone. The microphone changes the complex wavev into a similar electrical complex-wave which isampliiled in amplifier 2li. The ampliiledcomplex electrical wave is fed to a modulator Ill, intowhich, too, is fed a pulse wave from pulse generator ZIB. 'I'he pulse wave has essentially a uniform harmonic content. The output of the modulator will be a large number of frequencies caused by all of the harmonics beating together. Assuming, for example, that the frequency of the pulse is 20,000 cycles per second'and only certain. of the generated frequencies are desired, the filter 22| will reiect the unwanted frequencies and pass only those frequencies. caused by the sum of the fundamentals. YConsidering that the monotone human voice frequency is 200 cycles per second, the sum of the pulse frequency and the voice frequency will be-20,200 cycles per second. The snmofthesecondharmmicswillbe40,400cycles per second; the sum of the -third harmonics will be 60,600 cycles per second, and so on. `Those frequencies which pasa through the filter make up a complex wave whose fundamental frequency is 20,200 cycles per second.

I recognize that the pulses generated by the pulse generator do not contain all harmonics in equal magnitudes. The filter 220 is so designed that it will attenuate some frequencies which it passes more than others to compensate for the dencienx's in the attenuated harmonics. I can control the nlter, furthermore, by means operated by the musician in order that the quality of the singing voice issuing from my instrument may be altered if desired by the musician. The complex wave coming from the illter 220 is now treated precisely the same as the complex wave electronically generated. If the musician presses the key for A, having a frequency of 440, then a pulse from the pulse generator 22| generated at a fundamental frequency of 20,660 cycles per second, is fed t0 the frequency changer 222. 'Ihe complex wave from the filter is fed through channel 223 to the frequency changer for scanning by the pulse wave as heretofore described. When the pulse wave of 20,660 cycles per second scans acomplex wave whose frequency is-20,200 cycles per second, the resulting frequency will be 440 cycles per second, and multiples thereof, assuming, of course, that the illter 220 isproperlyH adjusted.

'I'he series of pulses having an envelope similar to the complex wave being fed to the frequency changer and a wave whose frequency is 440 cycles per second, is fed from the frequency changer through channel 224 to the ampliiler 225 and thence to the loud speaker 226 v of the instrument.

'I'he modulator 2li is one such as the squarelaw modulator in order that it does not produce distortion.

Suppose now that a woman musician wishes to speak into the microphone 2|6 and her voice is pitched higher than 200. Let us say, for example, that it is 400 cycles per second. She merely turns the control on the pulse generator 2|9 to lower its fundamental frequency to .19,800 cycles per second. This is very simple to do since the pulse generator will oscillate at a frequency controlled by the locking or control oscillator whose frequency can be very simply adf asumo usted. With the pulse generator thus adjusted when the woman speaks in a monotone at 400 cycles per second, the harmonics of her voice are 'added to the harmonics of the pulse giving a fundamental of 20,200, and harmonics of 40,400, 60,600, 80,800, and the like, exactly as was the case when the voice whose fundamental was 200 cycles per second spoke into the microphone. The words can be spoken at any frequency monotone, though' preferably not below say 200 cycles per second because of the demands made upon the filter 220 in rejecting unwanted frequencies that arev close in frequency to those that it is desired to pass.

In order to insure the musician that his monotone is properly tuned with the pulses from the pulse generator 2I9, I provide an'integrator and rectifier 221, the output of which passes through channel y228 to a cathode ray tube 229, sold under the trade-name Magic Eye."

Ii' the musician slurs his monotone up, say

20 cycles per second, making it 220 instead of 200, and the pulse generator2l9 is left at 20,000 cycles per second, then the frequency A coming from the loud speaker will slur up from 440 to 460, making it possible for the musician to slur tones. This, of course, must be used by him with care for if he is playing a. chord which he slurs up 20 cycles per second from 200 cycles per second, he will add 20 cycles per second to all of the tones in the chord, tending to upset the harmonious relation that these tones previously had. This factor, furthermore, enables the musician to obtain weird or blue" effects not heretofore possible. The slurring, furthermore, can be readily tolerated within limits and in fact may be beneficial in tending to temper the exacting of pitch which is inherent in electronic instruments, and which ultimately makes them tiresome.

If the musician wishes to play the instrument without supplying a complex wave to the micro-` phone, he merely turns the frequency control on the pulse generator 2I9 so that its fundamental frequency is say 20,200 cycles per second, or any other predetermined frequency, and the instrumentls then ready to be played with the pulse itself as the complex wave. Its harmonic content can be controlled by the controls on the filter 220.l

I may have a console of more than one man'ual with a complex wave generator for each manual. With this arrangement the musician can play singing or humming voices on one manual and the accompaniment on the other manual or manuals. used, each manual. may be arranged as shown in Fig. 15, so that two people can speak into the Wheratwo or more manuals are tones and the human voice.

two microphones and the singing controlled on or a special vibration pick-up may be-used to pick up tones from a violin or other instrument.'

If the violin is used, one tone is constantly bowed. This may be done mechanically if desired. The musician keys the instrument to produce the music.- A single stretched string, thereioreg'may be employed which may if desired be struck as 'in a piano or plucked. It: tone may then be as the origin of the complex wave.

of the filter, it is possible to reproduce the mu-f picked up by the microphoueymaking it possible to obtain percussionl effects. Similarly, chimes, bells and the like may beused as the origin of the complex wave. Countless,rv effects may be achieved by controlling they harmonic content in filter 220. v v v I havev mentioned above that words spoken were not made up entirely of harmonicallyrelated4 frequencies. 'I'he vowel sounds of the voice are made up of harmonically related `frequencies and these sounds really make vocal music. By proper yfiltering the unharmonic sounds may be reinserted after the lvowels have gonethrough the frequency changer.

Care must be ltaken in an instrument using an independent sound source and a microphone, that sounds do not feed back from the loud speaker -to the microphone causing a howl,suc h as is experienced in public address systems. Directional microphones that are relatively insensitive to anything but sounds made directly into them may be employed.

By means of my invention I am enabled to give the technique of known instruments such as the pipe organ, piano, chimes, bells, harpsichord,rviolin, and the like, as well as synthetic In using the human voice as the origin of the complex wave, the

musician simply speaks in a monotone into the vwill be emitted from the loud speakers of the organ. If two keys are pressed while words are voiced by the musician, the organ will produce two different pitches of the same voice, so that a single musician may sing a duet, quartet, and the like, employing but a single voice By means sicians voice with fidelity or to give it another timbre by emphasizing some harmonics and attentuating others.

If two people talk into the microphone at the same time it will serve as two vcomplex Wave generators, even though they say different words the organ. will producevthese words in singing if the keys are played properly. Any number of voices may be used if desired, adding richness due to human inaccuracies in keeping pitch.

It will be seen that I have accomplished the 'objects of my invention. An instrument according to my invention is all electronic and may employ vacuum tubes which are dependable, inexpensive, uniform and have long life. 'I'he circuits are made principally of inexpensive resistors and condensers and -are as a whole not critical in their adjustments, so that they can be serviced by a radio repair man without special.

stops is easily obtainable. Keying does not cause objectionable transients, and one method of keying offers two attack times, two release timesv and second touch. All of the tone producing circuito are made of standard radio parte which are produced in mass production for the radio industry, making them available for the manufacture of my instrument at a low cost.

Every tone produced has true harmonics and not tempered harmonics, et the tones may be synthesized in a simple anner. Every tone in the scale may have exactly the same quality or the harmonic content may be made to vary from octave to octave to produce substantially the same quality variations as other .musical instruments do. The efficiency ofthe instrument is increased as the number of stops is increased and as the number of manuals is increased. All of the tones may be caused to slur up or down by a `Vsingle control, making trombone smears,

and innumerable effects possible. The variation in pitch of all of the tones in a simple manner makes it possible to tune my instrument with a xylophone, piano or other instrument not at International pitch.

The entire instrument may be given a vibrato or combined vibrato and tremolo with a single vibrator or one tube. The vibrato may be applied to individual manuals or, if desired, to each interval of the octave. The vibrato may be slightly different on each to give an enrichening effect. .When the vibrato is applied to the complex wave generator, a celeste effect is obtained.

The human voice or other complex wave generating means may be employed as the origin of the complex waves, providing an instrument which will challenge all of the talent and imagination which musicians possess.

It will be understood that certain features and subcombinations are of utility and may be employed without reference to other features and sub-combinations. This is contemplated by and is within the scope of `my claims. It is further obvious that various changes may be made in details within the scope of my claims without departing from the spirit of my invention. It is, therefore, to be understood that my invention is not to be limited to the specific details shown and described.

Having thus described my invention, I claim:

l. A method of generating musical tones having a desired wave form and desired audible frequency comprising the steps of generating an electric wave having the desired wave form at a frequency relatively high with respect to thedesired audible frequency, generating a succession of electrical impulses, impressing said impulses upon said electrical wavev at a frequency which will produce an electrical wave having a component which is of the desired wave form and desired audible frequency, and converting said component into the desired musical tone.

2. A method of generating musical tones including the stepsof generating a complex electrical wave of a predetermined wave form at a super audio frequency, generating a succession of pulses at a frequency different from. the freplex wave by a number of ycles equi pitch of a desired musical tone, means for impressing saidpulsesuponsaidcomplexwave, produce a series of pulses having an envele which is a wave 'having the wave form of complex wave and a frequency which is that the desired pitch within the audible rante, means for amplifying the electrical wave of audio ftequency, and means for converting said electrical wave into a sound wav'e of the desired pitch and timbre.

4. An electronic instrument including in ccmbination a complex wave generator for generating av complex wave of. a predetermined wave format ay super audioirequency, a plurality of pulse generators for generating a plurality of series o! impulses at respective frequencies different from the frequency of said complexwave by respective amounts equaling the frequency of various notes in a musical scale..key controlled means for selectively impressing said impulses upon said complex wave, said key controlled means including a thermionic mixer tube the output oi' which includes an electrical wave having the wave form of the complexy wave and a frequency within the. audio range determined by the keys, means for amplifying the output of said mixei' tube, and means for converting the amplified electrical wave of audio frequency to a sound wave of the desired pitch and timbre.

5. An electronic instrument including in combination a vacuum tube connected in an oscillating circuitto produce a continuous'wave rich in harmonics and at a predetermined super audio frequency, a filter having a plurality of parallel tuned circuits, -tuned respectively to the fimdamental and harmonics of said continuous wave,

means for controlling the amounts of thefundamental and respective harmonics passed by said filter, the output of said filter being a continuous wave of a predetermined wave form at a predetermined super audio frequency, vacuum tube means for generating a series of positive impulses at a super audio frequency higher than said complex wave frequency by an amount equal to the frequency of 'a desired musical note. a thermionic mixer tube, having a grid, means for impressing the continuous complex wave upon the grid of said tube, key controlled means for impressing the series of pulses upon the grid of said tube to produce an electrical wave having the form of the complex wave and a frequency equal to the desired musical pitch, means for amplifying said electrical wave of the desiredy wave form and musical frequency, and means for .converting the amplified electrical wave into a sound wave of the desired timbre and pitch.

quency 'of said complex wave by a number of cycles equal to the desired pitch, impressing said pulse wave upon said complex wave to produce an electrical wave having a component which is ofthe form of the complex wave and at a frequency of the de'sired pitch, and converting said component into the desired musical tone.

3. An electronic musical instrument including in combination a complex wave generator for generating a wave of a predetermined wave 4i'orm at a super audio frequency, an^oscillator for generating a series of positive pulses at a frequency different from the frequency of said comfor feeding said complex wave to said thermionic `mixer tube, key controlled means for selectively,4

feeding said series of positive pulses to said mixer tube, means for generating a vibratory continuous wave at a low frequency.. means for selectively feeding the low frequency vibratory wave to said thermionic mixer tube whereby the output of said mixer tube will comprise an electrical wave` of the wave form of said complex wave and at a frequency of the desired musical pitch 7. An electronic musical instrument includingl in combination means for generating a con-y tinuous complex wave of a predetermined wave form at a super'audio frequency, a thermionic mixer tube, means for feeding said complex wave to said mixer tube, a plurality of thermionic tubes connected in respective circuits for profrequency ofsaid pulses and thefrequency of said sound waves, a filter for rejecting undesired frequencies. and harmonics thereof, the

output of the filterbeing a modulated series of pulses of a vpredetermined frequency and a preducing a series of sharp positive electrical pulses at respective frequencies higher than the frequencyof said complex wave by respective amounts equal to respective pitches of musical notes, a plurality of keys, a plurality of dielectric housings, separate means for mounting said housing for rotation in response to the actuation of said keys, three conductors entering each of said housings, means for electrically connecting one of said'conductors to the output of a pulse generator, means for connecting the second rof said conductorsto ground, means for. connecting the third of said conductors to said thermionic mixer tube,a body of mercury in said housing, said body of mercury normally capacitatively coupling the output of a pulse generator to ground, the construction being such that upon the actuation of its respective key, a housing will rotate to move said body of mercury to a position capacitatively coupling thel output of said pulse generator to said mixer tube to produce an electrical wave having the wave form of said complex wave at a selected musical frequency, means for amplifying the output electrical wave of the mixer tube, and means for converting said amplified wave into a corresponding sound wave having the desired pitch and timbre. f

A 8. In an electronic musical instrument. a microphone adapted to convert sound waves into electrical waves, thermonic tube `means for creating a plurality of positive pulses at a predetermined frequency, means for modulating said pulses by said electrical wave corresponding to said soundwave, the output ofl said modulating means being a series of pulses having an envelope ofthe Wave form of said sound wave and at a predetermined frequency equal to the fre-4 quency of said pulsesplus the frequency of said sound wave, means for generating a second series of pulses of uniform height and at a frequency higher than the output frequency of said modulator by an amount equal to the pitch of a desired musical tone, a thermionic mixer tube.

determined wave form, a second pulse generator for creating a plurality of electrical impul'sesat a frequency higher than the filtered pulse frequency by an amount equal to the pitch of a musical tone, a thermionic mixer tube, means for feeding the output of said filter and the output of said second pulse generator to said mixer tube to produce an electricalwave having the wave form of the filtered wave anda frequency equalv to the pitch of the desired musical tone, means for amplifying the mixer tube output wave, and means for converting said amplified electrical wave into a sound wave having the form of said filtered wave and a lpitch equal vto the desired musical tone.

10. An electronic musical instrument as in claim 9 in' which there are a plurality of second pulse generators for generating a plurality of series of pulses at frequencies higher than the filter pulse frequency by4 respective amounts `equal to the frequency of the notes of a musical scale, and key control means for selectively im pressing the outputs of the respective pulsegenerators upon said mixer tube.

11. In an electronic musical instrument, means for generating sa complex wave at super audio frequency, including in combinationan oscillator producing a continuous wave rich in harmonics at thedesired -frequency, a filter including a plurality of Iparallel tuned circuits tuned respec-A tively to the fundamental and various harmonics of said frequency', resistors connected across said parallel tuned circuits, means for selectively shunting a portion of the respective resistors whereby to control the harmonic content of the means for feeding the output of said modulator 9. An electronic musical instrument including allel tuned circuits, said voltage dividers govern-- 1 said voltage dividers to produce predetermined in combination a microphone for converting aA sound wave into a corresponding electrical wave K of a predetermined wave form and pit ch, means for amplifying said electrical wave, means for generating a series of electrical pulses at a super audior frequency, means for 1 modulating said series of pulses by said amplified microphone electrical wave to produce a modulated series of Wave passed by the filter; key controlled means for changing the vfrequency of the filtered wave to a predetermined frequency within the `audio range, means for amplifying the audio frequency electric wave, and means for converting the ampliiied electric wave into its corresponding sound wave.

12. An electronic musical instrumentas in claim 11 in which said shunting means comprises a plurality of switches, and common means for operating all of the switches simultaneously.

13. An electronic instrument including in com'- bination an oscillator connected in a circuit to produce a continuous wave rich vin harmonics at a predetermined super audio frequency, a. filter connected across the output of said oscillator comprising a plurality of parallel tuned circuits tuned respectively to the fundamental and the various harmonics 'of said predeterminedifrequency, voltage dividers -across each of said par-i ing the respective harmonic components of the output wave of the filter, a plurality of stop means operable by the musician for operating complex waves, key controlled means for selectively changing the frequency of the. filtered complex wave to a predetermined frequency within the musical range, and means for converting the electrical wave within the musical range to a musical note of a pitch determined by the key controlled means and a timbre controlled by the stop means.

14. An electronic musical instrument including

Images