US2094244A - Sound reproducing film and a device for recording sounds photographically - Google Patents

Description

p 7- A. POULSEN, ET AL 2,094,244

H SOUND REPRODUCING FILM AND A DEVICE FOR RECORDING SOUNDS PHOTOGRAPHICALLY I Original Filed Jan. 25, 1932 @92 s 'ZQl L,

' mpcrophone A i n'r' m mum- A M n M AV i L -J 3.

flRNDl-D POUI-SEQ.

,lji 8.

Am. cam. atone erme- Patented Sept. 28, 1937 UNITED STATES SOUND REPRODUCING FILM AND A DEVICE FOR RECORDING SOUNDS PHOTOGRAPH- ICALLY Arnold Poulsen andAxel Carl Georg Petersen, Copenhagen, Denmark Application January 25, 1932, Serial No. 588,792

Renewed February 17, 1937. In Germany February 26, 1931 In the reproduction of tones (sound oscillations) recorded on films it is well known that a more or less prominent tingling or crackling, the so-called back ground noise is produced' Part of this noise may originate from the microphone or amplifier used in the recording, or from the light-sensitive cell or amplifier used in the reproduction process. By careful construction, however, of the amplifier, the microphone and the light-sensitive cell thebackground noise due to these parts may be reduced to a very small value.

In addition to the above mentioned causes the background noise may originate from irregularities of or impurities in the layer of emulsion on the film, or from so-called scratches in the same, which unavoidably are produced by wear. The prominence of noise due to the last mentioned causes depends on the transparency of the sound reproduction film after it has been developed and fixed. Scratches and irregularities in the emulmanifest themselves as background noise, when a relatively large part of the sound films used for the reproduction is translucent.

The background noise originating from the emulsion irregularities and from the scratches in the film is generally independent of the magnitude of the sound amplitudes and, consequently,

becomes more prominent with decreasing ampli-- density of the negative film with decreasing sound amplitude and, consequently, the corre-' sponding density of a positive film, i. e. a reproducing film copied from said negative film in any ordinary manner will increase with decreasing sound amplitude. The transparency of the reproducing film will be reduced when the sound amplitudes decrease, and the background noise will then be reduced at the same rate.

Theaverage density of the film does not necessarily have to be varied during the sound recording itself. An ordinary negative or positive sound film of the variable density typ on' which .the sound has been recorded in the usual manner,may be used and a new positive or negative film produced therefrom by conversion of the sound recorded on the original negative film into corresponding electric currents, by means of which the sound is recorded on the new film with varia- 55. tion of the average density of the film.

sion coating of the film will therefore greatly.

ably reduced" by decreasing the average degree of 4 Claims. (01. 274-41.6)

The above mentioned variation of the average density of the photographic film is equivalent to a displacement of the working point on the socalled blackening curve of the light sensitive layer of the fi1mi. e. or a curve'representing diagrammatically the dependency between the intensity of the exposure of the film and the corresponding degree of blackening of the developed film. This displacement is effected so that a portion of the varying electric current, into which the sound is converted for the purpose of being recorded, is in known manner rectified and caused to act on the zero-position of the sound recording member, i. e. the position about which the said member oscillates during the recording or on the mean value of the sound recording bundle of rays-i. e. the value about which the intensity of the bundle of rays of light is varying in accordance with the sound oscillations.

' It will be suitable accordingto the invention .to govern the sound-recording member (the light displacement of the working point as quickly as,

or more quicklyfithan the increase in the magnitude of the sound amplitude concerned,awayfrom the position occupied by the working point during the pause-which generally will mean the position corresponding to full blackening of the sound-record track.

The sound concerned will therefore be recorded with certainty, and when a sound dies slowly away the density will vary so slowly that it cam not reach the maximum value during the pauses, before the audible oscillations to be recorded have ceased.

This'adjustment of the controlling system for a sound-recording member can be attained simply by mutual adjustment of the coeflicients of conductivity for the circuit or one or more of the circuits, by way of which the above mentioned rectified current is transferred to or is caused to actuate the sound-recordlng member. plished in such a manner that the variations in the rectified current-are effected. more quickly This is accomwhen the said current increases than when it decreases.

Usually, a rectifiervalve is'used for rectification of .the above mentioned current and thecontrol may be efiected by inserting in the anode circuit of the rectifier valve a smoothing condenser con- 1 nected'in parallel to an ohmic resistance. The value of the resistance is so adjusted depending on the ohmic resistance of the rectifier valve, that the charging of the condenser during the passage ofa positive half wave through the rectifier valve will be effected more quickly than the subsequent discharge of the condenser during the occurrence of a negative half wave. I

The details of the invention-appear from the following description of the same in connection with the drawing showing curves and diagrammatic constructions of recording systems according to the invention. I

Fig. 1 shows the blackening-curve of a negative tone film and the position of the working point on this curve, for recording sounds according to the present invention,

Fig. 2 shows diagrammatically, a system constructed, according to the invention, for exposurev of a film during recording of tones (sound oscilla tions) thereon according to the varying intensity 4 method,

Fig. 3 shows in front elevation, an exposure slit 'used in this system, and

I Fig. 4 shows a system for controlling the recording member in such a manner'that the working point of the blackening curve is removed .farther away from a given initial position the greater the tone amplitudes are.

Fig. 5 diagrammatically represents the shifting of the zero line during the increase and decrease of the sound amplitude;

. Fig. 6'is an .outline of a variable width record prepared in accordance with the present invention;

Fig. 7 represents the variable width rec'ord prepared in accordance with the invention and '.Fig.' 8 represents a variable density record corresponding to the one shown in Fig. '7.

In Fig. 1 a is the blackening-curve for instance for a silver-bromide emulsion applied to a nega- -tive film. The abscissae J represent the intensi ties of lighting, and the ordinates S represent the degrees of blackening measured as a percentage of the maximum density, which latter corresponds to complete opaqueness of the light-sensitive coating, after this has been developed and fixed.

When recording sounds according to the wellknown variable density process, the point C' Of the blackening-curve, at which the work is being performed, i. e. the point on the curve on both sides of'which the blackening varies uniformly during recording, is a fixed point of the curve, irrespective of the magnitude of the tone amplitudes.

As shown in Fig. 1, the point c is situated at a relatively low point on the curve d-ordinarily at the central point of the rectilinear portion of the curve w-s'othat onthe negative film .the sound records corresponding to sound oscillations of smaller amplitudes appear relatively very opaque.

I creases.

Disregarding temporarily the ordinary variations in the exposure due to the variations of the varying average lighting of the negative film during the recording of sound oscillations of various amplitudes, then it will be seen directly that the light passing through the negative film during the printing of the same will vary from about 100% to aboutv50%, or, as the point 0 is not situated amplitudes of the sounds and considering only the quite as far down as the initial point of the curve a, from for instance corresponding to the point 0 and to about 50% corresponding to the central point 0 of the-rectilinear or nearly rectilinear part of the negative blackening curve.

By blackening in the present specification is meant not what is ordinarily understood by this word. If the blackening is indicated by S and the amount of light impinging on a certain area of the film is indicated by I and the portion of said light penetrating or passing the film is indicated by P,

i. e., s is. the ratio 'between the amount of light impinging on the film and the portion of said light which is absorbed by the film. In Fig. 1 of the drawing S is a percentage of I.

When recording sound oscillations on films according to the variable density -method, the sound, as is well-known, is converted into varying electric current, wl ich is caused to act upon a system by means of which the film is exposed in such a manner that the intensity of exposure of the film is modified in accordancewith the current. In Fig. 2 a system of this nature is illustrated, in which an oscillating mirror e is used which issuppor-ted by an oscillograpli loop 1 and is exposed, through a-slit g in a screen h, to a source of light 2 an image of which is formed in the plane of the mirror by means of a lens 111.

The mirror projects animage-g' of theslit g on to a screen I disposed between the mirror and-the film la, in such a manner that the image g; partly covers a slit m in the screen Z. A cylindrical lens n produces on the film; k a linear image i1 of the image of the source of light formed on the mirror by the lens m'.-

Assuming that the-mirror swings about the longitudinal axis of the p f. the image 9 will be caused to cover alternately more or less of the slit m and, consequently, the part of the bundle of light reflected from the mirror e and passing through this slit will vary in accordance with the oscillations of the mirror. In other words, the quantity of light concentrated on the film k by the lens n and, consequently, the light intensity of the image i1 will vary in accordance with the oscillations of the mirror.

According to the well recording processes, the mirror, when at rest occupies such a position that one edge of the image known variable density .g coincides approximately with the center line q '75 of the slit m. Acording to the present invention edge q of the slit m, so that the intensity of the lighting on the film will' correspond to the point 0 of the curve a, Fig. 1. When sound oscillations act on the sound-recording device, then the edge of image 0' that originally was situated along the edge q has to pass the edge of the slit m to an extent increasing with the amplitudes of the sound oscillations, in correspondence with the above described fact that the point 0 has to b displaced on the curve a. 4

This displacement may, for instance, beeffected in the manner illustratedin Fig. 4, in that a portion of the current, into which the sound is converted for the purpose of being recorded, is directed across a transformer T to a rectifier valve 8, in which the current is rectified and amplified, and then directed across a resistance t inserted in parallel with a condenser u in the anode circuit of the valves. The anode current of the valve s produces voltage variations across the resistance t, which are impressed upon the grid of an amplifier valve 1). The anode circuit of this amplifier valve o is closed by way of the oscillograph loop f, Figs. 2 and 4, to which the secondary winding of a transformer T is conperform in order to be able to record the sound will depend on the alternating current supplied by way of the transformer T of the loop. The initial voltage on the grid of the valve 12 is selected in such a manner that a relatively high plate ourrent will flow in the valve uwhen the current rectified bythe valve s is small, i. e. when the sound oscillations occurring are weak. The mirror e is then rotated at a relatively wide angle, for example to such an extent that the edge of the image 9" will be situated only slightly inside of the bottom edge of the slit m.

The exposure of film k is therefore very weak,

corresponding to a very considerable blackening of a positive film produced by copying from the film It. When the sound amplitude increases, 1

then the current rectified by the valve s will also increase, and the drop of voltage across the resistance t will reduce the initial grid voltage in the valve 0, whereby the anode current of the latter will be 'reduced. The mirror is thus turned so that the image y will be moved further across the slit m.

In the above it was assumed that the edge of the image g, when the mirror is at rest, is situatedat the bottom edge of the slit m or only slightly below the same. When the edge of the ating impulses surpass a certain minimum value.

In the above the-assumption was that the displacement of the working point 0 was proportional to the amplitude of the sound.

With no direct current from the anode circuit of the valve 12 passing through the loop 1, the

mirror is adjusted to such a position that the amount of light passing the aperture m is sufiicient to expose the film to a degree corresponding to a density of per cent. The valve 1: is then switched into circuit and the anode current is adjusted tosuch a value that when passing through the loop ,f, the mirror is moved to the position in which the upper edge of the crosssection of the light beam 9 .is at the lower edge of the aperture m. When the intensity of the sounds to be recorded reaches a maxi'mum,'the current from the rectifier s will be a maximum thus causing a maximum negative biasing potential to be'applied to the grid of the valve 17. The plate current of that valve will thus become zero. At values between zero and the maximum value of plate current in the valve 11, it will be understood that the light beam g will take up corresponding positions in relation to the aperture m.

As already stated, it is preferable that the mean density of the record corresponding to the loudest sound to be recorded should be not greater than 50%. To avoid distortion of loud sounds, the apparatus will in any case-be so designed that, at an assigned maximum sound intensity,- the total extent of the oscillation of the edge of the cross-section g of the beam of light, consequent on the sound vibrations, is not greater than the width of the aperture m, and it is only necessary,

to obtain the above result, that the maximum (or initial) value of the continuous current flow ing in the anode circuit of the valve V shall be not greater than the maximum peak value of the alternating current in the secondary winding of the transformer, andthat the continuous current shall be zero at the assigned maximum sound intensity. 7 If these conditions are satisfied it. will be evident that the mean position of the edge of the beam of light will be at or near the middle of the aperture m when the assigned maximum sound intensity is reached. It is to be noted that if this assigned maximum should be exceeded the I mean position of the beam of light is not altered,

because the anode current has already been brought to zero. This would not be the case if the anode current were varied directly as the intensity of the sound, as in this case an exceptionally loud sound, would cause an exceptionally heavy anode current which might move the beam' of light off the aperture m altogether.

The nature of the sound record produced in accordance with the present invention will more ciearly appear from Figs. 5-8.

In Fig. 5 the dotted line 0 represents the zero line of an amplitude record. At the left side we have shown a few sound waves, the volume of which is suddenly increased from d to c and at the same time the zero line is displaced from the zero point a: to the zc:o point :c. It is assumed that prior to the increase the amplitude 12 of the waves is constant and that after the increase the amplitude c of the waves is also constant until it again decreases at the zero point :c' The zero line is shifted from the zero point a: to the zero point :r' in a period of time represented by the distance A measured in the direction of the movement of the film. This direction is indicated by the arrow P.

- The variation of the amplitude of the regis- 15 and at a certain rate (A or A1).

tered variations, i. e., the diiference c-:-d, takes place within the period of time represented by A. At the right-hand side of Fig. 5 we have shown what happens when the volume of sound (16-- 5 creases from the magnitude it (which is equal to c) to the magnitude b(which is equal to d). It is assumed that the variation a+b representing the de'creaseof the volume of the sound takes place during the time represented by A1, i. e.,

the horizontal distance between the two zero points :12" and $4.

If A1 is equal to A and c:-d is equal to a-:-b, then a.+b and c:-d, respectively, represent a variation of a certain magnitude (a+b or 0+d) 7 It will be. clear that the zero point a." is not shifted back to a position a:' corresponding to the original position-cc within the period of time A1, but considerably more slowly, viz., in the course of the time represented by B. The shift of the zero line caused by the increase is equal to :I:+:I:, and the shift of the {zero line during the decrease and within the same period of time is equal to a:"-:a:,' and this shift is considerably smaller in the transversal direction of the film than ;v'+a:. For the sake'of simplicity it is assumed that the volume of the sound represented in the right-hand portion of Fig. 5 decreases from a to b in the course of two consecutive waves to and 101. After this the volume bis kept constant;

Fig. 5 illustrates that in accordance with. the

present invention a certain variation of the vol-' ume of the sound is not only of a certain value (a+b and c+d) with respect to the amplitude variation, but also of the same value with respect to the time (represented by A and A1, respectively) ,-within which said variation takes place.

A variable density record according to the invention is shown in Fig. 8. Since the graphical 40 identification of the inventionis well-nigh impossible in such illustration, the variable width counterpart of this record is shown in Figs. 6 In practice such translation of-variable and '7. width records into variable area records is -ef- 45 fected by special equipment well known to those I skilled in the art.

record and Fig. 6 a line representing the corresponding sound. It will be notedin Fig. 6 that be shifted more rapidly towards the center line 55 of the film than towards the border line.

However, two consecutive sound waves I and. II may Fig. '7 shows the corresponding variable width be found the maximum amplitudes of which have a certain diiference p, see Fig. 6, which 'is equal to the difference between two other consecutive sound waves occurring during the decrease of the sound, see the two waves III and IV in Fig. 6.

These two equal differences p appear within the same period of time A.

If the record is prepared in such a manner that the shift 'q of the zero position corresponding to the difference p during the increase of the sound, see the left-hand side of Fig. 7, is larger than the shift qi of the zero position correspondit is not possible in a scratching made by hand toindicate the differences in the average density by shifts with amplitude changes in the sound, and

in which said shift occurs more rapidlywhen the sound volume increases by a certain value than when it decreases by said certain value.

2. A sound record film in which the mean density of the sound track varies between the density corresponding to the central point of the one end of the rectilinear portion of the said' 7 characteristic.

3-. A sound record film on which the average transparency of the sound trackis shifted at a higher rate than thevolume of the sound varies, when the said volume increases, and is shifted at a lower rate than the variation of the volume of the sound, when the said volume decreases.

4. A sound-record film in which the mean transparency of the sound track-varies between the transparency corresponding to the central point of the rectilinear portion of the density characteristic of the said film and the transparency corresponding to one end of the rectilinear portion of the said characteristic.

ARNOLD POULSEN.

AXEL CARL GEORG PETERSEN.

Images