US2258124A - Communication system - Google Patents

Description

Oct. 7, 1941. H, J. NICHOLS COMMUNICATION SYSTEM Filed Sept. 20, 1959 5 Sheets-Sheet l I I G. n

(1111117777) 5517iiiiiiiiiiiiiiii7iii fiiiiiiiiiihiiiiliiliibiiiiiiriiiiiiiliiiPiii FIG. 2.

IINVENTOR.

A TTORNE Y Oct. 7, 1941. H. J. NICHOLS 2,258,124

COMMUNICATION SYSTEM Filed Sept. 20, 1939 5 Sheets-Sheet 2 FIG. 5b.

. lawrok.

' ATTORNEY Oct. 7, 1941. J NlCHCLs 2,258,124

COMMUNICATION SYSTEM Filed Sept. 20, 1959 5 Sheets-Sheet 3 llll llll jllllml I F IGJIO.

IN V EN TOR.

- A'TTORNEY 0d. 7, 1941. J NICHOLS 2,258,124

COMMUNICATION SYSTEM Filed Sept. 20, 1939 5 Sheets-Sheet 4 INVENTOR.

FIGS.

Oct. 7, 1941. H. J. NICHOLS 2,258,124

COMMUNICATION SYSTEM Filed Sept. 20, 1939 5 Sheets-Sheet 5 71a 86 O Q F IG- INVENTOR. fiwfl v :4 TTORNEY Patented oer. 1, 1941 UNITED STATES PATENT OFFICE COMMUNICATION srs'rm Harry J. Nichols, Binghamton, N. Y.

- Application September 20, 1939, Serial No. 295,723

11 Claims. (Cl. 178-11) The present invention relates to a communication system and, more particularly, to facsimile transmission systems and the like.

The invention embodies novel scanning and transmitting means for the production of facsimile whereby high precision and high speed of transmission are obtained.

More specifically, the device embodying the invention is provided for the purpose of supplying novel facsimile producing means including novel scanning means, novel means for producing a, signal varying precisely with every tone value corresponding to the respective elemental areas scanned and novel means for producing half tone facsimile reproduction.

Similar devices of the prior art have been subject to errors in scanning, have required the movement of the material scanned in both of two directions at right angles to each other, and have utilized an elongated, stationary source of light, producing uniform illumination of low intensity, instead of utilizing a moving beam of high intensity or such prior devices have necessitated the use of elaborate systems for moving the light source. Such devices have not only been cumbersome, but have failed to produce a sharply defined facsimile, due to errors in scanning and failure to produce a signal, varying precisely with every change in tone value of the respective elemental areas, or in the reproduction of graphic material, have failed to produce sharply contrasting records of light and dark tone values.

In View of the above conditions prevailing in the prior art, one of the objects of the present invention is to provide a novel facsimile system whereby the foregoing undesirable characteristics are eliminated.

Another object is to provide a simple and highly precise system for the transmission of graphic material such as pictures, writing, printing, documents, records and the like over various types of communication systems.

A further object is to provide a novel system adapted to record and reproduce in both full tone, that is in contrasting black and white, and-in half tone, that is where graduations of tones from white through gray to black are obtained.

Still another object is to provide a novel method of scanning wherein a beam of light is rotated about two axes at right angles to thereby produce straight line scanning of material moved over a curved surface.

' A further object is to provide novel scanning means comprising a relatively moving, intersecting slot and slit for defining sequential elemental areas and means for maintaining a moving beam of light in perfect coincidence with said intersec tion at all positions thereof. Another object is to provide novel means producing a signal varying with the variations in the tone values of the respective elemental areas scanned and means for producing a signal varying inversely with said values upon the scanning of every alternate line element.

A further object is to provide a novel system whereby sharp changes in tone values of the respective elemental areas scanned, such as in printing or similar graphic material, produce a signal varying as sharply as the tone value changes.

A further object is to provide a novel facsimile transmission system in which the signals occupy only a limited frequency band in combination with saturable means controlled by the evaluated tone values whereby the necessity of demodulatlng or rectifying the received signals is avoided.

The above and further objects and advantages of the invention will appear more fully hereinafter from a consideration of the detailed description which follows, taken together with the accompanying drawings wherein are illustrated several embodiments of the invention. It is to be expressly understood, however, that the drawings are for the purposes of illustration and description only, and are not designed as a definitionof the limits of the invention, reference primarily being had for this purpose to the appended claims.

Other objects of the invention will be pointed out in the following description and claims and illustrated in the accompanying drawings, which disclose, by way of example, the principle of the invention and the best mode, which has been contemplated, of applying that principle.

In the drawings wherein like reference characters refer to like partsthroughout the several views. v

Fig. 1 is a plan view of a simplified embodiment of the transmitting apparatus in accordance with the present invention, cut away in part to more clearly illustrate the interior mechanism.

Fig. 2 is a side view of the transmitting apparatus of Fig. 1 illustrating the scanning apparatus in further detail.

Fig. 3 is an end view of the transmitting apparatus, as shown in Figs. 1 and 2.

Fig. 4 is a detailed view of a portion of the scanning belt. a

Fig. 5a is a diagram illustrating the manner in which the rotary mirror imparts motion of translation to the scanning beam.

Fig. 5b is a diagram illustrating the normal deviation from perfect linearity of the path of motion of the scanning spot in devices of the prior art.

Fig. 5c is a cross section taken on line VcVc of Fig. 50 showing one means for correcting the normal deviation illustrated in Fig. 5b.

Fig. 6 is a schematic circuit diagram showing one arrangement for producing facsimile signals in accordance with the invention.

Fig. 7 is a similar diagram illustrating another embodiment of the signaling arrangement and a circuit reversing mechanism connected thereto.

Fig. 8 is a similar diagram illustrating another embodiment of a signaling arrangement particularly adapted to produce signals representing a full-tone reproduction of the graphic original.

Fig. 9 is a diagram representing graphically the nature of the signals produced by the arrangements illustrated in Figs. 6, '7 and 8.

Fig. 10 is a schematic circuit diagram showing a preferred form of control of the starting operations at the transmitter.

Fig. 11 is a simplified plan view of the reproducing apparatus forming part of the invention.

Fig. 12 is a side view of the apparatus illustrated in Fig. 11.

Fig. 13 is a schematic illustration of the recording arrangement forming part of the invention.

Fig. 14 is a schematic diagram of the signal receiving and control circuits of the recording apparatus.

Referring to the drawings and, more particularly, to Figs. 1, 2, and 3 wherein the scanning apparatus of the sender is shown in somewhat diagrammatic form for facility in illustration and description, the material of which a facsimile is desired may comprise a record represented by the sheet of paper 20, and termed the record sheet, which is assumed to be placed record face downwards on the table 2| of the scanning apparatus. The record sheet is slid along the table from left to right as viewed in Figs. 1 and 2 until registered in position over the scanning slot 2|s whereupon the leading edge (ordinarily the top edge) of the sheet engages feed rollers 22, 22a. The table 2| is preferably arched transversely to provide a cylindrical surface as is clearly shown in Fig. 3. During the scanning process, the record sheet is moved continuously across the scanning slot at a suitably slow rate by paper feed mechanism including the feed rollers 22, 22a and subrollers 23 and 23a. The latter are mounted on an axle 2 carried by hinge bracket 25 pivoted on axis 26 and provided with a spring 21 or other suitable means to press the sub-rollers against the feed rollers with a predetermined pressure.

Feed rollers 22 and 22a are mounted on shaft 28 driven by motor 29 through suitable transmission means shown in Figs. 1 and 3, as for example by motor shaft 30 (Fig. 3), worm 3|, worm gear 32 (Fig. 1), horizontal shaft 33, worm 34, worm gear 35, vertical shaft 36, and bevel gears 31, 38. Feed rollers 22, 22a are preferably metallic discs formed with roughened edges, as for example by knurling, while sub-rollers 23, 23a are preferably of yielding material such as rubber. Shaft 28 is also provided with a metallic sleeve 39 insulated therefrom, which is lightly contacted by a pair of insulated fingers 40, 40a, the combination forming paper switch XI (see Fig. 10) which forms part of the sending control arrangement.

Sleeve 39 closes the circuit between fingers 40, 40a in the absence of a sheet of paper, but as soon as a sheet passes between the sleeve and the fingers the circuit is opened.

Novel scanning means are now provided whereby a focused beam of light is synchronized with means defining an elemental area of the material scanned and whereby perfect linearity of movement of the focused beam is produced over the scanned material as applied to a curved surface.

The scanning mechanism is preferably located underneath table 2|. A scanning belt 4| is suitably positioned underneath table 2|, and moves closely underneath scanning slot 2| 3 in suitable guides 2|g mounted beneath table 2|. The scanning belt is positively driven by sprocket wheel 42, preferably mounted on horizontal shaft 33, and is held taut in position by idler wheel 42a, and mirror drive-wheel 43. A light source 44, preferably in the form of an incandescent lamp with concentrated filament, or other form of lamp having an intense, small-area light source, is mounted in a suitable focusing sleeve 45, which also carries a lens system 46 adapted to concentrate a converging beam of light 41 on rotary mirror 48. The axis of light beam 41 is at an angle to the surface of rotary mirror 48, and the reflected beam 49, termed the scanning beam, is directed obliquely on the under side of the scanning section of the scanning belt 4|, (see Fig. 2).

Referring to Fig. 4, scanning belt 4| is provided with drive notches 4|n, and with equally spaced scanning slits 4|s whose spacing longitudinally of the belt is substantially the width of the widest record sheet for which the machine is designed. The scanning slot 2|s in table 2| is somewhat shorter in length than the spacing of the scanning slits, thus providing a periodically recurring underlap period in the scanning cycle. The scanning slits have a dimension longitudinally of belt 4| substantially equal to the elements into which the record is to be resolved, but are preferably elongated transversely of the belt to enable light to be projected on the record and to be diffusely reflected therefrom to a light sensitive device 50 (Fig. 2). The lens system 46 is such that the light passing therethrough is converged on the mirror surface and after reflection therefrom is focused substantially at the surface of the element being scanned. The scanning belt being located between the rotary mirror and the point of focus, and between the point of focus and the light sensitive device 50, the scanning slit effectively controls the passage of the focused light to the light sensitive device. Stray light from the light source is shut off from the light sensitive device by the cooperation of the scanning belt and a baffle 5| which is substantially in contact along its upper edge with the scanning belt.

Rotary mirror 48, preferably of annular form as regards the mirror surface, is provided with a special contour having a cusp 480 (Fig. 5a) at one point, and has a slight, variable radial slope of the mirror surface for compensating purposes, as will be more fully described hereinafter. The optical arrangements are such that as the mirror rotates through one revolution, a reflected beam 49 projects a moving spot of light along the scanning belt. The design of the mirror is such, and its motion is so coordinated with the motion of the scanning belt, as for example by means of drive wheel 43, shaft 52, bevel gears 53, 54 and vertical shaft 55, that the projected spot of light, termed the scanning spot, falls on a scanning slit 4|s Just before it begins its movement longitudinaliy of the scanning slot lie and follows the slit Us in its movement to a point just beyond the distant end of the scanning slot, whereupon another beam starts at the point of origin to pick up the succeeding scanning slit. The arrangement provides for intense illumination of the scanning point with a minimum of stray and wasted light.

The light sensitive device 50 such as a photocell or the like may be adjustably mounted for adjustment to the optimum point and is preferably located at the center of curvature of surface 2| and slightly out of the path of the rays directly reflected from the scanning point so that cell 50 receives the diffused light 56 from the" point on the record being scanned instead of directly reflected light. In other words the light is conducted from the scanning point at an angle which differs from the angle of incidence thereupon. Suitable shielding is provided, such as hood 5! and baffle 5| to prevent extraneous light affecting the response of the photocell. The light and optical arrangements are such that as a scanning slit Ms moves along, a bright spot of light of the size of the basic elemental area (as defined by the scanning slit and scanning slot) traverses the record sheet, and thereupon light in varying amounts, as determined by the light diffusing qualities of the record, is conveyed to the photocell. Thus the light reaching the photocell is substantially restricted to that diffusely conveyed from the elemental area illuminated at the scanning point, while the scanning slits, in conjunction with the scanning slot, accurately define the elemental area being scanned at any instant. This permits a simple, efficient and inexpensive optical system to be used. and avoids the necessity of delicate and uncertain adjustments.

The record sheet is meanwhile moved continuously by the paper feed mechanism, the speed ratio between the scanning belt and paper feed preferably being such that the paper progresses one elemental line for each traverse of the paper by the scanning slit. The record is thus systematically analyzed line by line and element by element, as will be readily understood, the light intensity supplied to each elemental area being precisely uniform while the light variations received by the photocell are substantially proportional to the light and shade values of the record. These light variations are converted into electrical signals, generally known as picture signals, by novel transmitting means described hereinafter.

The details of the mirror of the novel scanning means will be more easily comprehended by reference to the diagrammatic illustrations in Figs. 5a, 5b, 5c.

Referring now to Fig. 5a wherein the contour of the rotary mirror 48 is shown in linear development for illustrative purposes, it is assumed that the annulus of the mirror structure is parted at cusp 48c and straightened to form a bar having a curved surface, the median plane being in the plane of the paper so that the figure represents a mid-section of the developed annulus. Let it be assumed that the direction of the axis of the incident beam of light (projected into the plane of the paper) is indicated by line 41; then in the initial position of the mirror, that is at the beginning of a traverse of slot 21s by a slit 4ls, the reflected scanning beam 49 is deflected at an angle to the left of the line of incidence, following well known optical laws, and the scanning spot will fall at point a on scanning are so.

Now if the mirror be considered as translated to cated by dotted line 49a) the reflected beam is deflected to the right by the same angular amount as its original deflection to the left. At the end of traverse, therefore, the scanning spot will fall at point D. As the cusp 48c passes the line of direction of beam 41 (remembering that in actuality the mirror surface presented to the incident light is continuous because of rotation of the mirror) a new beam 49 will begin its traverse of slot 2ls. Byanalysis of the motions involved and the application of well known optical laws, it is found that when the mirror contour 0 takes the form of an arc of a circle of proper radius, the scanning spot will move along the scanning arc so at a uniform rate when the mirror 48 is rotated at a uniform rate. Since mirror 48 is driven by scanning belt 4| through drive wheel 43, it is apparent that the arrangement described provides the desired coordination in motion between the scanning spot and the effective scanning slit, provided the scanning spot moves exactly in the scanning arc described by the scanning slit.

It can be shown by descriptive geometry that with the simple mirror contour, as shown in Fig. 5a, the scanning spot will not follow exactly a right circular element of a cylindrical surface, (which in geometric terms describes the path of the scanning slits across the scanning arc) but will describe a path curved with respect thereto. This curvature is due in part to the obliquity of the plane of movement of scanning beam 47, and in part to the fact that the reflecting point of the mirror varies in level during the rotation. Referring to Fig. 5b, the rectangular figure represents a surface upon which the scanning spot is projected as, for example, a portion of the under surface of the record sheet when in position for scanning. The dotted curve pa represents the curved path traced by the scanning spot when no corrective measures are taken, while full line p represents a line element of the record sheet which it is desired to scan (also the path of a scanning slit across the record sheet). The curvature error 6 may be corrected in various ways, the preferred method with the arrangement shown being to provide the rotary mirror with a constantly varying radial slope.

This is illustrated in Fig. 5c in which angle 2' represents the radial correction angle at the center of the mirror contour. Fig. 50 represents a cross-section of the rotary mirror at the low point of the contour, this being the point requiring the greatest correction under the assumptions of the example illustrated and therefore requiring the maximum radial inclination. The diagram illustrates the manner in which the radial inclination of the mirror accomplishes the required correction of the position of the scanning spot. The correction is, of course, produced in a direction at right angles to the line of motion of the scanning slits. The proper angle of inclination to correct the curvature error at any point can be determined by graphical methods or by trigonometric methods. The amount of correction required depends, of course, on the geometrical and dimensional relations of the particular case, but the inclinaamounting to only a few degrees.

It wmue pp rfit w'tndse skilled in the art that by'following'the teaching of the invention, the scanning-element's shown may be combined in various ways, and that variations may be made in these components. For example, for the purpose of scanning transparencies such as photographic film, the rotary mirror would preferably be adapted to project the scanning spot perpendicularly to the surface of the record, and the photocell (of elongated type) would be placed above the table, while the scanning belt could be placed either below or above the table. While the formof the apparatus described represents a preferred embodiment of the invention relating to methods and apparatus for scanning nontransparent copy, it is to be understood that the invention is not restricted to the particular forms of apparatus or material described.

Referring now to Figs. 2, 3, and 10, operating control of the scanning apparatus and of the transmission of signals is provided by a control arrangement, comprising paper switch XI, in series with control relay 58, battery B4, switch SI, and relay 59, can switch X2 in series with switch X3, controlled by relay 58, and controlling a shunt circuit around relay 59 and switch Xi, the controlled elements comprising light source 41 supplied with energy by the secondary of transformer 60, and the signaling circuit of a transmitter T, such as for example that illustrated in Fig. 6. Relay 58 is also provided with contacts X adapted to control the signaling circuit while relay 59 is provided with contacts X4 controlling the light source energy through the series circuit comprising the secondary of transformer 60, light source 41 and switch S2.

The sending operation is started as follows: Let it be assumed that power is connected to motor 29, hence that the paper-feed and scanning mechanisms are driven at proper speed. Switches SI and S2 are then closed, the former energizing relays 58 and 59 through battery B4, the latter closing the circuit through light source 4? if switch X4 is closed. The contacts of relays 58 and 59 are shown in Fig. in the positions they assume when 58 and 59 are deenergized hence when these relays are energized, contacts X break the signaling circuit, contacts X4 disconnect power from light source 41, and contacts X3 place cam switch X2 in circuit with battery B4 and the winding of relay 58.

To start scanning, the record sheet is slid along table it until it is engaged by feed rollers 22, 22a which thereafter advance the record sheet gradually. As soon as the leading edge of the sheet passes under roller 39, the contacts of paper switch Xi are opened and current is cut off relay 59 which is thus deenergized. enabling contacts X4 to close thereby applying current to light source 41. Relay 58 remains energized by means of the circuit including B4, switch SI cam switch X2 and switch X3. Cam iii of cam switch X2 which is driven by drive wheel 43 continues to rotate, however, and opens contacts X2 before a scanning slit 4 is begins the scanning cycle, thereby deenergizing relay 58. Contacts X are thus enabled to close at the beginning of the underlap period, and the transmission is initiated, for example, during the underlap portion of the cycle, since for the sending of a synchronizing signal the scanning beam may be entirely cut off from the record by the belt 4! just prior to the beginning of the traverse of the light beam and no light reaches the photocell 50. Other means of introducing a synchronizing pulse may, of course,

also be utilized. Since contacts X3 were opened when relay 58 was deenergized, subsequent operation of cam switch X2 is ineffective and the control circuits remain unresponsive until control is subsequently regained by paper switch XI. Meanwhile, periodic synchronizing signals and picture signals are permitted to pass to the line through the closed contacts X.

The sending machine continues scanning until the record sheet has been fed beyond paper switch Xi whose contacts close after passage of the record sheet. Current is thereby applied to relays 58 and 59, reversing the starting procedure by interrupting the sending circuit and by removing current from light source 41.

The arrangement shown, therefore, provides for automatic starting of the signaling under con trol of the record sheet, for initiation of the transmission by the sending of a synchronizing signal, and for automatic shut-oil after transmission.

Referring now to Fig. 6, this shows diagrammatically one embodiment of the novel means comprising a transmitter at the sending station for producing synchronizing signals and picture signals, varying directly with every variation in tone value corresponding to the respective elemental area scanned. This transmitter may comprise a photocell which may be of any suitable type, a photo-tube having a concentrated anode 50a and a photo-emissive cathode 500 being shown in conventional representation. The current variations in photo-tube 50, being usually of a feeble nature, are amplified by amplifier unit U which may be of any suitable type and may have one or more stages, the dotted rectangle enclosing tube Tl representing the amplifier unit U. Am plifier tube Tl may be of any suitable type, such as the usual triode vacuum type having a hot cathode 0, plate or anode a, and control grid 9. Battery Bi provides the plate potential for tube Ti while a battery B applies a normal electron accelerating potential to photo-tube 5. Tube TI is normally worked with a value of grid potential, supplied by battery Ba, suitable to limit the plate current to a safe working maximum. This grid potential is usually slightly negative, depending on the type of tube, hence battery Ba is shown as supplying a negative bias to tube TI. The anode 50a of photo-tube 5D is preferably connected to grid 9 as shown. A resistor RI is preferably connected between the cathode and grid of tube Tl, this resistor serving to limit the maximum current through photo-tube 50, under con- 'ditions of maximum illumination, and to provide a control potential for grid g substantially proportional to the amount of light striking phototube 50.

A saturable reactor 62 constitutes the controlled element, this reactor comprising a 'core which can be readily magnetically saturated by a moderate amount of direct current flowing in the saturating winding. The reactor 62 is preferably in the form of a closed magnetic core 62a having two windings 62b of an equal number of turns mounted on the outer legs of the reactor core, and a third winding 62c termed the saturating winding, mounted on the middle leg of the reactor core. The outer windings 62b, termed the A. C. coils, are connected in series aiding relation, while saturating winding 620 (or D. C. coil) is in series with the plate circuit of tube Tl. An oscillator 63, of any suitable type to produce the frequency and output desired, is provided to produce the signaling currents. This oscillator may be a vacuum tube oscillator, a tuning fork oscillator, an inductive generator, or any other type of oscillation generator suitable to produce A. C. of the desired signaling frequency, for example, 1200 cycles. The reactor is interposed between the oscillator 63 and the outgoing circuit, as for example the line. The principle of operation may be explained as follows: With no current flowing in the saturating winding, the A. C. coils and magnetic core form a reactor whose impedance is roughly determined by the number of turns in the A. C. coils and the magnetic characteristic of the core. The impedance of the reactor permits a small fraction of the full signaling current to flow, say for example. When direct current is passed through the D. C. winding, the magnetic core is saturated by the increased flux in the core to an extent roughly proportional to the ampere turns of the D. C. coil. Hence the effective impedance presented to the A. C. currents is diminished, and stronger signaling currents flow to the line. By proper design of the reactor in harmony with the oscillator and tube Tl, the amplitude of the signaling currents can,

- within a certain range, be controlled substantially proportionally to the current flowing in the plate circuit of tube TI.

The operation of the signal transmitter is as follows: Assuming that photo-tube 50 is receiving maximum illumination, a certain amount of current flows from the positive side of battery B, through battery Ba, resistor RI, and photo-tube 50 to negative side of battery B. This current causes a potential drop across resistor R! which potential is applied to control grid 9 of tube Ti in such a manner as to produce a relatively negative potential thereon, and the plate current through tube Ti is accordingly reduced to a minimum. The small plate current flowing through winding 62c produces practically no saturating effect, and hence the effective impedance of reactor 62 is a maximum, reducing the signal level to the minimum. Now when the light reaching photo-tube 50 is reduced, as when scanning a dark area of the record, the current flowing through photo-tube 50 is reduced proportionally, the potential of grid 9 is raised, and the plate current of tube Ti is increased, thus producing saturating eiiects in reactor 62, and permitting increased signaling currents to flow to the line. lhus it is seen that a decrease of light at phototube results in a proportional increase in the amplitude of the signaling currents. This circuit may therefore be utilized to produce a strong pulse of synchronizing current during the underlap period when no light is received by tube 50.

If we assume that all light is removed from photo-tube .50, the impedance thereof becomes practically infinite, no current flows therethrough, and grid g assumes its normal bias. The normal bias being set to allow the maximum permissible current flow in the plate circuit of tube TI, the saturating effect of winding 620 is a maximum, the impedance of reactor 62 is reduced to the operating minimum and the signaling currents are raised to maximum amplitude. Thus it is seen that the arrangement provided enables photo-tube 50 to produce a strong synchronizing signal and also to produce signals varying directly with every variation of tone value corresponding to the respective elemental area scanned.

By providing equal (balanced) A. C. coils on reactor 62, it will be found that transient efiects due to changes in the signal amplitudes are virtually suppressed, and no appreciable shift in the frequency of the signals occurs. By proper design of the reactor and associated circuits any tendency towards phase shift with sudden changes in signal level likewise can be reduced to negligible proportions. A further advantage is that the control power required is only a small part of the controlled power. Thus the'transmitter arrangement of the invention Just described generates and transmits signals of constant frequency whose amplitude precisely rep resents the tone values of the particular elemen- 'tal area being scanned. The nature of these signals is shown graphically in line I of Fig. 9.

Referring now to Fig. 7 which shows another embodiment of a transmitter of the invention adapted to produce signals similar in character to those produced by the arrangement of Fig. 6, photocell 50 again controls the out-put from oscillator 63 by means of amplifier unit U, and a saturable coupling device which in this case consists of a transformer of special characteristics. Amplifier tube Ti is normally worked with a negative bias, provided by battery Ba or other suitable means, of such value as to maintain the plate current at a predetermined minimum when photocell 50 is not exercising control. Photocell 50 is connected between the grid 9 of tube Ti and a point on battery Bl suitable to provide a normal accelerating potential. The controlled element in this case is a special coupling transformer 62a provided with a primary 82p (into which oscillator 63 works), a secondardy 62s connected inthe signaling circuit, and a saturating winding connected in the plate circuit of tube Ti Transformer $20; has a core of magnetic material which can be readily magnetically saturated by a moderate amount of direct current flowing in the saturating winding. This effect is facilitated by using a minimum amount of magnetic material commensurate with the alternating currents to be carried, and by the use of magnetic alloys of high permeability but in which the permeability is progressively reduced when a relatively low magnetic flux is exceeded. The desired condition is that some saturating effect be obtained due to the alternating currents in the absence of any current in the saturating winding. In this respect the coupling transformer differs from the ordinary audio coupling transformer in which the amount of core material is desirably made ample to prevent saturating effects by the combined action of the alternating and direct currents present. The primary and secondary windings may be divided into two sections and con nected in series aidingrelation in a manner similarto the A. C. coils of reactor 62 so that transient effects due to changes of current in the saturating winding are balanced out.

The principleof the control of the signaling currents in this embodiment differs from the embodiment of Fig. 6 and may be explained as follows: With no current in the saturating winding, the magnetizing eiiect of the alternating currents in the primary and secondary is just suflicient to work the magnetic core material near the point of the maximum permeability, hence the coupling efficiency is substantially at its maximum and maximum signal amplitudes are obtained in the secondary circuit. When direct current is passed through the saturating winding, the magnetic core material is worked beyond the point of maximum permeability, hence the coupling efiiciency is reduced, and the amplitude of the signals in the secondary is reduced. This effect proceeds substantially in proportion to the saturating current over a certain range, then the proportionality disappears, and at full saturation some degree of coupling still remains, hence a certain minimum signal level (say 10% of the maximum level) remains when the core material is substantially completely saturated.

The operation of the signal transmitter of Fig. 7 is as follows:

Assuming that photocell 50 is receiving maximum illumination, a certain amount of current flows from the positive connection between photocell D and battery Bl, through photocell 50, through resistor Ri, battery Ba and returns to the negative side of Bi. The potential across resistor RI due to this current opposes that of battery Ba, hence the potential of grid 9 is raised, permitting the maximum allowable current to flow in the plate circuit of tube TI. This current flowing in saturating winding 52c produces substantially complete saturation of the core of transformer 62a, and reduces the signal level to a minimum as described above. Now when the light reaching photocell 5D is reduced, the current flowing through photocell 50 and grid resistor RI, the potential of grid g, and the plate current of tube Tl are all reduced. The saturation effects in transformer 820 are likewise reduced, the coupling increased, and the signal level increased. When all light is removed from photocell 5U, practically no plate current flows, the saturating effect is a minimum, and the signal level a maximum. Thus it is seen that phototube 50 is enabled to produce a strong synchronizing signal and also to produce signals varying directly with every variation of tone value corresponding to the respective elemental area scanned.

Referring now to Fig. 8, there is illustrated another embodiment of a transmitter of the invention which is particularly adapted to full-tone rendition of graphic subject matter, as for example the rcproduction of writing, printing and line drawings. In this arrangementtwo amplifiers Ul, U2 are provided, amplifier Ul for the photocell currents, and amplifier U2 for the oscillator currents. Battery Bl provides plate potential for tube TI and an accelerating potential for photocell 50; battery Ba a normal negative bias for grid 9 of tube Tl; while batteries B2 and BI) provide plate and biasing potential respectively for tube T2. These potentials can, of course, all be derived from a voltage divider in well understood manner. Photo-tube 50 is preferably connected to tube Tl in the manner described in connection with Fig. '7. Tube Tl preferably works into plate coupling resistor R2, and is coupled to tube T2 in the special manner described hereafter. The output of oscillator 63 is coupled to the input circuit of signal amplifier tube T2 by any suitable means, as for example by transformer Tr2 as shown. A grid resistor R3 is preferably connected in the grid circuit of tube T2. The output circuit of tube Ti and the input circuit of the tube T2 are coupled together under certain conditions by a gaseous discharge device, such as for example a neon lamp 64. A device of this nature has the characteristic that no current flows until a certain critical potential, called the break-down potential, is applied to its terminals, whereupon it becomes ionized and conducting, and current continues to flow until the potential is reduced to a lower critical potential, called the extinction potential, whereupon the device is instantly deionized and becomes nonconducting.

The operation is as follows:

Assuming photocell 50 to be suddenly illuminated, current flows therethrough raising the potential of grid g and causing increased plate current to flow in tube Tl, thereby increasing the voltage drop across resistor R2. When the break-down potential of neon lamp 64 is reached, current flows in the circuit including resistor R3 and neon lamp 84, thereby suddenly reducing the potential of grid 92. This drop renders grid 92 negative to such an extent that the plate current in tube T2 is greatly reduced, and the alternating voltage supplied by transformer Tr2 produces a comparatively feeble signal output from tube T2. The signal level is thus reduced to a minimum, termed the low level. When the light reaching photocell 50 is abruptly reduced, the potential of grid g is likewise reduced, causing diminished plate current to flow in the output circuit of tube Tl, thereby reducing the potential impressed across neon lamp 64 below the extinction point to produce deionization. The potential of grid g2 is thereupon restored to normal value and high level signals are resumed. The change of light values at the photocell in passing to and from sharply defined black and white areas, as in scanning print for example, occurs with great rapidity. Because of the rapid. decisive change in the potentials on grid g2 due to the ionization and deionization of neon lamp 64, the abrupt changes in the signal level from high to low, and vice versa, faithfully represent such changes in the light values at the photocell. It should be observed that the signal control circuits comprise only elements having negligible inductance and capacity, and as a consequence, the abrupt changes in signal level are accomplished without the introduction of transient effects of any practical importance. The arrangement may be considered as analogous to a signal keying arrangement in which transients (commonly called key-clicks") are suppressed. From the foregoing, it is apparent that the picture signals comprise groups of oscillations of the same frequency but of two distinct amplitudes, being further characterized by their time of occurrence relative to the periodic synchronizing signals and by the duration of each group. The

particular nature of the signals produced by the transmitter arrangement of Fig. 8 is graphically represented in line 11 of Fig. 9.

Referring next to Figs. 11 and 12, these figures illustrate one embodiment of a novel facsimile recorder according to the invention wherein the recorder mechanism is mounted on a frame including a table 65. A roll of paper 56, on which the subject matter is to be recorded, is mounted on an axle 61 moving diagonally in two guide slots g cut in the side members of the frame. The paper feed mechanism comprises feed roller 68 for feeding the paper continuously past the recording point, and control and drive means therefor. Feed roller 68 is automatically raised and lowered (to feed the paper) by duplicate lift mechanisms, one mounted on each side of the frame, hence only one mechanism will be described. Each lift mechanism comprises a bell crank 69, pivoted on drive shaft 10 and having roller 68 mounted at the end of one arm, while the other arm is pivotally connected to a toggle pair comprising links II-Ha and pivoted lever 12. The link mechanism is actuated by solenoid 1! whose plunger is connected to the hinge Joint of the toggle by clevis 18. A retractile spring 18 tends to bring the toggle to the straight position against stop 12a as shown, in which position roller I8 is lifted clear of the table. Feed roller 88- is driven from drive motor 18 by a transmission system including drive shaft 11, worm 18. worm gear 18, vertical shaft 88, worm 8|, worm gear 82. shaft 18, spur gear 88, idler gear 84 and final gear 85, the latter being fixed on the axle 88a of feed roller 88.

The reproducing arrangement for producing the facsimile has a general function the reverse of the scanning arrangement at the sender, and includes a recording belt 86 mounted above the table and provided with stylus points 88;) (see Fig. 13) preferably spaced longitudinally of the belt at equal distances. Belt 88 is positively driven by sprocket wheel 81 and is held taut in position by idler wheel 81a. The edge of the recording belt moves transversely across the paper along the median line of a narrow guide slot 85s,

Fig. 11, extending partly across table 85, this slot corresponding to the scanning slot 2ls at the sender. A light but rigid pressure bar 88 of inverted T-section and somewhat shorter than the width of the paper 66 (see Fig. 13) moves freely in guide slot 65s (see Fig. 12) and is mounted on and driven by the armature 89 of one or more magnetic motor units 88 (two being shown see Fig. 13). These armature pieces 88, extended at right angles by thrust rods 89a, reciprocate pressure bar 88 in response to printing signals, which bar in turn presses the paper against the recording belt as required in the reproduction process. Since the printing signals are of a relatively high frequency, the effect of the motion of the pressure bar is to forcibly vibrate the paper strip, called the record strip, along a transverse linear element, the paper striking and rebounding from the stylus effective at the moment. The actual recordingis preferably done by means of an inked ribbon at stretched taut across and above the paper along the path of traverse of the stylus points 86p although other recording media, such as carbon paper or the like, may be used. The paper tit and ribbon M are thus squeezed between the pressure bar and the moving stylus point, the recording being in the form of a single dot, or groups or chains of dots. The recording belt thus acts as the anvil or platen while the pressure bar acts as the hammer. .The ends of the stylus points may be of various forms, as for example, rounded, pyramidal, or diamond-shaped.

to produce dots of any desired shape. The rib bon 9! is fed slowly along its length from reel 82 to reel 82d and vice verse. by reversible ribbon feed mechanism (not shown) which feed mechanism may be similar to any one of numerous well known mechanisms and therefore is not described users; v

entira surface of the ribbonin much the same manner that the stylus points cover the surface in detail. The preferred arrangement of the inked ribbon 9i with reference to the recording line of the stylus points 86p is for the ribbon to be positioned so that the stylus points move diagonally from margin to margin of the ribbon as disclosed in applicant's copending application Serial No. 72,349, filed April 2, 1936, now Patent No. 2,176,680 granted Oct. 17, 1939. The preferred arrangement is shown in Fig. 11 and referring thereto it will be noted that the stylus points move so as not to catch the edges of the ribbon. The combined motions of the stylus points and ribbon, which latter is moved at such a rate that it is advanced substantially the extent of one line for each traverse of a stylus point, cover the of the paper. In this way, a fresh inked surface of the ribbon is presented to each succeeding stylus point, and all the surface and ink of the ribbon are effectively used in the reproduction proc- 888.

Referring again to Fig. 12, the reproducer is also provided with mechanism for driving and synchronizing the recording belt. Sprocket wheel 81 which drives the recording belt is driven from motor 18 by a transmission system comprising drive shaft 11, bevel gears 88, 84, hollow shaft 88, then through a phase correcting gear 88, stubshaft 81, and vertical hollow shaft or sleeve 88 to which sprocket wheel 81 is aflixed. A clutch or friction device 88 (details not shown) of any suitable design is interposed between stub shaft 81 and vertical shaft 88, this device permitting release magnet M2 to stop shaft 88 and sprocket wheel 81 in a predetermined angular position while stub-shaft 81 is rotated, in which position any one of the stylus points 88p is at the beginning of a traverse. A rotary cam switch combination I08 (described in detail later) is also driven by shaft 88, this switch combination comprising one of the elements for establishing and maintaining synchronism between the recorder and the sender.

Referring now to Fig. 14, at the receiving station, the windings w of the magnetic motor units 98 (one being shown) may be connected directly to the line, or in case the signals are carried at a low level, or are attcntuated in transmission, a vacuum tube amplifier (not shown) may be inserted between the line and the recording circuits. A coupling transformer Tr3, preferably of high impedance, has its primary bridged across the line in shunt with unit 80, while its secondary is included in the input circuit of a relay, preferably of electronic type as shown. While electronic relay T3 may be one of various types, the preferred type is a gaseous triode comprising a hot cathode, a control grid, and an anode or plate; all mounted in a sealed envelope containing at atmosphere of gas or vapor. Triodes of this type provide the advantages of ease of control, quick response, large current carrying capacity, and a high ratio of control power to controlled power. The. control grid is normally maintained below the ionizing point by a negative bias provided by battery Bc, while the plate potential is provided by battery B3 or other suitable source. The output circuit of T3 includes battery B3 and a plurality of branch circuits connected in sequence by the rotary cam switch M0. The cams iOl, M2 and 33 of this switch are preferably mounted on the same hollow shaft 98 as sprocket wheel 81, as indicated.

Corrector magnets MI and M3 are connected in the branch circuits including contacts 181a and IBM, respectively. Release magnet M2 is connected in the branch circuit including contacts Mia. Release magnet M2 is of the quick acting, slow release typefcapable of being sustained in operation by periodic impulses, energy being stored in capacitor C2 to maintain the magnetic flux between impulses. Release magnet M2 is provided with a pawl member I84 (see Fig. 12) adapted to arrest rotation of sleeve 98 and also to operate contacts 14a which connect solenoid 13 into circuit with battery B3. The functions of the receiving and control circuits will be more fully disclosed in the description of the operation of the recorder.

Referring now to Fig. 12 which shows in somewhat schematic manner the transmission andthe system of gears previously described drives feed roller 68. Drive shaft 11 also, by means of bevel gears 83, 84, drives sprocket wheel 81. A phase corrector mechanism 26 is interposed in the transmission system between drive shaft I1 and shaft 01. This mechanism may be of any desired type capable of either advancing or retarding the phase position of the controlled member. A mechanism suitable for use in connection with the present invention is shown in my U. 8. Patent No. 2,111,153 granted March 15, 1938. As shown in the instant application, the mechanism comprisesa pair of corrector magnets MI and M2, one adapted upon energization to cause an increase in the speed of the controlled member (shaft 81) with respect to its drive means (shaft 11) and the other adapted to decrease the speed thereof.

Shaft 8! controlled by the phase corrector mechanism and controlled in its rotation thereby with respect to drive shaft 11 drives, by means of a friction device 89, sleeve 98 on which is fixed sprocket wheel 81. Pawl member I04, actuated by magnet M2, is adapted to engage a stop piece 98s on sleeve 98, thus bringing sleeve 98 to a stop in a predetermined angular position. In this position, rotary cam switch IMI, whose cam members are fixed on sleeve 98, stops in such position that contacts i02a are closed, (see Fig. 14). Drive sprocket 81 is also fixed on sleeve 98, hence recording belt 86 can be coordinated with the cam members in such relation that one of the stylus points is in the underlap position during the motional interval when contacts l02a are closed. The arrangement is such that the recorder is stopped in and released from a predetermined initial position whereby approximate synchronism is established in response to the first received synchronizing signal. More exact synchronism is effected and thereafter maintained by the phase corrector arrangement as described in connection with synchronization.

Referring now to Figs. 12 and 14, the operation of the receiving apparatus in utilizing the synchronizing signals is as follows: In stand-by condition, release magnet M2 is not energized, permitting pawl member iii! to intercept stop member 88s thus stopping sleeve 38 in a predetermined angular position. In this position, cam

I02 closes its contacts i02a as indicated in Fig. 14.

When the transmitter sends the first synchronizing signal, transformer Tr3 applies a positive swing to grid g3 of electronic relay T3, overcoming the bias and tripping T3 which thereupon becomes conducting. Current flows from the positive side of battery B3 through the winding of M2 via contacts l02a and tube T3 back to battery, thus energizing M2. Magnet M2 is quick acting, hence actuates pawl I04, quickly releasing sleeve 98, thus connecting driving power to sprocket wheel 81 and rotating the cams of switch I00. When contacts i020 open by reason of the rotation of cam I02, relay T3 is deionized, grid g3 meanwhile having been restored to normal bias by battery B0. The energy stored in capacitor C2 is sufficient to hold magnet M2 operated for more than one revolution of sleeve 98. Because of the slight delay incident to the operation of magnet M2, the cam shaft may lag slightly and thus when the second synchronizing signal is received, (at the end of the traverse at the sending end) contacts Illa may-be closed. TI is tripped as before but this time part of the discharge current flows through corrector magnet Ml via contacts Illa. Magnet Ml actuaics phase corrector II to advance the phase, regaining part or all of the initial lag. Phase correction is repeated on succeeding synchronizing signals, if necessary, until accurate unison is established, whereupon all the local pulses are applied to magnet M2, sustaining same in operated condition.

Thereafter, no change occurs unless the receiving apparatus tends to lag or lead relative to the synchronizing signals, whereupon magnet MI is energized in case of lag, and magnet M2 in case of lead, these magnets controlling the phase corrector to restore unison. Recording belt 88, sprocket wheel ll, cam sleeve 98 and rotary cam switch Hill are kinetically coordinated in such manner that when contacts i020 close coincidentally with the period of the local synchronizing pulses, recording belt 86 moves in unison with scanning belt 4| at the transmitter. Without further analysis, it will be clear that the periodic synchronizing signals are first utilized to establish synchronism, and are thereafter utilized 'to maintain synchronism.

From the description of the various parts and mechanisms of the recorder, it will be seen that the complete operation of therecorder is as follows: In standby condition with the motor running, sleeve 98 is stopped by release magnet M2 in such position that contacts [02a connect release magnet M2 in circuit, and one of the stylus points 86:) is positioned at the beginning of the underlap portion of the scanning cycle. Spring 15, operating through lever 12, holds feed roller 08 clear of the record strip, hence the paper feed is inoperative. When the first synchronizing signal is received, electronic relay T3 (see Fig. 14) is tripped, applying current to the circuit of release magnet M2 which is energized thereby and pawl member ill releases sleeve 58 whichis quickly set in rotation by friction device 99, thus setting recording belt 86 in motion. Pawl member I04 also closes contacts iiila connecting solenoid 13 in circuit and energizing the same. Solenoid l3 pulls down toggle pair H, "I la lowering feed roller 68 into contact with the record strip thereby starting the paper ieed. When cam I02 is rotated through a small angle, contacts i02a open, cutting off current to tube T3 and permitting it to deionize. Meanwhile capacitor C2 has been charged sufficiently to hold release magnet M2 energized for a period of two or more revolutions of cam sleeve 98. Synchronism of the recording belt with the scanning belt is established and maintained as previously described; whereupon the local synchronizing pulses are applied to release magnet M2, sustaining same in operated condition so long as the synchronizing signals are regularly received.

Assuming now that the recording belt has been synchronized with the scanning belt, as the picture signals are received from the line, the current waves pass through the winding of magnetic units whose armatures 89 are set in vibration, alternately thrusting the pressure bar 88 towards and pulling it away from, the paper 66. In normal position, pressure bar 88 is adjusted (by means not shown) to have its upper edge substantially in contact with the paper. Thus when picture signals representing dark areas are received, the pressure bar vibrates the paper, pressing the paper and inked ribbon against the recording belt at the point where a traveling stylus point 86p projects. The length of the dot or dash resulting from such engagement depends upon the duration of the dark signals, while the depth of recording; namely, the dot size and amount of ink deposited, depends upon the amplitude of the dark signals. The stylus points move along in synchronism with the movement of the scanning slits at the transmitter, and therefore reproduce in extent, position and shading, the light and dark areas of the original material being scanned. For half-tone recording, the operating conditions are established so that faint or shallow dot recording occurs even at the lightest areas of the subject, the depth of recording, increasing as darker areas are scanned. For fulltone recording, operating conditions are adjusted so that the low level (white signals) do not record, while the high level (black signals) record at uniform intensity, producing white or black areas as the case may be.

The adjustment of the various factors to obtain the desired half-tone or full-tone effects depends in part on the subject matter being transmitted,

and in part on the characteristics of the transmission lines, signal level, and various operating factors. The principles to be followed in providing desiredoperating conditions are well known in the art and consequently do not require detailed explanation.

It should be noted that it is not necessary to switch the recorder units out of circuit when the synchronizing signals are being received since the synchronizing signals are received during the underlap interval at which time the recording points are not within the range of the pressure bar. Likewise, when picture signals are being received, the electronic relay T3 is ineffective due to the fact that the plate circuit is incomplete at the cam switch I00.

At the end of the transmission, when the synchronizing signals cease, magnet M2 is deenergized and contacts IMa are permitted to open, breaking the circuit to solenoid 13 which is thus deenergized, permitting spring 15 to operate lever T2 and links H to oscillate bell crank 69 to thereby raise feed roller 58, stopping the paper feed. The strip of paper upon which the facsimile was produced can then be pulled out and torn off at tear-strip I05 or other desired means used to cut off the record sheet.

In the case of half-tone recording, it is sometimes desirable to have the elementary dot areas arranged in checkerboard fashion, that is, each line of dots is displaced a predetermined amount (usually one-half the breadth of the elemental area) with relation to the adjacent lines, thus providing the typical appearance of pictures produced by processes involving the use of halftone screens. Such effects can readily be obtained with the system of the invention. For example, the spacing between successive scanning slits may differ by the proper amount, producing signals displaced by an increment of time when successive lines are scanned, which signals at the recorder are translated into an increment of position. Another method is to have the spacing between successive stylus points differ by the amount of the desired displacement in position of the dots in successive lines.

A third method is to provide novel means comprising in combination with the transmitter, a rotary reversing switch adapted to in effect reverse the connections to the magnetic recorder units between the recording of successive lines. By this means the alternations effective for recording are reversed for each line of dets, thus producing the desired staggered position of the dots. The novel means including a reversing switch capable of producing the desired reversing action is illustrated in schematic form in Fig. 7. Referring thereto, conductors W and WI of the signaling circuit are connected to half-rings R and RI respectively, while the line terminals are connected to rings R2 and R3. Rotary brush arms A and B, insulated from each other and each carrying a pair of differently spaced brushes, respectively contact each of the half-rings and ring R3, and each of the half-rings and ring R2. The brush arms are driven at half scanning speed, that is to say, they complete one revolution during the scanning of two successive lines, and are preferably positively driven by shaft 33 upon which drive wheel 42 is mounted (see Fig. 1) and are coordinated therewith in such manner that the reversal of connections occurs during the underlap interval. Thus the signal alternations are, in effect, displaced during the scanning of successive lines, causing the dots at the recorder to be displaced by a half space to provide the desired staggered effect. It will be evident that the same results can be obtained by placing the reversing switch at the recording end of the system, but it is generally preferable, particularly when the system is to be used for both half-tone and full-tone recording, to have the control of this feature at the transmitting end. Then by means of a throw-out clutch (not shown) to stop the rotary reversing switch, the system can be changed over at will from staggered to nonstaggered pattern in accordance with the subject to be transmitted.

Novel means are therefore provided whereby precise facsimile reproductions can be made in both full tone and half tone, and wherein the material to be copied is linearly scanned by a concentrated beam of uniform light intensity while signal variations are produced precisely proportional to the variation of tone values corresponding to the respective elemental areas scanned.

Although several embodiments of the invention have been illustrated and described, various changes and modifications in form, materials, dimensions and relative arrangement of parts and circuits, which will now appear to those skilled in the art, may be made without departing from the scope of the invention.

Reference is therefore to be had to the appended claims for a definition of the limits of the invention.

What is claimed is:

1. In a facsimile system, means for producing a concentrated scanning beam of light, and rotating optical means intercepting said light beam and rotating said scanning beam along a path in one plane and means comprising a part of said rotating optical means simultaneously controlling the movement of said beam along a path in a plane at an angle to said first plane.

2. In a facsimile system, means for producing a concentrated scanning beam of light, and rotating optical means having reflecting contours continuously deflecting said'scanning beam along a path in one plane and simultaneously deflecting said beam along a path in a plane at right angles to said first plane.

3. In a facsimile system, means for producing a concentrated beam of light, and a rotating optical member located in the path of said beam comprising a surface gradually rising and falling with respect to the plane of rotation of said member and gradually tilted out of said plane.

4. In a facsimile system, means for producing a concentrated beam of light, and a rotating optical member located in the path of said beam, said member being so constructed that its linear development comprises a surface, concave in one plane, said concave surface sloping away from a plane at right angles to said first plane at a varying angle and gradually along one extent of the linear development thereof and reversely sloping along another extent thereof.

5. In a facsimile system, means for producing a scanning concentric beam of light, and a rotating optical member located in the path of said scanning beam, said member being so constructed and arranged that said scanning beam is defiected along a path generically in one plane and simultaneously tilted in a plane generically at right angles to said first plane.

6. A facsimile system of the character described for reproducing a copy of a graphic sub- Ject comprising means for evaluating the tonal values of respective light and dark areas of said subject and producing current variations proportional to said tonal values, an electronic discharge device, and control means for said discharge device comprising a gaseous discharge device, controlled by said current variations whereby electronic flows through said electronic discharge device of two amplitudes only are produced, said amplitudes varying in duration in proportion to the number of successive light or successive dark areas evaluated.

'7. A facsimile system of the character described for producing a copy of a desired object comprising means evaluating the tone values of the sequential elemental areas of said object including a slot, means moving said object with respect to said slot to define progressive line elements, means forming the boundaries of a moving slit crossing said slot and defining said sequential elemental areas by the intersection of said slot and slit, and means producing a concentrated moving beam of light continuously illuminating said intersection within a desired field of view.

8. A facsimile system of the character described for producing a copy of a record comprising a curved surface for mounting said record, a slot in said surface, means moving said record across said slot to define progressive line elements, means forming the edges of a moving slit crossing said slot and defining sequential elemental areas of said record by the intersection of said slot and slit, means producing a concentrated beam of light, rotating optical means in the path of said beam driven in unison with said record, the surfaces of said optical means being so designed as to move said beam of light continuously and in coincidence with the movement of said intersection within a desired field of view.

9. In a facsimile system of the character described comprising means for sequentially scanning elemental areas arranged in line elements of a field of view, means controlled by said scanning means producing electrical quantities varying directly with the light and dark tone values of respective areas, and means synchronized with said scanning means, and operative upon the scanning of alternate line elements only, regulating said producing means to thereby produce electrical quantities varying inversely with the light and dark tonal values during scanning of said alternate line elements.

10. In a facsimile system of the character described comprising means for sequentially scanning elemental areas arranged in line elements of a field of view, means controlled by said scanning means producing an electrical output at a pair of terminals varying with the light and dark tonal values of said respective areas, and means synchronized with said scanning means for connecting said terminals to a transmission medium and reversing the connection of said terminals to said medium each time a new line element is scanned.

11. In a facsimile system of the character described comprising means for sequentially scanning elemental areas arranged in line elements of a field of view, means controlled by said scanning means producing an electrical quantity varying with the tone values of said respective areas and including an electron discharge device, switching means for periodically reversing the effect of the output of said electron discharge device, and means synchronized with said scanning means for automatically controlling said switching means to produce a reversal of eflect upon scanning of alternate line elements.

HARRY'J. NICHOLS.

Images