US1790723A - Facsimile system - Google Patents

Description

3, 1931. HHRANGER 1,790,723

FACS IMILE SYSTEM Filed Feb. 7. 1929 3 Sheets-Sheet 1 nomto'z RICHQRD ROWL ND RANGER 1 T3513 his flb f wj Feb. 3, 1931. RANGER 51,790,723

FACSIMILE SYSTEM I Filed Feb. 7. 1929 s Sheets-Sheet 2 anvemtoz mum uownuo RANGER Feb. 3, 1931.

R. H. RANGER 1,790,723

FACSIMILE SYSTEM Filed Feb. '7, 1929 3 Sheets-Sheet 5 A IllllIIH|IIllllllllllllIllllllllllllllIlllllllllllllllkllllllllllllllllllllllllllllllllllllllllllllllJ|ll|Illllllllllllllllllllllll B i I I l F I v awuenboz RKHARD mum) RANGER Y Patented Feb. 3, 1931 UN TED STATES PATENT-l OFFICE BIGHABD.EOWLAND RANGER, OF NEWARK, NEW JERSE Y, ASSIGNOR TO RADIO COR- PORATION OF AMEIRICA A CORPORATION DELAWARE FACSIMILE SYSTEM Original application filed July 16, 1924, Serial No. 726,235. Divided and this application filed February I 7, 1929. Serial No. 338,072. I

The present'invention forms a divisional part of my-copending application, Serial No. 726,235, filed July 16, 1924, and relates to a new and improved-system ,of duplicating or reproducing pictures and .the like and, al-

though my invention may be used for the purpose of producing duplicates of pictures or the like at the place where the originals arefound, it is also particularly adapted to produce such pictures at a considerable distance from the originals utilizing for the purpose of control of the reproduction in existing telephone, telegraph, radio, or wired radio circuits. 1

The invention herein disclosed. and claimed is an improvement on the invention disclosed and claimed in copending application of E. F. W. Alexanderson andv R. H. Ranger, Serial No. 669,688, filed October 20, 1923, for picture transmission by wire or radio, and copending application of R. 11. Ranger, .Serial No. 695,17 5, filed February 26, 1924, for methods and means for repro ducing and transmitting pictures, and over the disclosure in copending application of R. H. Ranger, Serial No. 338,071, filed February 7, 1929, and is a continuation in part of the latter application.

The novel features which I believe to be characteristic of my invention areset forth with particularity in the appended claims; my ihvention, however, both as to its organization and method of operation, will best be understood by reference to the following description taken. in conjunction with the ac- 'companying drawing, in which Fig. 1 shows one form of apparatus. embodying my invention; I v

Fig. 2 shows another and some respects more preferable form thereof;

Fig. 3 is an illustration showing the method by which I may reproduce varying values of light intensity according to thisYinven- Fig; 4 is a set of curves explainingthe operation of my system; I

Fig. 5 is an example of a picture reprodu ced according to my-invention; and Fig. 6 diagrammatically illustrates a series of consecutive lines of apicture built up in accordance with the representations shown by Fig. 3, and particularly in accordancewith the representations shown by portion E of Fig. 3. v

My invention contemplates as a new article of manufacture a picture or pictorial representation in which the efiects of vary ing intensities of light are produced in a novel 'mannerand'contemplates also a proccess by which such pictures may be produced and apparatus by which the process may be carried out. The pictures which I produce and the process by which I produce them render my invention particularh useful for transmitting pictures to a distant point by means of telegraph, telephone-- or" radio circuits, particularly the latter.

While I will herein refer to the use of my invention and the article of manufacture.

produced as a picture; and the like itshould be understood that the invention is not lim- 1 ited thereto but that the article of manufacture produced may be any sort of a duplicate, for instance, copies of' motion picture films,'manuscripts, letters, documents, finger prints, fashion plates, magazine covers, financial statements, prints and labels and, in fact, all similar subject matter. I, therefore, .wish it to be understood that I have used the term picture throughout the specification and claims in a broad generic sense as defined above to include the various.

objects given by way of example and others, as will be understood by those skilled in the art to which the invention relates.

It is an object of this invention to produce a picture or visual record in which the effects of varying light intensity are produced .in a novel manner.

It is another object ofthis invention to produce a practical and simple method or vide a method and apparatus adapted for 'the reproduction and transmission of pictures, by the use of which a more faithful reproduction of the original is obtained.

It is another object of my invention to provide a method and apparatus for the reproduction of pictures which is adapted for use in transmitting pictures to a distant point by the use of telephone or telegraph circuits.

It is still another object of my invention to produce a. method and apparatus for the reproduction of pictures whi'chis particu iarly adapted for use in transmltting p1ctures to a distant point over existing radio circuits at a. high of speed.

Still a further object of my invention is to produce reproduction of an originally transmitted picture which reproduction closely corresponds to the original and, at the same time, closely approximates the hi hly desirable half tone prints.

iltill another object of my invention is to provide, as an article of manufacture, a reproduced. transmitted picture in which the varying light or tone intensities are produced thereon in a noi'el manner and in which certain tone intensitics,lying on one side of a. predetermined light or tone value, are reproduced by one type of marking impulse and in: which light or tone values lying on the other side of' the predetermined value are reproduced in a different type of marking impulses,

Still other and ancillary objects of-my invention will become apparent from a reading of the following specification when considered together with the accompanying drawings, which are given byway of explanation.

As above stated, the present invention forms a divisional part of my copending application, Serial 1%).726235, filed July 16,

1924:, and this last namedapplication, in turn, forms an ii'nprovementover my copendlng applicatiom -Serial No. 695,175, filed February26, 1926, for method and means for reprbducing and transmitting pictures and over copending application, Serial No. 338,071, filedFebruary], 1929, disclosing as a divisional part, a system particularly relatin to the picture produced by means of the arran ement shown in copending application, erial No. 695,175.

The'novel features which I believe to be characteristic of my invention are set forth with particularity in the appended claims but the invention both as to its organization and method of operation willbe best understood by making reference to the above described drawings in connection with the following description. i

In my copending appllcation, Serial No. 338,071, filed February 7, 1929, abovereferred to, I have disclosed a picture made up of a mass of dots of substantially uniform size and varying in concentration to produce the efl'ect of varying intensities of light and shade. To expedite the understandingof the present invention which embodies in part the principles disclosed in the above named copending applications,'Serial No. 695,175.

pulses separated. by spacing periods of varying duration, or, to express it difierently, successive portions of the picture are translated into successive electrical impulses of substantialiy uniform duration and of a varying number per unit of time. This arrangement presents the pictures and the like to be transmitted in accordance with the above outlined systems of transmission utilizing at all'times the full power of the transmitting apparatus, whereby distortions of tonal variations produced by static,,linenoises, etc., is minimized. I

The picture may be reproduced from such a series of impulses by utilizing them to control a mechanical or other suitable recorder, as more specifically described in the said applications. In its simplest form. such a recorder consists of a drum carrying upon it the fabric on which the picture is to be reproduced. A recording pen or stylus is controlled by any suitable means, such as an electromagnet, in such a way as to be brought into contact with the fabric during the time that an'impulse is being received. The recelver drum is rotated in synchronism with the transmittin drums and longitudinal travel between t is pen and drum is provided in proportion to the advance of the picture at the transmitter Such a system produces a picture made up of dots and bearing a certain resemblance to the well known half-tone en raving. It will be noted, however, that in t e case of a picture produced according to the process just described the dots are all of uniform size, but variably spaced; whereas, in a half-tone engraving they are of varying size and spaced a uniform distance between centers. I have discovered in carrying out the invention that in order to obtaln heavy black in the said process the dots should be sufficiently close together so as to merge substantially into a single line and that the speed at which the dots can be transmitted and received over heavy black portions of the picture determines the limit of speed at which the picture can be transmitted. Moreover, since heavy black values are transmitted as a rapid series of dots, it will be apparent that this system involves considerable operation of relays and the like with conratus that I may term the photo-modulator,

V which includes the apparatus for translating shadow.

light intensities into current impulses, but i also on the transmitting apparatus itself.

I have discovered that these objections, and others not specifically enumerated, may be overcome by providing a system wherein light values below a certain intensity'that is to say, values darker than an intermediate gray, as will be more particularly explained hereafter-are represented, not by a series of dots close together as in the preceding arrangement, but by impulses varying in length according to the intensity of .the

Briefly, high lights are transmitted as uniform short impulses or dots separated by varying spacing periods, as before, until an intermediate gray value is obtained, at which point the impulses and the spacing interval'between impulses are of the sametime duration. -For values of shadow more intense than this, the value is represented by increasing the-length or duration of the impulses, andmaintaining the spacing interval uniform. This may readily be seen by reference to Fig. 3, wherein line 'A represents successive values of light intensity of the transmitter, starting with black and gradually fading out to white, then returning to dark gray; B represents the power .or current output of the transmitter working in accordance with the invention disclosed in my prior application No. 695,175, and Serial No. 338,071; C represents the record produced from the series of impulses as shown in B; D represents the power or current impulses produced according to the invention disclosed and claimed herein, and E represents the record produced when the transmission is as shown in D.

It will be noted that the extreme left hand portion of'line A is represented in 'lineD by a single long dash which is reproduced as shown in E. The next portion of less intensity is represented in the corresponding portions are represented by dashes instead of dots there is no obstacle to increasing the speed of the transmitter and receiver as much as desired. This could not be done with the arrangement disclosed in my copending application No. 695.175 for the reason that if the apparatus were speeded up to too great an extent, the dots, represented by the left hand portion of line C, would be separated by intervals which would appear on the record as whitespaces and, therefore,-the entire picture would be in efiectlight; that isto say, all of the relatively dark portions of the picture would be given a lighter value than they should have and changed to a greater extent than the lighter portions of the picture whereby the tonal balance of the picture would be destroyed:

However, according to this invention, it is possible to speed upboth transmitter and receiver to about twice the maximum speed obtainable with the previous apparatus without destroying the tonal balance of the picture. This is clearly seen from line Hwhich represents a record produced by current impulses, as shown by line F, when the apparatus is speeded up and will also be seen from Fig. '5, which is a reproduction of a-picture made according to this process in about half I the time necessary to make the same according to my prior process. I

It will be seen by comparing line H with line G that the dark portions of the picture still retain their proper values, and the tonal balance of the picture is not afi'ected. This point is of great practical advantage'since it will be obvious that a decrease in the time necessary to transmit a picturerequires less .power from the transmitter and a smaller charge to take care of the overhead expenses of the transmitting and receiving apparatus, thereby resulting in a cheaper cost for the transmission of pictures. I

F ig. 6 more particularly illustrates in a rather conventional manner aseries of lines of a graphical representation built up in accordance with the disclosure of the present parts of D and E and it will benDted t at "invention. More particularly, Fig. 6 illusthe impulses are of shorter duration; The next portion of line A may be taken as the intermediate portion where the changeover occurs and at this point the impulses and the spacing periods are-of uniform duration and equal'to .each other. a

From this point on to the right, the-impulses are of uniform duration and variably spaced in the same way as in line B. The

' final right hand portion of a value darker and if carried on at slow sp would appear as in line G.

' than the intermediate is again represented by dashes. Line F represents the current or power impulses produced-upon a gradua- I tion of the light intensity from light to dark eed, the record Itwill be seen, however, that since the dark trates several lines of the picture built up from a series of marks of the general type shown by Fig. 3, and particularly those produced in line E of Fig 3, wherein variable length marking pulses equally spaced repre-' sent light values darker than a predetermined normally light value,and equal duration marks variably spaced represent light values lighter than a predetermined normal. value. The showing of Fig. 6 has been portrayed merely to illustrate the successive manner of producing a series of adjacent lines on the picture withoutregard to repre senting any particular image, since it has been found that no true manual representation of the picture may be made.

Referring now particularly to Fig. 1, I I

have shown apparatus adapted to translate the intensity of light into current impulses of the type described; that is, wherein values of light lying on one side of an arbitrary boundary zone are transmitted as impulses of uniform duration separated by variable spacing periods, and intensities of light lying on the other side of said zone are'trans mitted as impulses of varying duration separated by uniform periods, while the intensities of light lying directly upon or in the boundary zone are transmitted as uniform impulses separated by uniform spacing periods, both of which are substantially equal in length.

In this figure, 1 represents a light sensitive cell of any suitable type, such for in stance as that shown in my prior appli ation. It is provided with a light-prooi 6 ing or container 2 having an opening 3 the in adapted to admit light thereon,

l/Vhile l have not shown an specific ar rangement for projecting light trom the picture upon this cell, any suitable arrangement may be used, such for instance as that shown in my prior application, in which 1 have dis.- closed a drum carrying the picture and driven by asuitable motor, and a source of light arranged to project light from the picture on the cell. Various lenses may be provided and the drum is arranged for rotational. and translational movement simultaneously, the lenses and source of light being so arranged that light from all portions of the pictureis successively thrown through the opening 3 and upon the light sensitive cell 1. A suitable source of potential 1, which may be a direct potential of approximately 300 volts, is arranged to tend to force a current through this cell and resistance 5 is connected in the circuit including the source a and light sensitive cell 1 in such a manner that a potential drop is produced in this resistance by the flow of current therethrough.

The grid of a suitable vacuum tube amplifier 6 is connected to one portion of resistance 5 and the filament thereof is connected to another portion of the resistance in such a manner that when a potential drop occurs in resistance 5 duc'to a flow of current through light sensitive cell l, the grid of tube 6 will be maintained at a negative potential with respect to the filament. To the output or plate electrode of vacuum tube 6 there is connected bias battery 9 and resistances 10, 11, 12 and bias battery 13 as shown.

A second vacuum tube amplifier 7 has its plate or output electrode connected to'a point between resistance 12- and bias battery 13 and its grid or control electrode through bias battery 8 to a point between resistances 10 and 11. Any suitable source of potential for energizing the plates of this tube, such as indicated by +B, is connected between I resistances 1,1 and 12. As shown, the control or grid electrodes of vacuum tubes ll. and 15 are connected respectively to the terminals of bias batteries 9 and 13 remote from resistances 10 and 12.

Between the output or plate electrodes of tubes let and 15 and the filaments of the same I I connect condensers 16 and 17 respectively and a circuit is also established from the plate electrode of tube 14, to one terminal of bias battery 20, through relay'coil 32, thence to contact 23 and, in the position of armature 22 shown, to the common filament lead and the negative terminal of the source trodes connected to the negative terminal of bias batteries 20 and 21 respectively, the posltive terminals of which are connected to one plate of condensers 16 and 17 and also to the plates o1 tubes 14,- and 15.

The plate or output electrodes of tubes 18 and 19 are respectively connected through relay coils 25 and 26 to the common source of plate potential +B. Relay coils 32 and 33 control the position of armature 29,211:- tractmg it against either contacts 30 or 31,

thus selectively energizing line 35 by current reversals from battery 34 for the purpose of transmitting the picture to a distance.

In the arrangements shown, plus current.

may be used for signalling and the minus current for spacing. Obviously, however, this is merely by way of example and any other I suitable arrangement may be provided.

Having now described the circuit arrangement, the operation will be briefly set forth. Fundamentally, the arrangement comprises two systems, each of which is substantially similar to the system disclosed in my copending application No. 695,175, jointed together in such a way that one system controls or varies the duration of the impulse periods without afiecting the duration of the spacing periods, While the other controls or varies the length of the spacing periods without'af fectinglthe duration of the impulse periods. The systems are coordinated and controlled 111 a manner to be explained hereafter, so that for light values lying on one sideof an intermediate predetermined value,- only one of the systems is modulating, or varying the duration of its periods, the other system operating-to produce periods of minimum andsubstan'tially constant duration. When predetermined value. This will be clearly' the light intensity shifts to' the other side of the predetermined value, conditinnsare reversed; that is to say, the system which was previously modulating orvarying the duration of its periods operates to'produce periods of minimum and substantially constant duration, while the other modulates or varies the length of its period.

Thus it'will be seen that results are produced which are similar to those which would be obtained if one system were, so to speak,

inverted when the light intensity reached a seen from line D of Fig. 3, in which it will be seen that the'right hand portion ofthe line is practically the same as the left hand portion except that spacing and marking periods are inverted or the record is turned upside down.

Turning now to the operation of the system, it will be seen by reference to Fig. 1 that in the positionof relay 22 shown condenser 16 is completely short-circuited. This circuit may be traced from the lower plate of condenser 16 to armature 22, contact 23,

. through relay. coil 32 and to the upper plate of condenser 16. Therefore, no charge can be accumulated upon this condenser. Bias battery 20 is chosen of such a value that under these conditions, grid of the tube is highly jnegative and plate current flowing in relay- 25 is substantially zero.

On the other hand, condenser 17 is not short-circuited and is therefore free to take or store up a charge. Since the negative terminal of source B is connected to one side of ment that tube 15 is not drawing any plate current. Ultimately, condenser 17 will be charged to such a value that its potential 1s equal to the potential of source B. As the charge approaches this value, the potential upon grid of tube 19, which was maintained highly negative by bias battery 21, will grad- .ually be overcome .and as the condenser grid of tube 19 will be carried beyond the.

reaches its maximum charge, the potential of knee of the curve, and the. plate circuit of tube 19 will begin suddenly to take current. This current, acting through winding 26, will attract armature 22 away' from contact 23 against contact 24:.

It will be seen that the closure of armature 22 against contact 24 establishes a circuit from one side of condenser 17 through relay coil 33, contact 24, armature 22 to the other side of condenser This condenser, it will be remembered, has just been charged to its maximum potential. It will therefore be discharged through the path just traced and the rush of current due to the discharge will draw armature 29 against contact 31, thereby energizing line 35, thereby providing for throwing the transmitter key in one direction. If desired, resistances may be inserted for slowing down the condenser discharge.

More partlcularly, a resistance, may be illserted in the lead from armature 22 to the negative terminal of the source shown by -B. In this case only one resistance is necessary.

It may at this time be pointed out that the operation of armatures 22 and 29 is practically simultaneous. In fact, if desired, atmature 29 may be omitted entirely and armature 22 used to control thetransmitter, by the provision of a suitable set of contacts thereon. v

It will now be seen that tube 19 and its associated circuits have been rendered ingperative. No charge can build up on condenser 17 this condenser being short-circuited by armature 22. Bias battery 21 therefore maintains the grid of tube 19 highly. negative and this tube ceases to draw plate current for the time being. An identical cycle of operation then occurs with respect to tube 18; that is to say, condenser 16 begins to charge at a rate determined by resistance '27 and the current drawn' by vacuum tube 14. When it attains its maximum charge, the negative bias impressed upon the grid of tube 18 by bias battery 20 will be overcome. The plate circuit of tube 18 will begin to take current,

armature 22 will be retracted against contact 23 andcondenser 16 will discharge through relay coil 32,- thus drawing armature 29 against cont-act 30 and closing the other circuit to throw the armature key to the other position. I v

i It will now be apparent that if condensers 16 and 17, tubes 18 and 19 and resistances 27 -and28 are similar, the time taken by con densers 16 and 17 to attain their maximum charge will be the same and an oscillating system will be obtained; that is to say, armatures 22 and 29will move back and forth sub: stantially together from left hand contacts to right hand contacts and the period of time intervening between reversals of position of armatures 22 and'29 will be determined by the value of resistances 27 and 28 and the size of condensers 16 and 17 assuming that tubes 14 and 15 draw no plate current. Under these conditions and providing the transmitter (not shown) is controlled in any suit able way, uniform impulses will be transmitted, separated by uniform spacing pe-' riods, the spacing and. marking periods being equal to each other.

This condition may be regarded as the basic 'l or starting'point in explaining the function of tubes 14 and 15. If means can be provided whereby tube 14'will controlthe length of desired.

the marking impulses and have no effect on the spacing impulses and tube 15 can be so arranged as to control the length of the specing impulses without afiecting the marking impulses, the system will give the results It was-previously stated that when armature 22.1noved to the right away from contact 23, condenser 16 began to charge and the greater the positive potential'on the grid of tube 14, the greater the amount of current .taken by the plate circuit of tube 14 and the longer the time taken by condenser 16 to charge. 1 7

It on theother hand, no current isdrawn by tube 14, and-current is drawn by tube 15, the length of time required for condenser 17' when ithad acquired its maximum charge," to attain maximumcharge and move the its voltage was suflicient to entirely overcome l the voltage of bias battery 20 so that the tube 18 suddenly began to take current. It, how'- ever, tube 14 is drawing current at the same time that condenser 16 is charging, sinceboth these currents must be supplied from the source B through resistance 27, it will be apparent that agreater length of time will be required for condenser 16 to attain its maximum charge. The amount of current drawn by vacuum tube 14 and, therefore, the timetaken for condenser 16 to reach a point of maximum charge, is controlled by the intensity of light in the picture through the photoelectric cell.

For high lights, a high negative potential is impressed on the grid of tube 6 through bia'son vacuum the action of photo cell 1 and resistance 5. This cuts 0d the plate current flowing in tube 6 and the potential of the plate willrise due to the decrease in the drop through resistances 10, 11 and the internal resistance of the tube. This is an eflect wellknown to those familiar with vacuum tube amplifiers. If bias battery 9 is so chosen that as this change occurs, the potential of the grid of tube 14: will, be carried to a point more positive than the point at which the tube begins to-take current, then it will be seen that the amount of current takenby tube 14 will be the greater, the greater the intensity of light projected upon the photo cell and therefore the longer the time taken by cbndenser 16 to reach its fully charged condition.

t will be seen, assuming that vacuum tube 15 is not taking current, that the spacing and marking impulses or the operation-of armatures 22 and 29 will no longer be uniform as before. The contacts will move to the left,

opening the circuit around condenser I 17 which will charge in the same time as before, and thereby causing the armatures to be drawn to the right hand position in the usual time. v

If now the conditions with respect to tube 14: are as described, the condenser 16 will charge only slowly and the armatures will remain against the right hand contacts for .a considerably longer period than before, the

length of the peried being controlled for the ose of this invention by the intensity of light; that is, the more intense the light on the picture is, the greater wlll be the negative tube 6, the greater the POSI- relay contacts to the right will be greatly increased, exactly as pointed out with respect to vacuum tube 1 1; To provide that these two systems shall not work together; that is, to arrange that the marking periods shall not begin to increase in duration until the spacing periods have reached their minimum duration, it is only necessary to provide that when one of the tubes 14 or 15 stops taking current by reason of its grid becoming negative, the other of these two tubes is just beginning to take current; the efi'ect of this is that in the operation of the system, either condenser 16 or 17 will alwaysbe charging at its normalrate; i

The operation by whiehthe intensity of light controls the current flow through tube 14 has already been described. The controlof tube 15'is similar except that the phases are reversed. It will be remembered that for a relatively intense light, the plate of tube '6 increases its positive potential, thereby increasing the positive potential on the gridof tube 14. It will be seen that in a similar manner the positive potential on the grid of tube 7 will likewise be increased. This will cause a flow of plate current in tube 7 and a drop in potential of the plate of the said tube, thereby causing the grid of tube 15 to become sufiiciently negative to prevent the flow of current.

It will thus be seen that tube 14 is active in controlling the timeof charge of condenser 16 for intense light on the photo cell and that tube 15 is active in controlling the rate of charge of condenser 17 when no light is thrown on the photo cell. Under this latter condition, tube 14 has been renderedinoperative by its grid becoming negative.

For the purpose of making the action more clear, it may be considered that the grids of tubes 14 and 15 are out o'f'phase. Theoretically, the grid of one or" these tubes should not become sufficiently positiveto allow cur-- rent to fiow therein until after the grid of the other has just become sufficiently negative to cut off the flow of plate current.

The action of the various tubes may be seen from Fig. 4, in which the grid voltages of tubes 6, 7 1 1 and 15 are plotted for "vari-- ations of light as'curves a, b, a and 0? respectively. The dotted vertical line, represents the point at which the impulse periods are equal in duration to the spacing periods.

However, I find it advantageous" not to "tive potentialof the plate of vacuum tube 6 work with plate current cut-ofl' in tubes 14 and 15 exactly at this point, but to provide for a slight overlap, as shown by curves 0 and d of Fig. 4 so thatthe grid of one tube begins to become sufliciently positive to allow a slight plate current to flow before plate current has entirely ceased to flow in the other tube.

It may be pointed out at this time that if the amount of overlap is increased, the changes in spacing and marking time are not independent of each other. For instance, if this overlap is greatly increased starting with impulses of relatively. short length and widely spaced, as the spacing period decreases, the impulses are increased in length, and conversely, when the impulses are relatively long and separated only by small spacing periods, as the impulses become shorter in length the spacing periods between them will increase.

I find it advantageous to work only with a slight overlap, as shown, since by so doing only one of the groups of impulses is varying in accordance with light intensity and under these conditions, it is easier to adjust the apparatus to obtain the desired results.

The reason for this is that in all representations of light intensity in terms of current, it is advantageous'that the summation of the current periods shall be a true representation of the light intensity. This, may be stated as meaning that there should be a linear relationship between the value of the length of time the current is on, in any given unit of time, with respect to the strength of light. It is easier to make this ratio approach the theoretically correct ratio when the spacing or marking periods are changed independently of each other, except for the limiting conditions in the center of the light scale. At this point it is necessary to have a certain amount of overlap to maintain a lin-' ear ratio between the time the current is on in a given unit of-time and the light intensity due to the bending of thetube characteristics at the limiting position. In this connection, the problem of distortion in audio frequencyamplifiers, and its solution by the use of push-pull amplifiers, is analogous.

For some cases, I have found 'tdesirable to avoid the use of mechanical relays in order to obtain more reliable operation, and for other reasons; and an arrangement is shown in Fig. 2 in which armature 22and contacts 23 and 24 are eliminated and their function is performed by vacuum tuberrelays. In this arrangement the light sensitive cell with 1ts associated tubes 6 and 7, resistances 10, 11 and 12, bias batteries 9 and 13 and 'tubes36 and 37 are substantially identical with the arrangement shown in Fig. 1 and will not be described in detail.

. Connected to the plate or output electrode of tube 36 I provide resistances 38' and 40.

'The positive terminal of the sourceof plate 'istin Con enser 42 is charged through resistances Condenser 43, in a siinpotential is connectedto the free end of resistance 38. The other end of resistance 40 is connected to one plate'of condenser 42, the other plate of which is connected to the common filament lead. To the output or plate 7 the positive side of the source of plate potential through resistance 48.

The grid of tube 46 is connected through a suitable bias battery 44 to the upper plate of condenser 42. A circuit connected to the plate of tube 36 is arranged in a similar way including relay coil 52, condenser '51 and leading to the ,plate of vacuum tube amplifier 47, the grid of which is connected through suitable bias battery 45, to the top plate of condenser 43. The positive terminal of the source'of plate potential is connected to the f plate of vacuum tube 47 through suitable resistance 49. Coils 52 and 53, armature 54 and contacts 55 and 56 and line 57 correspond to coils 32 and 33, armature 29, contacts and 31, line shown in Fig. 1, their purpose and structure beingthe same. i The basic principles of operation of this arrangement are-the same as those given for Fig. 1. The left hand tube 46 determines the spacing interval, the right hand. tube 47, the marking interval andplate current in these tubes, is controlled by the potential exon condensers 42 and 43 respectively.

38 and in series. ilar manner, is char ed through resistances 39 and 41 in series. that the grid of tube 36 is sufficiently negative so that this tube takes no current, it will be seen that condenser 42 will charge at a.- rate determined by the size of resistances 38 and 40 and the potential of the source B. Q

The potential onthe top plate or condenser '42 will rise until it begins to overcome the effect of grid bias battery44 to allow tube ssuming for the moment.

46 tod'raw plate current throughresistance 48. This condition having been reached,'we will now pass over to the right hand side of the system to tube 47 which is assumed to be at such a position on the cycle that condenser 43 now begins to charge, assuming that condenser 42 has takenits mairimum charge.

mined by the value; of resistances 39 and 41 and the ,potential of the source B. -When condenser 43 has charged to such a value that the efiect ofbias battery 45 is overcome and;

Condenser 43 will charge at a rate detercurrent, a new situation develops. When the tube 47 begins to take current, it does so very rapidly, which in turn causes a sudden drop in the potential of the plate of the v tube.

' of source B.

connected directly to the plate oftube 47. so that this drop, which is a drop through resistance 49 due ,to plate current, will be passed through condenser 51 in such a way that condenser 51 will take a portion of the charge which has already been built up upon condenser 42. In other words, through the actionof tube 47, which begins suddenly to take plate current, condenser 51 becomes.

suddenly active as a condenser shunted around condenser 42. I

This condition may be set forth more clearly by an explanation on the basis of the electron theory. Assuming that condenser 42 is charged to its full value, there will be an excess of electrons on'the lower plate of condenser 42 and a deficiency on the upper plate corresponding to the potential of the source B. As long as vacuum tube 47 is taking no current, the potential of the left hand plate of condenser 51 will be the same as that of the upp erplate of condenser 42; that is to say, there will be a deficiency of electrons on this plate corresponding to the positive potential of source B. Since, however, the right hand side of condenser 51 isconnected also to the positive side oijsource B,'both plates of this condenser are at the same potential and no charge exists thereon.

Assuming that vacuum tube 47 now begins to take current in its plate circuit, there will be a rush of electrons to the plate of tube 47 and to the right hand side of condenser 51, thus partially making up for the deficiency of electrons thereon previously. The right hand plate of condenser 51 thus drops in potential, disturbing the electrostatic equilibrium between these plates and the change in equilibrium of this condenser, resulting from the ilow of electrons to the right hand plate, causes a flow of electrons from the left hand plate through coil 52 to the top plate of condenser 42, thus partially neutralizing the charge already upon condenser 42. This in turn lowers the potential'of the grid of tube 46 and the plate current flowing therein decreases. However, condenser 42 immediately begins to charge again, tending to carry the grid of tube 46 to such a potential that this tube begins to take plate current again. When tube 46 begins to take plate current again, the potential of the plate of this tube begins to drop and changes the'charge equilibrium of condenser 50, both sides of wh ch were previously at the potential of the posltive side The action is identical with that previ- The right hand plate of condenser 51 is weaves ously explained with respect to condenser 51, condenser 50 becoming suddenly active as a parallel shunt to condenser 43 and taking a part of the charge from condenser 43 in the same manner in which condenser 51 took part of the charge from condenser 42. s

We thus have a cycle of operations consisting of a charging of condenser 42 to such a point that tube 46 begins to take plate current, whereby the charge on condenser 43 is partially taken by condenser 50, cutting oil the flow of plate current in tube 47. Condenser 43 then begins to charge again, and when it ischarged, tube'47 begins to draw plate current again, which causes condenser 51 to take a part of the charge from condenser 42, cutting off the plate current from tube 46. Condenser 42 begins to charge again and the cycle is re eated.

lhe time taken y condenser 42 to charge determines the period elapsing between the time when plate current begins to flow in the tube 47 and the time when it is cut 0E. by the start of plate current of tube 46, and conversely, the time taken by condenser 43 to charge determines the period elapsing between the time when plate current begins to flow in tube 46 and the time when it is cut oil by the start of plate current in tube 47.

The practical effect of this is that plate current flows in tubes 46 and 47 alternately, and the change-over is not slow and gradual, but sudden andvery sharp, even when the system is producing a small number of oscillations per second. It is evident that there must be a limiting action taking place in these cycles, such that condensers 42 and 43 never rise in their total voltage value beyond a certain point. We will assume, for example, that we have come to thepoint in a given'cycle such that condenser 42 has its full charge so as to cause plate current to flow in tube 46. This causes condenser 50 to take a part of the charge from condenser 43 and stop the flow of plate current in tube 47 as explained above.

Under these conditions, the right hand plate of condenser 51 rises in voltage so that it virtually gives b'ackthe charge to condenser 42 which it previously took therefrom. The result of this reversing action is cumulative to cause a very rapid rise in the voltage on condenser 42, once it has attained such a charge tha't the effect of bias battery 44 is ovecome and tube 46 begins to take plate curren A. quasi-regenerative action may be obtained by properly proportioning coils 52 and 53 by making them in the right proportion to build up the charge on condenser 42 to a higher value once it starts to rise above the potential suflicient to allow plate current tc flow in tube 46. But if it remained at this action takes place which prevents the charge on condenser 42 from exceeding a certain value. This is due to the fact that the grid tube 46 is connected directly through the C battery to the top plate of condenser 42. condenser 42 rises in potential beyond a certain point, the grid circuit of tube 46 will begin to pass current and thus operate to.

that one side of the system, for instance, the,

marking side, will control the duration of the marking periods strictly in accordance with light intensity, regardless ofwhat may be the.

conditions in the spacing side and vice versa. The various resistances are so proportioned that this condition is always realized. In

other words, when condenser 51 gives back a charge direct to condenser 42, any excess charge will pass 0E through the grid circuit of tube 46 so that there is always a definite and identical starting point. The same is true for tube 47 and condenser43. To assist this action I find that it is sometimes of advantage to have tubes 46 and 47 so constructed as to be able to pass a comparatively large amount of grid current. A satisfactory way of providing for this is to use tubes having a larger number of turns of wire in the grid than is usuallyfound in the common types oftubes now on the market, but obviously this result may be obtained in other ways, as for instance by the use of a woven screen grid, perforated plate grid, or the like.

The marking and spacing intervals may be varied as in the arrangement shown in Fig. 1 by tubes 36 and 37. The greater the plate current taken by tube 36, the longer it will take for condenser 42 to reach its maximum charge and initiate the next portion of the,

cycle, and the same is true of tube 37.

Armature 54 is thrown from contact 56 to contact and back again by the surge of current through coils 52 and 53 whenever the condensers 42 and 51 and 43 and 50 respec' tively divide their charges, as explained above. Obviously'thi's relay. may control the transmitter directly or it may be used to control a second relay which controls the transmitter. It is of advantage to use a second relay to control the transmitter to act as a bug trap, well known in the telegraph art, to prevent reactions from the line connections into the tube circuits, and also to enable relay 54: to be operated with lower currents than would be permissible if this relay were required to control the transmitter circuit directly.

For the sake of simplifying the diagrams and descriptions, I have omitted various parts and circuits which are used in practice, such as filament heating sources, filament rheostats and the like, but as these matters are common knowledge in the art, ll think it unnecessary to describe them. i L

While I have shown anddescribed my invention with suflicient particularity to enable those skilled in the art to practice the same and obtain the benefits thereof, various modifications and changes may be made as will be apparent to those skilled in the art without departing from the spirit and scope of my invention.

Having fully described my invention, what I claim is:

1. A picture comprising a mass of dots varying in concentration in a manner to produce effects of varying light and shadow on,

one side of an arbitrary intermediate value, and a mass of lines varying in length to produce effects of varying light and shadow on the other side of said arbitrary value.

2. YA picture comprising a background having disposed thereon a mass of substantially uniform dotsto produce varying light values above a predetermined light value, and a mass of lines varying in length to produce efi'ects of shadows darker than said predetermined value.

3. A picture made up of substantially uniform dots and lines ofvarying length, values of light intensity on one side of a predetermined value being represented by dots of varying concentration and values of light intensity on the other'side of said predetermined value being represented by lines of varying length.

4. A picture. comprising parallel rows of marks of substantially uniform width and of substantially equal length and variably spacedto produce efiects of varying light intensity on one side of a predetermined value and of varying length and uniformly spaced to produce effects of.varying light intensity on the other side of said predetermined value.

RICHARD HOWLAND RANGER.

Images