US3364402A

US3364402A - Orthogonal code transmitting array

Info

Publication number: US3364402A
Application number: US553472A
Authority: US
Inventors: John S Davis
Original assignee: Bunker Ramo Corp
Current assignee: Bunker Ramo Corp; Eaton Corp
Priority date: 1966-05-27
Filing date: 1966-05-27
Publication date: 1968-01-16
Anticipated expiration: 1985-01-16

Description

Jan. 16, 1968 J. s. DAVIS ORTHOGONAL CODE TRANSMITTING ARRAY Filed May 27, 1966 11b \lc Hoi 5 m5 N H O J /lllii United States Patent 3,364,402 ORTHQGONAL CGDE TRANSMETTHNG ARRAY John S. Davis, Glendale, (Ialifi, assignor to The Bunker- Ramo Corporation, Stamford, Conn, a corporation of Delaware Filed May 27, 1966, Ser. No. 553,472 Claims. (Cl. 317-261) ABSTRACT (IF THE ENCLOSURE An apparatus defining a two dimensional surface useful with a movable probe for developing horizontal and vertical signals representing the position of the probe on the surface. The apparatus is comprised of first and second spaced sets of parallel conductors. The conductors of the first set extend perpendicularly to the conductors of the second set. Each set of conductors is woven on a loom with the conductors preferably forming the warp. The shoot is preferably formed by nonconductive elongated elements which serve to insulate the warped conductors from one another and to maintain them in place.

The present invention relates to an orthogonal code transmitting array and a method for making the same. More in particular, the invention refers to a device and a method for manufacturing such a device that is capable of generating digitally coded data in response to the path of a sensor placed upon or moved along a preferably plane surface of said device that may sequentially generate coded data descriptive of the location or the path of the sensor along said surface. The said sequentially generated coded data may, according to the invention in its broader aspects, be utilized in different fashions.

In the. first place, it is possible to decode the data which may preferably represent orthogonal coordinates that at each of sequential instants describe the location of the sensor and generate analog electrical signals in response to each of the data. These electrical signals may then be applied in the form of deflection signals to orthogonally disposed deflection elements of, if desired, remotely located cathode ray tubes so as to trace the path of the sensor at a remote location. A person can then, by manipulating the sensor along said surface, write or describe curves at the remote location. Of course, his manipulation may be guided by a display device that the operator can see, e.g., also in the form of a cathode ray tube having an image of suitable persistence or being equipped with controllable image retaining storage elements well known in the art, so that he can orient the sequentially generated portions of presentation, e.g., several parts of a handmade drawing, a set of curves, several portions of a single curve, etc., properly with respect to one another.

In situations of the general nature indicated in the foregoing paragraph, the communication for which the device of the invention is a tool or helpful means is primarily between persons. Another application of the device of the invention which is presently considered to be of even more interest is in the field of communication between a person and a computer. There are many situations in which the communication between a person and a computer may be greatly facilitated by the aid of an arrangement incorporating a device according to the invention. For instance, in situations where a digital computer is utilized to control a plant, there may be a plurality of input parameters and a plurality of output parameters. It may be known from experience gained in manual operation of the plant that, in response to a step change of the input parameters a certain change in one or more output parameters is desirable and possible to achieve which may or may not require simultaneous adjustments to one or more of the "ice other input variables. The exact mathematical relationships between the variables may not be known, such as because of the fact that the behavior of the plant, though partially understood, is too complex for a complete analysis. Nonetheless, the operator may have an idea of the change in the input parameters that will produce a desirable change in output parameters either instantaneously or over a given period of time. In such cases, it can be extremely useful to the operator if he can communicate with a computer that controls the plant by feeding to it presentations in the form of graphs showing the desirable and feasible response characteristic of, for instance, one parameter of the process. In other situations, the computer may adjust several input parameters simultaneously on the basis of information presented to it in the form of bundles of graphs. As the computer can receive information in digital form only, it is desirable that the information presented by the operator in graph form be readily converted to digital form.

In other situations, the operator may draw a curve from which the computer derives certain data in time sequence. These output data may then be displayed e.g., on a cathode ray tube, alone or in combination with the input data presented to it. In all these situations, the time necessary for programming the computer, e.g., for controlling a process, may be greatly reduced. For instance, a program may be quickly evaluated and, if thereby the comparison of graphical material fed into the computer and the graphical presentation produced by the computer shows obvious discrepancies due to incorrect programming or shortcomings leading to a less desirable form of control of the process, necessary improvements in the programming may be readily made. Some of the advantages of an analog computer in which a certain relationship between input and output data may be brought about by manually adjusting a number of potentiometers are thus available to digital computers without sacrificing any of their versatility because of their applicability to a great variety of programs.

Devices of the general type described capable of generating sequences of digital signals in' response to manually described graphs or other pictorial material are known from an article in Proceeding of Fall Joint Computer Conference, 1964, pp. 325-331, by M. R. Davis and T. 0. Ellis, entitled, The Rand Tablet: A Man-Machine Graphical Communication Device. The entirety of the subject matter disclosed by this article including specifically the driving and signal sensing arrangements referred to and shown therein is incorporated in this specification by reference.

This article describes an arrangement incorporating a planar nonconductive sheet carrying a plurality of closely spaced conductive strips formed on opposite sides of the sheet at right angles. Each of the individual strips is being supplied with a coded signal which identifies a particular strip and which is being picked up by a capacitive sensor that is held in close proximity to the sheet so that the signals from two strips located on opposite sides of the sheet define within the resolution of the system the coordinates of the sensor with respect to the sheet. The sensor, which is manipulated by an operator, generates a rapid succession of coded signals that successively define the sensor path. These digital signals may be fed into a computer which may be programmed to operate according to information supplied by an operator in the graphical form in the manner described.

The aforementioned article describes a planar sheet or tablet provided, at opposite surfaces thereof, with conducting strips formed by etching. The interstices between the strips, in a manner essentially similar to that followed in the manufacture of printed circuits, are removed from a continuous layer of copper, silver, or other conductive metal by etching techniques. A resolution of 100 lines per inch can be achieved in this manner, which is reasonably adequate for the tablet described and which allows the decoding of the movements of the sensor with an accuracy commensurate with that to be expected from the manipulation of the sensor.

Nonetheless, the etched tablet appears to have certain disadvantages which the present invention seeks to overcome.

One disadvantage is that, especially in the case of large tablets, it becomes very costly to make the originals, which are made on an enlarged scale and from which an image is formed on a laminated sheet, whereafter the etching on said sheet is induced in areas defined by the original. In the second place, the optical equipment required to reduce the image size becomes extremely expensive when both a high resolution and a high accuracy are required for matrices exceeding a dimension of approximately 12 X 12 inches. Furthermore, in employing etching techniques, certain difficulties are encountered in keeping close tolerances on the thickness of the conducting layer with attendant variations in impedance.

Attempts have been made in the past to overcome the disadvantages mentioned by using weaving techniques so as to form tablets or matrices of the type described. In these attempts, the matrices were formed by interweaving first and second sets of parallel conducting wires, each of the wires being covered with insulation, with one set forming the warp and the other set the shoot of a weave. Difiiculties were then encountered in maintaining the desired accuracy. If a piece of cloth is woven, this generally shows the tendency to deform in pillow-like fashion after it is taken from the loom; that is, one dimension (e.g., the width) measured across the lengthwise center of the rectangular piece tends to be less than the width measured at the corners. It is also very difiicult with this technique to keep the random errors low, such errors being caused, for instance, by diiferences in tension between individual warp wires or between individual shoot wires. Any wire that is woven in at greater tension will have a greater tendency to shorten its length in the woven piece and cause a slight local distortion in the shape of the wires it crosses by, so to say, pushing them out of the way. Furthermore, it becomes extremely difficult to prevent short circuits between individual shoot and warp insulated wires. Both the wires and their insulations have to be extremely thin because of the required resolution of these wires, which makes it diflicult to prevent damage in the weaving process.

It is an object of the invention to overcome such disadvantages of known weaving techniques for the manufacture of orthogonal matrices of the type indicated.

It is another object of the invention to provide a novel device in the form of a generally orthogonal matrix, comprising first and second sets of parallel conductors, said sets located in first and second closely spaced surfaces, each of the conductors of a set being closely and accurately spaced with respect to neighboring conductors and closely spaced in the completed framework of the matrix.

It is also an object of the invention to provide a method for making such matrices, said method allowing extreme cost reduction, manufacture of the said matrices in practically any size that might be required, and a high degree of accuracy.

According to certain broader aspects of the invention, each of the orthogonal sets of parallel conductors is individually woven on a loom, with the metal conductors preferably forming the warp. The shoot is preferably formed by nonconducting elongate elements and serves mainly to maintain the warp conductors in place. The nonconducting shoot elements may be formed of .spun fibers, but preferably a synthetic filamentis utilized because of its more closely held dimensional characteristics and mechanical properties. The shoot elements preferably have a much lower modulus of elasticity than the metal wires with which they are interwoven and are preferably also much thinner so that the metal wires are not substantially permanently deformed in the weaving process, and the resulting weave comprises metal wires that are almost perfectly straight. Also, according to a further aspect of the invention and in preferred embodiments thereof, the spacing of the nonconducting wires may be much closer, e.g., ten times as close as that of the conducting elements.

Two of such woven pieces are spaced closely and held with the conductors running at a right angle to form an assembly that-in conjunction with a suitablecapacitative sensor or probe and in conjunction with additional equipment for thegeneration of electrical signals identifying each conductor-allows the sensor to pick up signals to define its position on the path it describes as a function of time.

A preferred embodiment of the invention will now be described so as to facilitate the understanding thereof. However, it should be understood that other embodiments of the invention are possible and that the inventive concept is not limited to the example described. The surrounding equipment, such as the means for generating the signals to individual conductors and the sensors utilized to pick up the signals, is not described in detail inasmuch as a full disclosure of equipment useful for such purposes is to be found in the aforementioned article by M. R. Davis and T. 0. Ellis. It should also be understood that, whereas an important application of the invention appears to be in the field of communication between an operator and a computer, other applications are possible, e.g., in communication from person to person, in teaching and the like.

Reference is now made to the drawing, in which:

FIGURE 1 is a sectional schematic isometric view of a small section of a weave that may be utilized in building a device of the invention;

FIG. 2'is a cross section of the weave shown in FIG. 1 taken across the warp, showing the preferred geometry with greater accuracy;

FIG. 3(a) shows, in schematic form, a plane view of a portion of an assembly incorporating orthogonally oriented weaves of FIG. 1;

FIG. 3 (b) shows and end view in section of FIG. 3(a); and

FIG. 4 shows one way of making terminations to the conducting wires.

Referring now to FIG. 1, there is shown a plurality of parallel conducting wires 11a, 11b, 11c, and 11d, preferably made of low resistance material, such as copper. These wires form the warp wires of a weave. As will become more apparent when describing the structure formed by the shoot filaments, the warp wires should be closely spaced so as to define maximum resolution. (The diameter of the warp wires will be less than 0.005 inch.) No insulation is necessary for these warp wires, in sharp contrast to earlier proposed techniques in accordance with which a thin layer of insulation was necessary therefor. Preferably, in the weaving process, the tension of the warp wires may be controlled by utilizing a creel, which is device that may be used for individually controlling the tension of the warp wires during the weaving process by releasing the warp wires from individual spools.

As is customary in the weaving process, the spacing of the warp wires is controlled by the reed (not shown), which in this case should be manufactured with extreme accuracy (for a resolution of lines to the inch the reed should be manufactured to an accuracy of .00025 inch or thereabout). The spacing of the warp wires in FIG. 1 is shown considerably larger than in a preferred actual Weave according to the invention, for the purpose of a clearer presentation. There is further shown in FIG. 1 a plurality of shoot filaments 12a, 12b 12 woven at right angles with the warp wires so as to form a plain basket weave. Generally, the spacing of the shoot fila ments may be selected closer than that of the warp wires, up to about ten times as close (e.g., 1000 lines per inch). In the weaving process in customary fashion, the reed packs the shoot filaments one by one each time after a shuttle has projected a filament between a set of odd and even warp wires toward the fell of the cloth. In this process, the shoot filaments may be relatively tightly packed.

The structure resulting from this procedure is characterized by extreme resolution (determined, among other things, by the diameter of the metallic wire, which can be of the order of 0.002 inch or less). High accuracy and utility of the structure are not influenced by optical distortion and the impedance of the separate conductor. The thickness and width of the separate conductors as obtained by etching may vary to some extent. Furthermore, extremely large widths and even greater lengths can be woven without any loss of absolute accuracy. In weaving a 30 x 30 inch plane, a resolution of 6000 lines in total (200 lines per inch over 30 inches in two directions for two sets placed at right angles) can be easily achieved. No limitation exists on the length of the piece, whereas cloth looms are available that will allow the manufacture of pieces 20 feet wide or more.

It is possible, after a set of parallel conductors is made in this fashion, to protect the resulting structure against accidental damage by some suitable bonding process that applies a coating, or the like.

FIG. 2, which is a schematic cross section taken through a plane perpendicular to the warp of a preferred form of weave, shows the warp wires much more closely spaced than their thickness. In fact, the spacing can be determined by the thickness of the shoot filaments, the configuration of the loom, and the elasticity of the shoot filaments. The latter, of course, will undergo relatively large deformation when the reeds push them into the fell of the cloth, and, though the tension resulting therefrom is relatively small because of the small diameter and relative strength of the shoot filaments and their high degree of elasticity, there are limits in this respect, and too high a tension on the shoot filaments should be avoided.

The mesh in the shoot direction may be either higher or lower than that in the warp direction (the latter being determined by the degree of resolution required). The generally thin filaments used for the shoot can be shaped at a higher density. However, it should be understood that, when they are very closely spaced, the transverse tension executed by the shoot filaments on the warp wires will also increase, resulting in some slight deformation and a high pressure necessary for the reed to bring the individual shoot filaments into the fell of the cloth. On the other hand, it would be possible to decrease the density of the shoot filaments for such purposes as decreasing the cost of the weave and reducing deformation of the warp wires to a minimum, and in some cases the spacing of the shoot wires may be less than that of the warp. This is especially feasible if measures are taken to maintain the shape of the weave after it is removed from the loom, such as framing it or by applying a suitable coating.

FIG. 3(a) shows schematically one form of assembly of a conductor plane or matrix and shows one way of mounting the finished piece of cloth to a connector. The connector 17 in the form of a frame may be a piece of insulating material having recesses 18 at one plane surface thereof spaced at a distance to match the spacing of the Warp wires 12; that is, the spacing defined by the reed of the loom. The recesses each receive a bare metal warp wire. The ends of of the warp wire may be cut to alternate lengths for the odd and even wires, as shown. Connecting wires Ida-14a, having insulating sleeves l5a-1Sd, are placed in contact with the warp wires. Wires 14a-14d are connected to a suitable signal driving source (not shown) of a type, for example, referenced in the aforementioned article by M. R. Davis and T. 0. Ellis. The wires therefore constitute means for driving the matrix in the manner contemplated by the present invention. The connecting wires and the warp wires may be coated with tin, and an induction heated roller used to connect them while they are positioned in a suitable tool. On the other hand, it would be possible to connect the warp wires, after cutting them individually to a suitable length, to a connector pin placed on a support.

Furthermore, it Will be readily recognized that alternate techniques might be utilized for making terminations. For instance, for situations in which the total number of warp wires is very large, it may be advantageous to make a terminal device comprising in itself a woven structure such as to facilitate the supplying of individually different signals to each of the warp wires.

FIG. 4 shows the configuration of the assembly formed by two sets of parallel conductors, each set forming a separate plane or matrix. Two sets of matrices 2t] and 21 are shown, having their metallic wires 11 and 11' running at right angles to one another. They are held spaced in a framework comprising

rectangular frames

23, 24, 25, and 26 which may be held together at the four corners by bolts 27. Terminations may be made in the manner described in conjunction with FIG. 3. For instance, recesses in the surfaces of

frames

23 and 24 that are facing one another may be utilized to receive the parallel wires that form the

sets

20 and 21. Between the

frames

23 and 24, an insulating sheet of Mylar or a similar material may be clamped suitable for keeping the sets of wires apart.

In operation, the two sets of parallel conductors orthogonal to one another may cooperate with a capacitive pickup device connected to a high input impedance amplifier. Signals impressed on one of the conductors that is closest to the pickup are impressed thereon. The signals that are supplied to the conductors or generally the warp wires may be coded in serial Gray code so that, if the pickup would be between two warp wires, an error not greater than one-half the interval from one warp wire to the next will result.

In another embodiment, the signal to each of the warp wires may be a timing signal, comprising a single pulse instead of a coded group of pulses. The first wire receives a pulse at time I, the second one at time 2, etc., from a distributing network connected to a counter that generates a series of pulses equal to the number of wires of a set. When a pulse is picked up by the pickup device from one of the wires, a counter at the receiving installation, which is running synchronously with the counter that supplies pulses to the wires in sequence, may be stopped, thereby identifying the one wire of the set that is in proximity to the pickup by the count registered on the counter at the receiving end.

Of course, many variations may be introduce-d in the coding scheme used for identifying the warp wires of a single set. For instance, it is possible to use the Gray code between groups of warp wires and identify wires within each of the groups by the counting scheme briefly described.

Though in the foregoing it has been described that the elements 12 are preferably nonmetallic filaments, which is better from a mechanical viewpoint, specifically in that they allow more stretch, it should be obvious that other combinations are possible; e.g., the conducting Wires 11 could be provided with an insulating sheath, in which case metallic wires could be substituted for the nonconducting, more elastic filaments 12. This could be done especially under circumstances where the mechanical requirements on the elements 12 are less severe.

It has been described, particularly in conjunction with FIG. 4, that the two sets of orthogonally oriented conductors cross one another at a small distance, necessarily resulting in a smaller capacity between the wires of the set more remote from the pickup and the latter than the capacity for the wires of the other set. Taking into account that there is a certain screening or shielding effect by the wires that are closer to the pickup upon the signal transmission from the more remote set to the pickup, the signal from the more remote set of wires will come through at a strength generally about three times lower than that of the set that is close to the pickup if no measures are taken for compensation. Compensation may be made by varying the sensitivity of an amplifier (not shown) connected to the pickup or by varying the relative signal strength between the sets of wires. The coding signals identifying an individual wire of one set may, for nstance, alternate with the coding signal identifying a Wire of the other orthogonal set, and the compensation may be induced to become operative for one of the sets only, so that the signals identifying the location of the pickup in the orthogonal coordinates are recorded at substantially equal output levels.

It is furthermore obvious that the speed with which the location of the pickup is recorded should be calculated so that movements of the pickup can be recorded quasicontinuous. When the pickup is moved by hand, this required only moderately fast or even relatively slow pulse repetition rates, depending also, for instance, upon the degree of resolution required, as this determines the number of bits in a Gray code required to identify each conductor of a set.

It will be apparent from the foregoing that, although there have been described plane sets of parallel conductors, different arrangements are possible. For instance, rather than being planar, the set or sets may be shaped to conform to the surface of a cylinder and disposed for rotation about an axis. The capacitive pickup might then be movable along the surface thereof and in a direction parallel to the axis of rotation of the drum. Alternatively, the cylindrical matrix may be stationary about its axis and the probes adapted to move around its periphery and be translated along the surface in a direction parallel to the axis of the matrix. Furthermore, first and second capacitive pickup elements might be used for each of the sets of conductors and the codes impressed on each of the wires might then be picked up and analyzed simultaneously instead of in alternate periods of a cycle. Of course, in this case, additional equipment may be necessary when it is still desired to analyze and decode the movement of a pen-shaped, freely movable pickup device, but many applications may exist in which the path of the pickup device is determined in different ways than by manipulation. The foregoingexamples and descriptions of different embodiments merely illustrate that the invention is not restricted to the examples given or to equipment particularly adapted to a specific type of surrounding equipment. The invention itself is defined in and by the following claims.

The embodiments of the invention in which an exclusive property or privilege is claimed are defined as follows:

1. An array useful in transmitting coded signals to a probe that is placed in proximity to said array, said array comprising:

a first set of parallel conducting wires, each connectable to a signal source capable of generating a signal that identifies a particular wire in said first set;

a second set of parallel conducting wires, each connectable to a signal source capable of generating a signal that identifies a particular wire in said second set;

first and second sets of elongate elements, interwoven respectively with said first and second sets of parallel conducting wires, so as to maintain the parallel conducting wires of each set at a predetermined distance from one another and thereby forming first and second weaves; means for electrically insulating said parallel conducting wires from one another; and

means for maintaining each of said first and second weaves relatively closely spaced from one another, with the conducting wires of said first weave extending perpendicular to the conducting wires of said second weave.

2. The array as defined in claim 1 wherein in each of the weaves the conducting wires form the warp and the elongate elements form the shoot of the weave.

3. An array useful in transmitting coded signals to a probe that is placed in proximity to said array, said array comprising:

first and second sets of elongate nonconductive elements, interwoven respectively with said first and second sets of parallel conducting wires, so as to maintain the parallel conducting wires of each set spaced at a predetermined distance and insulated from one another and thereby forming first and second weaves; and

4. The array as defined in claim 3 and having the additional feature that said nonconducting filaments are made of a material having a substantially lower modulus of elasticity than said conducting wires.

5. The array as defined in claim 4 wherein the diameter of said nonconducting filaments is substantially less than that of said conducting wires.

' References Cited UNITED STATES PATENTS 1,576,934 3/1926 Running 317 261 X 2,278,538 4/ 1942 Dublier 1741 17 2,669,646 2/1954 Ford 174--117 2,686,222 8/1954 Walker 17819 2,975,235 3/1961 Leitner 178-18 3,257,500 6/1966 Rausch 1741 17 V FOREIGN PATENTS 1,080,592 4/ 1960 Germany.

LARAMIE E. ASKIN, Primary Examiner.

E. GOLDBERG, Assistant Examiner.