Search Images Maps Play YouTube News Gmail Drive More »
Sign in
Screen reader users: click this link for accessible mode. Accessible mode has the same essential features but works better with your reader.


  1. Advanced Patent Search
Publication numberUS3551616 A
Publication typeGrant
Publication dateDec 29, 1970
Filing dateJan 15, 1969
Priority dateJan 15, 1969
Also published asCA922638A, CA922638A1, DE2001703A1
Publication numberUS 3551616 A, US 3551616A, US-A-3551616, US3551616 A, US3551616A
InventorsHans Y Juliusburger, Morris Krakinowski, George R Stilwell Jr
Original AssigneeIbm
Export CitationBiBTeX, EndNote, RefMan
External Links: USPTO, USPTO Assignment, Espacenet
Multiple switch encoding device
US 3551616 A
Abstract  available in
Previous page
Next page
Claims  available in
Description  (OCR text may contain errors)

United States Patent [72] Inventors Hans Y. Juliusburger Putnam Valley; Morris Krakinowski, Ossining; George R. Stilwell, Jr., West Nyack, N.Y.

[21] App1.No. 791,406

{22] Filed Jan. 15, 1969 [45] Patented Dec. 29, 1970 [73] Assignee International Business Machines Corporation Armonk, N.Y.

a corporation of New York [54] MULTIPLE SWITCH ENCODING DEVICE [5 6] References Cited UNITED STATES PATENTS 3,120,584 2/1964 Grunfelder et al 200/166.1X

Primary Examiner- Robert K. Schaefer Assistant Examiner-H. J. Hohauser Attorneys-Hanifin and Clark and John .I. Goodwin ABSTRACT: An encoding device is provided in which a first plurality of parallel conductive lines are located proximate to a second plurality of lines extending normal to the first plurality. The intersection points of the lines are selectively connected by switches to connect pairs of the lines to form current paths. The lines are also permanently connected together by contacts to reduce the number of output lines required for a given number of switches. The first set of lines is located on the upper surface of a substrate and the second set of lines is located on the lower surface of a diaphragm separated from the substrate by an insulating layer having holes at the intersection points of the lines. The lines are connected by pressing the diaphragm through the holes.


FIG. 2

ATTORNEY- Pmmmnmmm' 3.551.616

sum 3 or 5 FIG. 4


MULTIPLE SWITCH ENCODING DEVICE BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is in the field of multiple switch or binary encoding devices which selectively interconnect multiple SUMMARY OF THE INVENTION In conventional multiple switch encoding devices, pairs of electrical conductors are brought into contact by the switches. If there are N switches, the device will have ZN output lines. It is an object of the present invention to provide a novel multiple switch which has fewer than ZNA output lines. In the device of the present invention, the switch encoding portion is connected into tensors of lower rank by a scheme referred to as contact encoding.

The foregoing and other objects, features and advantages of the invention will be apparent from the following more particular description of the preferred embodiments of the invention, as illustrated in the accompanying drawings.

BRIEF DESCRIPTION OF THE DRAWINGS In the drawings: FIG. I is a schematic diagram of a conventional pushbutton switch arrangement which is used to explain the present invention;

FIG. 2 is a schematic diagram of a conventional multiple switch encoding device which is used to explain the present invention;

FIG. 3 is a schematic diagram of one embodiment of the present invention employing switch encoding and contact encoding; FIGS. 4A and 48, when combined as shown in FIG. 4, show a schematic diagram of another embodiment of the present invention employing switch encoding, contact encoding and self-encoding;

FIG. 5 is a partial view of a section of a particular encoding device which can be employed in the present invention; and

FIG. 6 is a schematic diagram of a switching arrangement which may be employed in the present invention.

DESCRIPTION OF THE PREFERRED EMBODIMENTS In many electronic devices and systems, a large number of signals are controlled and operated by reducing each signal to a code, for example. the binary code in which different combinations of electrical circuits are selectively and separately actuated. A switching device for encoding such signals is required and a matrix switch has often been employed.

A matrix switch is a useful encoding device which includes a first group of electrical conductors or lines usually arranged in parallel and a second group of parallel electrical conductors or lines arranged normal to the first group to form a grid having a number of possible intersection points. Each intersection point includes a switch element connected between a first conductor and a second conductor so that pairs of conductors can selectively be brought into electrical contact. Each switch element brings a unique pair of conductors into contact so that the signal produced by bringing a pair into contact is a coded indication of the particular closed switch.

If the first group contains I conductors and the second group contains J conductors, I X J intersection points are formed so that the number of switch elements P is equal to l X J. If each group contains the same number of conductors (l matrixsvjtch, expression l is g,

J then P is equal to I. The matrix switch is used in combination with electronic devices which utilize the code. It can be seen that if a matrix switch having a large number of switch elements is required, a large number of external conductors will be required and therefore the number of necessary electronic devices will also be large.

For example, if 64 switch elements are required, 16 cond uctors are necessary, eight in each group for a square matrix (8 8=64). The general expression for the relationship between the number of conductors I for a given number of switch elements P is v I= 2P" (1) The value 2" in the expression is determined FEET??? that each switching element connects two conductors thereby providing two outputs for each switching element. One of the important features of the present invention isthe reduction of the number of external conductors representing the P switching elements by increasing the number of outputs per switching element. Thus, if expression (1) can be generalized as I nP where n is the number of outfii tsper switch, then the number of external conductors for given values of P can be reduced.

For example, presume that a large value of P such as 4,096 is required. For 4,096 switching elements in a conventional I= 2X 4,096 2X 64= 128 However, if n can be 4, then Thus, 96 less external conditiors are conhected to the logic devices for the same value of P as a consequence of increasing n. Increasing the value of n becomes a topological problem as well as an electrical problem, particularly when the conductors occupy only two planes. An explanation of the theory of the present invention follows wherein the elements of the invention are described in terms of vectors and tensors.

When there are P switches each connected to one of P input lines, the resulting information produced by switch closure can be compressed at the most into log P output lines. This is defined as encoding. This encoding procedure can be arbitrarily broken down into two partial encoding steps, the first performed by switches, and the second by means of electronic devices, such as diodes and/or transistors. An illustration of a simple two stage encoding device is shown in FIG. I. The first stage 10 consists of pushbutton switches P,, P P and P each respectively connected to output lines L,, L L and L The output lines are connected by diodes in second stage 12 so that two-bit binary code words 00, 01, 10 and l l are produced when switches P, through P, are respectively closed.

Assuming that the switches P, through P are activated one at a time only, any input can be considered as a vector of rank land be designated P (i= 1, 2 I) The magnitude of any particular vector is given by the specific value of i. The total set of vectors represented by the switches may be called a vector field. The magnitude of the vector field inthe above example is represented by I. In FIG. 1, I 4, which is the same as the magnitude of the largest vector.

In the example illustrated in FIG. I, the output of the first stage 10 (i.e., the input to the second stage 12) is a vectors L, (i= 1,2...) where I=4.

In further developing the vector concept, it should be noted that advantage can be taken of the fact that a switch P has two contacts which both can be used as information bearing output lines. FIG. 2 illustrates a matrix switch having sixteen switch elements in first stage 14. The lines L,, L,, L, and L. of FIG. 2 and L',, L' L, and L are in different planes and are normally not connectedatthe l6 intersection pointibut contact can selectively be made by various methods. An ill ustration of a matrix switch of this type wherein contact is made by a stylus which presses the two lines together is described in the aforesaid U.S. Pat. No. 3,308,253, issued Mar. 7, I967, to M. Krakinowski. v

The switch described in the patent includes a group of parallel conductors printed on the upper surface of a substrate. A spacer layer is placed over the upper surface of the substrate and contains holes at the intended intersection Thus, P

points.A diaphragm of elastic material having a group of .parallel conductors printed on its bottom surface is placed on 1 the spacensu'ch that the two groups of conductors are normal These quantities maybe defined as vectors of rank 2.

The outputs of the first stage 14, (i.e., lines L,L,, L,-

'L',) aredefined as vectors of rank I referred to as L, and L, L respectively and provide a partially encoded code using a total of eight lines to the electronicstage 16. These output vectors 1 are related to the switch elements by the cartesian product P L,- X L The physical significance of the cartesian product is that activation of a particular switch P,,- causes electrical connections to be established between the switch P,, and the outputs L, and L,. For example, if switch P is closed, current can flow from L via the switch to L',,..

As previously stated, it is very desirable to perform as much encoding as possible in the first or switching stage to reduce the number of output lines in order to minimize the number of components in the second stage containing the relatively more costly electronic components. The number of electronic components is related to the number L of input lines (output lines from the first switching stage). Thus, the object is to reduce the number L and, as previously stated, this is accomplished by making the number n as large as practically possible. When n,is greater than two, the resultant switching vector becomes .more complex than P and requires more than two-dimensional space for appropriate representation and presents a serious problem when the conductor patterns are confined to printed wire arrangements on two two-dimensional layers.

' Also, on each of the two-dimensional surfaces such as in the topological limitations, the tensor concept to be described becomes useful. v

The use of the tensor concept adapts itself conveniently to the description of planar switching patterns in that through the utilization of subscripts and superscripts horizontal and vertical components of separated.

Since the output lines ofithe EDS can be considered as the encoding device are kept notationally horizontal and vertical, the output lines L, and L, will now be designated L" for vertical lines and L,, for horizontal lines.

:Each switch is defined by the cartesian product L,, X L and is designated by the notation P,,. 1

In the case where n is 4, each switch has four. outputs and the switching vector is written P i;-

Unlike the usual case where n is 2, the number of switches is not the cartesian product of two output tensors L, and L,. For an n 2, the switches are the product of intermediate horizontal and vertical tensors designated K. =K X K z t X where is a mixed tensor of rank 4 and K and K are rank 2 tensors represented by a pair ofconductors.

The K tensors must be further encoded into the L vectors by a scheme referred to as contactencodingsince the tensors are encoded by direct contacts between lines rather than by switches.

The multilevel encoding scheme using two planar surface line arrays can now be formally described as follows: Given an original input represented as a tensor field of the 1 2... i for n P hlhL a. first; encoding level uses switch encoding to partition the original-tensor field into two intermediate tensor fields K and K r such that the cartesian product of- A the intermediate tensors equal theEor'iginal tensor field =Kh1h z "I 71 Subsequent levels of contact encoding on the same planar surfaces perform further partitioning until the final set of tensors (or vectors)L Li ..L a.nd L",, L", L", is obtained,

The ultimate relationshipfi keferring fo FIG. 3, a switch encoding arrangeTnent is shown wherein the number of switches P 81 and n is equal to with n equal to 2 would require nine horizontallines and nine vertical lines, giving a total of l8'o'utput lines. In the encoding arrangement of FIG. 3, each switch element has four outputs and the total number of output lines is 4 X. 81% 12. However, to carry out the encoding, the device of FIG. 3 requiresa switch encoding stage and a contact encoding stage.

In FIG. 3, the encoding switch arrangement is shown in schematic form. In actual construction, one set of lines, for ex ample, the horizontal, are printed-on the top surface of a substrate. There are 18 horizontal lines divided into nine pairs of two lines each. A spacer consisting of insulated material having holes at predetermined points, such as holes 20, 22, 24, 26, 28, 30, etc., is placed over the substrate. A diaphragm" of deformable material is placed over the spacer. The diaphragm has 18 vertical lines printed on its bottom surface suchthat an array of horizontal and vertical lines isproducedas schematically shown in FIG. 3. r V V The horizontal and vertical lines are insulated from each other by the spacer, however, the intersections of the horizontal and vertical lines occur at the hole positions of the spacer. Each switch element actuation "consists of pressing the diaphragm locally above pairs of holes to bring two horizontal and two vertical lines into contact as the diaphragm is deformed into the two holes. Thus, switch 32 (depicted by the square area in FIG. 3) connects line 34 to line 44 via hole 20 and line 36'to line 46 via hole 22. Likewise, switch 38, when depressed, connects line 40 to line 44 and line 42 to line 46. All 81 switch elements in FIG. 3, when actuated, connect twb horizontal lines to two vertical lines and provide the tensors P 2 5: of rank 4.

2 The intermediate tensors K m], and KT, are

further encoded through contac t enc ciiin g. Contact en-' lines, preferably at the edge of the EDS. W

is shown in FIG. 3 in block 48 and tensor K is shown in block 50. The intermediate tensors 'rEf'iirik 2 and each has a magnitude of 3. The contact encoding is arranged so that each switch closure (one of 81 code) which connects four lines (four of 36'code) is further encoded into a four of 12 code. The 12 output lines are labeled'll 2, 1, H H'i "'2, and V,, V V;,, V,, V',, V',,. Each switch closure produces a unique connection ofI-I, V, H or V lines. For example, switch 32 connects H, to V, and H, to V',, switch 52 connects H, to V, and H, to V' Switch 56 connects H, to V,

and H, to V',. Switch 58 connects H, to V, and H, to V,. 1 Thus, for each closure of one of the 81 switches, current flows The contacts for the contact encoding can be constructed in a number of ways, for example, by plated through'holes or by a clamping device which permanently presses the diaphragm through holes at the contact points.

The essential feature of the encoding device is that unique pairs of lines are brought into contact. One manifestation of such contact is to have current flow in the connected lines. Thus, H,, H H;,, H',, H, and H, may be connected through resistors 59 to potential sources +V and V,, V V,,, V',, V, and V may be connected through resistors 60 to ground. No current will flow in the lines until a switch is closed to create a current path from +V to ground which produces voltage drops in the output lines. The lines H,, H H H',, H,, H',,, V,, V,, V;,, V',, V, and V, are then connected to electronic utilization devices, such as transistors, diodes, etc.

The foregoing discussion illustrated how an encoding device can be formed in two planes using a combination of switch encoding andcontact encoding, such that less output lines are required than conventional encoding devices. The encoding device was restricted to horizontal lines in one plane and vertical lines in the other plane. An alternate approach can be used to reduce the number of contacts used in the contact encoding, however, horizontal and vertical lines are located in both planes. The alternate approach is referred to as self-encoding.

Referring to FIGS. 4A and 43, an encoding device is shown having 256 switching elements (P 256, n 4, L 16).There are 32 horizontal lines and 32 vertical lines arranged in groups of two, such that each switching element has two intersection points and four outputs. As in FIG. 3, there are two intermediate K tensors, however, one is formed by contact encoding while the other is formed by self-encoding produced by having certain horizontal lines wrap around" via vertical line connections and certain vertical lines wrap around" by horizontal line connections. Thus, the second horizontal line 72 extends vertically and becomes the 16th horizontal line, again extends vertically and becomes the 18th horizontal line, and extends vertically once again to become the 32nd horizontal line. The fourth, sixth and eighth horizontal lines 74, 76 and 78 also wrap around" and extend through the switches a total of four times. 18

Thehorizontal lines which are not self-encoded (the odd lines in the example of FIGS. 4A and 4B) are connected together in groups of four by contact encoding. Thus, the first, third, fifth and seventh lines 71, 73, 75 and 77 are connected together by contacts as are the other three groups of four odd lines each. The vertical lines in the other plane are arranged identical to the horizontal lines but oriented by 90 90. The H,, H,, H, and H, output lines are derived directly from the first, ninth, 17th and 25th lines lab'eled' respectively 71, 73, 75 and 77. The H H' H' and H, lines are derived directly from the ends of the wrap around lines 72, 74, 76 and 78 respectively. The same is true in the vertical plane, V,, V V and V, labeled 81, 83, 85 and 87 and V',, V',, V, and V, labeled 82, 84, 86 and 88.

Each switch closure will connect a unique two pairs of two leads each to produce a four-bit output code (the electrical portion of the encoding device shown in FIG. 3 has been omitted from FIGS. 4A and 4B for clarity). Thus, closing switch 80 connects H, to V, and H, to V, permitting current to flow in these four lines. Closing switch 90 connects H, (via a contact and line 71) to V, and H, to V',. Closing switch 92 connects H, to V, and H, (via lead 72) to V,. Switch 04 connects H, (via a contactj to V, (via a contact) and H, (via lead 76) to V',.

In like manner, each of the '256 switch elements connect a unique group of four lines to establish current flow. There are a total of 16 output lines. In a conventional matrix encoder, 256 switches require 32 output lines. Since each of the output lines are normally connected to a transistor or similar electronic element which is relatively more costly than the encoding switch, it is more desirable, for example, to have a four out of 16 code rather than a two out of 32 code for the same number of switching elements.

FIGS. 3 and 4A and 4B illustrates encoding devices wherein n 4 and each switch element has four outputs. This is accomplished by two switch points actuated at the same time. For values of n greater than four, more lines and more switch points are included in each switch element.

It was previously stated that when the encoding device of the present invention is embodied in an EDS, the switches are actuated by means of a stylus which depresses the diaphragm through holes in a separator to make contact with the substrate. FIG. 5 shows a specific arrangement for closing the switches in an EDS. FIG. 5 is a representation of a sectional view taken through two holes of the separator layer. FIG. 5 is drawn for purposes of explanation and therefore the relative scales of the element are not uniform. In FIG. 5, the substrate contains the first two lines 102 and 104. The separator layer 106 is mounted over substrate 100. The diaphragm 108 contains the other two conductors 110 and 112. An elastom etric actuator sheet 114 containing conically-shaped bumps 116 and 118 is placed over the diaphragm 108, A relatively thick layer of soft elastometric material 120 is placed over the actuator sheet. Force is applied, for example, by finger or by Sty-- lus 122, and transmitted to the actuator bumps 116 and 118 which cause lines 110 and 112 to contact lines 102 and 104 respectively.

The structure of FIG. 5 is particularly useful if more than one contact is used for each line connection. If desired, in order to insure switch closure, more than one contact point may be used to each line connection. Thus, in FIG. 6, lines 124 and 126 are brought into contact through holes 128 and 130. Likewise, lines 132 and 134 are brought into contact through holes 136 and 138. In such case, the actuator sheet 114will contain four bumps. The thick elastomer 120 serves to spread the force applied by the stylus. The localized load will be supported by at least three or all four of the bumps. Any combination of the at least three bumps will bring the necessary two lines into contact.

What has been described is a unique encoding device including a plurality of switches wherein the number of output lines from the device for a selected number of switches is minimized. The encoding device has been embodied as a keyboard employing elastic diaphragm switch technology. The invention is not limited to this technology however. For example, an optical encoding device using the principles of the present invention may be formed wherein the horizontal and vertical lines are connected at their intersection by photoconductors which, when dark, provide anopen circuit condition. The switch closures can then be made by directing a light beam at selected photoconductors which, when illuminated, provide current paths between the lines. In the optical version, the same theory of switch encoding and contact encoding to reduce the number of output lines is employed.

While the invention has been particularly shown and described with reference to preferred embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

We claim:

1. An encoding device for selectively connecting electrical conductors comprising:

a first support member located in a first plane;

a first plurality of conductive lines disposed on one surface of said first support member; 7

a second support member located in a second plane proximate to said first plane;

a second plurality of conductive lines disposed on one surface of said second support member opposing said first plurality of conductive lines on said first support member;

a plurality of P switching elements for selectively connecting conductors of said first plurality to conductors of said second plurality, each one of said P switching elements adapted to connect a total of n conductors where n is an integer greater than 2;

a third plurality of electrical conductors for permanently conductors comprising:

a rigid substrate having upper and lower planar surfaces;

a first plurality of conductive lines disposed in parallel on the upper surface of said substrate;

a layer of insulating material mounted on said upper surface of said substrate and containing a plurality of holes arranged in rows and columns each column of holes being located above and colinear with a separate one of said first plurality of conductive lines to provide access to said lines through said holes;

a layer of flexible material having upper and lower planar surfaces mounted on top of said insulating layer and having a second plurality of conductive lines disposed on said lower surface in a direction normal to the direction of said first plurality of lines each of said second plurality of lines being located above a separate row of said plurality of holes;

a plurality of depressible switch members located above the upper surface of said flexible member, each switch member adapted to be depressed to press said flexible material through at least two adjacent holes in said insulating material to bring at least two of said second plurality of lines into contact with at least two of said first plurality of lines;

a third plurality of conductive lines, each one permanently connecting together separate groups of at least two of said first plurality of conductive lines to form output lines; and I a fourth plurality of conductive lines, each one permanently connecting together separate groups of at least two of said second plurality of conductive lines to form output lines.

3. An encoding device for selectively connecting electrical conductors comprising:

a first plurality of electrical conductors located in a first plane;

a second plurality of electrical conductors located in a second plane proximate to said first plane;

a plurality of switching means for selectively connecting conductors of said first plurality to conductors of said second plurality;

a third plurality of electrical conductors for permanently connecting together separate groups of at least two of said first plurality of electrical conductors in said first plane to form output conductors; and

a fourth plurality of electrical conductors for permanently connecting together selected ones of separate groups of at least two of said second plurality of electrical conductors in said second plane to form output conductors.

4. An encoding device according to claim 3 wherein each one of said plurality of switching means is adapted to connect at least two pair of said electrical conductors each pair consisting of one conductor from said first plurality of electrical conductors and one conductor from said second plurality of electrical conductors.

Referenced by
Citing PatentFiling datePublication dateApplicantTitle
US3676615 *Jul 13, 1970Jul 11, 1972IbmPushbutton keyboard switch array and associated printed circuit logic cards
US3684842 *Jul 1, 1970Aug 15, 1972Texas Instruments IncPushbutton keyboard switch assembly with improved over center diaphragm contact
US3693775 *Jun 30, 1970Sep 26, 1972IbmActuator for elastic diaphragm switch keyboard
US3696908 *Nov 9, 1970Oct 10, 1972Sperry Rand CorpCapacitive key
US3710137 *Aug 6, 1970Jan 9, 1973A StephensControl system and method
US3721976 *Feb 1, 1971Mar 20, 1973Omron Syst IncKeyboard coding and interlock system
US3731015 *May 18, 1971May 1, 1973Universal TechnologyControl keyboard switch with cantilevered contact and diode matrix array
US3797630 *Jul 12, 1971Mar 19, 1974Alphanumeric Holdings LtdKeyboard for electronic circuit
US4066850 *Jun 4, 1976Jan 3, 1978Ncr CorporationKeyboard switch assembly having interchangeable cover plate, indicating layer and actuator switch assembly in any operative combination
US4071718 *Oct 22, 1976Jan 31, 1978Bowmar Instrument CorporationFlat keyboard assembly having cover type membrane with protrusions to align switch components
US4293849 *May 23, 1979Oct 6, 1981Phillips Petroleum CompanyKeyboard encoder using priority encoders
US4727225 *Jul 10, 1986Feb 23, 1988Illinois Tool Works Inc.Electrical switch for actuating a plurality of circuits
US5193669 *Feb 28, 1991Mar 16, 1993Lucas Industries, Inc.Switch assembly
US5374796 *Oct 28, 1993Dec 20, 1994Picker International, Inc.Contact sensor with improved sensitivity
US5684279 *Sep 12, 1995Nov 4, 1997Key Tronic CorporationComputer keyboard with improved membrane keyswitch structure having deflection concentration feature
US5900829 *Jul 23, 1996May 4, 1999Motorola, Inc.Method of and apparatus for detecting key actuations
WO2004077679A1 *Feb 28, 2003Sep 10, 2004Infineon Technologies AgMethod and apparatus for decoding a key matrix
U.S. Classification200/517, 400/479, 439/43, 341/34, 361/805, 200/5.00A
International ClassificationH03M7/04, H03M7/00, H03M11/22
Cooperative ClassificationH03M11/22, H03M7/00
European ClassificationH03M7/00, H03M11/22