US3774116A

US3774116A - Electron beam addressable memory system

Info

Publication number: US3774116A
Application number: US00250803A
Authority: US
Inventors: F Marlowe; C Wine
Original assignee: RCA Corp
Current assignee: RCA Corp
Priority date: 1972-05-05
Filing date: 1972-05-05
Publication date: 1973-11-20
Anticipated expiration: 1990-11-20

Abstract

A block-oriented random-access information storage system is disclosed which includes a cathode ray tube, horizontal and vertical address registers for binary information identifying rectangular coordinate data storage locations on a target in the cathode ray tube, and digital-to-analog converters coupling the contents of the horizontal and vertical address registers to respective horizontal and vertical beam deflection means for the cathode ray tube. The electron beam is deflected to desired information storage locations on the target with great precision by first deflecting the beam to a nearby relatively large indexing electrode on the target, successively modifying the addresses in the horizontal and vertical address registers in response to a feedback signal from the indexing electrode until the beam impinges on a reference point on the target, and then changing the addresses in the address registers the known amounts needed to cause the beam to impinge on the desired nearby data storage location.

Description

[ Nov. 20, 1973 ELECTRON BEAM ADDRESSABLE MEMORY SYSTEM Primary Examiner John Zazworsky Attorney-FL Christoffersen et a1.

[75] Inventors: Frank Jerome Marlowe, Kingston;

Charles Martin Wine, Princeton, 57 ABSTRACT both of N .J A block-oriented random-access information storage Asslgnefl R Corporatlon, New York system is disclosed which includes a cathode ray tube, [22] Filed: May 5, 1972 horizontal and vertical address registers for binary information identifying rectangular coordinatc data stor- PP N03 250,803 age locations on a target in the cathode ray tube, and digital-to-analog converters coupling the contents of [52] CL 328/124 313/68 R 313/92 BI the horizontal and vertical address registers to respec- 51 Int. Cl 6.1 1c 11/26 hmimma and beam deflecim means 581 Field of Search 328/123 124- the cathode ray tube- The electm beam is deflected 313/68 R to desired information storage locations on the target with great precision by first deflecting the beam to a [56] References Cited nearby relatively large indexing electrode on the target, successively modifying the addresses in the hori- UNITED fT PATENTS zontal and vertical address registers in response to a 3,588,584 6/1971 Tubinis 328/124 X f db k Signal from the indexing electrode il h :E' g g: beam impinges on a reference point on the target, and 3 460 099 8/1969 mini] I l l I l i I 3284123 X then changing the addresses in the address registers 364l555 2/1972 Griff-m n 328/123 X the known amounts needed to cause the beam to impinge on the desired nearby data storage location.

10 Claims, 4 Drawing Figures I2 VERTICAL DEFLECTION HORIZONTAL TIMING DEFLECTION 82 OM 54 DATA OUT D/A 60 70 SERlALLY i. i T a H. ADDR. REG. 58w T" J V 76 SOURCE DATA ADDR.

64 SOURCE INDEX ADDR.

PAIENTEUmvzo I975 VERTICAL DEFLECTION D/A -54 DATA OUT [)/A /60 m SERIALLY H. ADDR. REG. W52

66 1 g 78 H. AR|TH.UNTT /A A A A A SHIFT REG.

' 76 SOURCE DATA ADDR.

' SOURCE INDEX ADDR.

Fia. l

ELECTRON BEAM ADDRESSABLE MEMORY SYSTEM BACKGROUND OF THE INVENTION It is known that digital information can be stored on a target in a cathode ray tube, and can be randomly accessed by horizontal and vertical deflection of the cathode ray beam. However, the density of digital information storage on the target has been limited by a lack of precision in the operation of known beam deflection systems. Beam deflection is imprecise because of nonlinearities in the electro-magnetic deflection apparatus, and because of changes in reference voltages and component values with time and temperature.

SUMMARY OF THE INVENTION SlOl'l.

T l-IE DRAWING FIG. 1 is a diagram of a precisely addressable cathode ray tube memory constructed according to the teachings of the invention;

FIG. 2 is a diagram of a cathode ray tube target, used in the system of FIG. 1, and having a plurality of beam position indexing electrodes;

FIG. 3 is a diagramof a single beam position indexing electrode and nearby digital information storage locations, and

FIG. 4 is a diagram which will be referred to in describing the operation of the system of FIG. 1.

DESCRIPTION Referring now in greater detail toFIGS. l, 2, and 3, the memory system includes a cathode ray tube 10, which may be a silicon-target storage tube having a cathode 12, horizontal deflection coil 14, vertical deflection coil 16 and a silicon target 20. The tube 10 may be similar to a vidicon TV camera tube with the exception that the photosensitive target of the vidicon is replaced by an electrostatic storage target 20. The target 20 may consist of a heavily-doped, highly-conductive silicon wafer 22 (FIG. 2), with a layer of thermally grown silicon dioxide dielectric 24, and a polycrystaline layer of conducting silicon 26 that is photoexposed and etched through to the dielectric 24' to form indexing electrodes 30 each adjacent to an array 28 of electrostatic digital storage locations (FIG. 3). The elemental storage locations may have a center-to-center spacing of five microns, so that-a Z-inch diameter silicon wafer may provide for the storage of 10 bits. The silicon layer 26 is etched to provide a block oriented random access memory (BORAM) in which a plurality of indexing electrodes 30 are evenly distributed over the 1 surface of the target 20, and each indexing electrode 30 is associated with an immediately adjacent array or block 28 of digital storage locations. In operation, the

indexing electrode associated with a block of information storage locations is addressed, and then all the information bit storage locations in the associated block are serially addressed for the writing or reading of a block of information.

A switch 32 shown in FIG. 3 connects the top layer or grid 26 of conductive silicon to a positive voltage source when it is desired to write a 1" in a single storage location in array 28, and connects the layer 26 to the ground potential of the wafer 22 when it is desired to write a 0. The electrically-floating surface of silicon dioxide 24 at an element in array 28 impinged by the electron beam charges to a potential close to that to which the layer or grid 26 is connected, provided that the beam has a high enough energy to result in a secondary emission gain greater than unity. To read out the stored l or 0, switch 32 is connected to the ground potentialof wafer 22 when the electron beam impinges on the storage element, and, if a 1 was stored, a charging current to the wafer 22 is sensed as a voltage change across the resistor R and stray capacitance C, and is amplified by amplifier 34. Readout is destructive, and if the information is to be retained, it must be rewritten into the memory element. More detailed information on charge-storage vacuum devices is given in the-book Electronic Image Storage by B. Kazan and M. Knoll published in 1968 by Academic Press, Inc.

The indexing electrode 30 is formed by etching a gap 36 in the conductive silicon layer 26 so that the encircled portion of the layer is isolated from the remainder of the layer or grid 26. The electrode 30 is maintained at the ground potential of the wafer 22 by a unity gain follower amplifier 38, which may serve all of the indexing electrodes 30. When the electrode 30 is used for indexing purposes in a manner that will be described, the switch 32 is placed in a position to connect the silicon layer or grid 26 to the ground potential of wafer 22. Therefore, the electrode 30 and grid 26 are maintained at the same potential, to avoid charge differentials thereon that would adversely affect the path taken by the electron beam. 1

When the electron'beam impinges on the indexing electrode 30, a current flows through amplifier 38 and no signal appears at the input of amplifier 34. However, when the electron beam impinges on the area 40 of the layer or grid 26, a current flows from ground through resistor R to the electrodes on the grid 26. The voltage across resistor-R is an input signal for amplifier 34. The indexing electrode may be viewed as a pair of

beam indexing electrodes

30 and 40 separated by a vertical gap 36' and a horizontal gap 36". The indexing electrode 30 has a horizontal deflection reference edge at the vertical gap 36, and has a vertical deflection reference edge at the horizontal gap 36". A horizontal deflection reference point 42 in the vertical gap 36' and a vertical deflection reference point 44 in the horizontal gap 36" will be referred to in describing the operation of the system.

Returning to FIG. 1, the electron beam in the cathode ray tube 10 is deflected in the horizontal direction an amount determined by the digital contents of a horizontal address register 52. The digital information in register 52 is supplied to a digital-to-analog converter 54, which produces an analog signal that is amplified in amplifier 56 and applied through horizontal deflection coil 14 to deflect the beam an amount corresponding with the digital contents of register 52. Similarly, a vertical address register 58 contains digital information which is applied to vertical digital-to-analog converter 60 and from there, through a vertical deflection amplifier 62, to the vertical deflection coil 16. The position to which the electron beam is deflected on the target 20 of the tube 10 is determined by the contents of the horizontal and

vertical address registers

52 and 58.

The horizontal and

vertical address registers

52 and 58 may be supplied with horizontal and vertical addresses of any particular one of the indexing electrodes 30 from a digital source 64 of index electrode addresses. The horizontal and

vertical address registers

52 and 58 may also receive digital addresses from

arithmetic units

66 and 68, respectively. The arithmetic units are conventional units capable of performing addition or subtraction of binary numbers. A first operand input of arithmetic unit 66 is supplied from the horizontal address register 52 over line 70. A second operand input to the arithmetic unit 66 is supplied from a shift register 72 through a switch 74. An alternative second operand input may be supplied from a source 76 of relative data addresses, i.e., relative to a nearby indexing electrode.

The shift register 72 may, for example, include seven flip-flop or bistable multivibrator stages connected in cascade and representing the seven binary bits of a number. A 1" may be loaded into the first or highorder stage so that the binary number 1 000 000 2 64 is stored and available from the shift register. The 1 may be shifted to the next stage so that the binary number 100 000 2 32 is stored and available from the shift register. Each time the 1 is shifted one place to the right, the binary number is halved, so that the

numbers

16, 8, 4, 2 and 1 become successively available from the shift register. The shift register is constructed to supply a predetermined sequence of progressively halved binary numbers on command from the timing unit 82.

The shift register 72 is arranged to also similarly supply, through switch 74, a predetermined sequence of successively halved binary numbers to the vertical arithmetic unit 68. The

arithmetic units

66 and 68 operate to either add or subtract the output of the shift register from the contents of the address register in dependence on a signal supplied thereto through a switch 78 from amplifier 34. Amplifier 34 also has an output 80 for data read out serially from memory elements located on the target of storage tube 10. A timing unit 82 is provided to control the sequence of events in the operation of the system.

OPERATION In the operation of the system of FIGS. 1 and 3, binary data stored in an area 28 of bit storage locations on the target 20 can be read out after a procedure is followed which determines the numbers which must be added to, or subtracted from, nominal addresses in order to precisely address the desired locations. First the nominal horizontal and vertical addresses of the corresponding adjacent indexing electrode 30 are supplied from the source 64 to the horizontal and vertical address registers 52 and 58. The addresses of electrode 30 supplied to

registers

52 and 58 correspond with the addresses of the reference point 42 in FIG. 3. These digital addresses, after being converted into analog signals in

converters

54 and 60 and applied to the horizontal and vertical deflection yokes, cause a deflection of the electron beam so that it strikes the target somewhere in the area bounded by the dashed line 84. The beam normally does not impinge exactly on the reference point 42 because of unavoidable nonlinearities and imperfections in the beam deflection system. However, the electrode 30 is large enough so that a beam deflected toward reference point 42 is certain to impinge within the area of dashed line 84 on either the electrode 30 or the area 40 of the layer or grid 26.

If the beam lands on the electrode 30, some presently-unknown digital number must be added to the contents of the horizontal address register 52 to cause the beam to be horizontally deflected to the vertical gap 36' including the reference point 42. On the other hand, if the beam lands in the area 40, some number must be subtracted from the contents of the horizontal address register 52. The amplifier 34 has an arbitrary low output signal level if the beam lands on electrode 30, which is electrically isolated by the gap 36 and is electrostatically isolated from the amplifier 34 by the unit gain follower amplifier 38. On the other hand, the amplifier 34 has a relatively high output signal level if the beam lands on area 40. The output of amplifier 34 is applied through switch 78 to horizontal arithmetic unit 66. When the'signal is low, the arithmetic unit performs addition, and when the signal is high, the unit performs subtraction.

A specific example of how the horizontal address is changed to deflect the beam to the vertical gap 36 will be described with references to the chart of FIG. 4 and to Table I. It is assumed that the horizontal address register contains seven binary digits, and that the nominal horizontal address of point 42 of the index electrode is 0 000 000 in binary digits, and that this binary number in the horizontal address register actually causes the beam to land on the index electrode 30 at point A. Point A is 15 units to the left of gap 36. Since the beam impinges on electrode 30, the output of amplifier 34 is low, and this signal applied through switch 78 to horizontal arithmetic unit 66 causes the unit to perform addition. The contents 0 000 000 of horizontal address register 52 is applied over line 70 to arithmetic unit 66, is added to the output 2 l 000 000 of shift register 72, and the sum is placed in horizontal address register 52. Now, the beam is deflected from point A to the right an amount equal to 64 units to a point B in the area 40 (FIG. 4).

Now the output of amplifier 34 is high, so that arithmetic unit 66 subtracts the new output 2 0 000 TABLE I CONTENTS OF HORIZONTAL ADDRESS REGISTER DIGITAL 2- BINARY A 0 0 000 000 ADD +64 2 +1 000 000 B 64 1 000 000 SUBTR 32 2= -0 100 000 c 32 0 100 000 SUBTR l6 2* -0 010 000 0 l6 0 010 000 SUBTR s 2 -0 001 000 E s 0 001 000 ADD +4 2 +0 000 100 F 12 0 001 100 ADD +0 000 010 G 14 0 001 ADD 71- +0 000 001 11 1s r 0 ()(ll lll of shift register 72 from the contents of horizontal address register 52. The beam is then deflected from point B to the left to point C. The procedure is repeated as shown in Table I, each time adding or subtracting an amount from shift register 72 which is one half of the value previously used. Finally, when the last cycle of operation is completed, the beam is at horizontal location H at the vertical gap 36', and the horizontal address register 52 contains the binary number 0 001 ll 1 which is necessary to precisely deflect the beam in the horizontal direction to the vertical gap 36'.

A precedure is then followed to change the contents of the vertical address register an amount necessary to precisely deflect the beam in the vertical direction to the horizontal gap 36". This is accomplished by first adding an amount to the horizontal address register 52 to cause the beam to move to the right from the vertical gap 36 to a position in the area 40 on a vertical line with the point 44 in horizontal gap 36". Since the beam then impinges on area 40 of the layer or grid 26, the amplifier 34 produces a subtract output which is applied through switch 78 to vertical arithmetic unit 68 to cause the unit to subtract the first high order number in shift register 72 from the contents of the vertical address register. The difference transferred to the vertical address register causes the beam to be deflected downwardly so that it impinges on the indexing electrode 30. This procedure is repeated, in the same manner as has been described in connection with horizontal deflection, until the contents of the vertical address register causes the beam to impinge at the point 44 in the horizontal gap 36".

The position at which the beam impinges on the target is now known to be precisely the point 44, and the horizontal and vertical address registers contain the numbers necessary to cause deflection of the beam to point 44. The horizontal and vertical displacements of the adjacent block of information storage element locations 28 relative to the point 44 are known. Therefore, the displacements can be added orsubtracted from the contents of the horizontal and vertical address registers to cause the beam to be precisely deflected to a starting point for scanning through the individual information storage locations of the block 28. The displacements are supplied from the source 76 of data addresses to the

arithmetic units

66 and 68, where they are added to or subtracted from the contents of the address registers. The sums and differences are then transferred to the address registers 52 and 58 to deflect the beam to the desired information storage location. If read-out of ina formation is being performed, the resulting l or 0" information signal is sensed by amplifier 34 and supplied to utilization apparatus (not shown) in serial fashion over lead 80.

The degree of precision with which the beam can be deflected to a given information storage location increases in accordance with the number of times the beam is shifted in halved increments toward each of the gaps 36' and 36", the closeness on the target 20 of the given information storage location to the associated indexing points 42 and 44, and the shortness of the time elapsed since the indexing operation was performed.

When it is desired to access information storage locations of another different block of information at another place on the target 20, the beam is deflected to the indexing electrode near the newly desired block of information storage locations and the above-described procedure is gone through to determine the contents of the horizontal and vertical address registers 52 and 58 needed to precisely deflect the beam at the new reference point on the indexing electrode. Then, known displacement amounts may be added to or subtracted from the address registers, as has been described, to precisely deflect the beam to desired nearby information storage locations.

What is claimed is:

1. In a cathode ray tube memory system, beam deflection means and an information storage target having a plurality of groups of information storage locations each at precise known displacements from a reference point on an adjacent associated beam indexing electrode, and deflection correction feedback means operative when the beam is deflected to a given indexing electrode to progressively alter the deflection of the beam toward said reference point thereon, whereby the beam can then be precisely deflected a displacement amount to a desired nearby information storage location.

2. A system as defined in claim 1 wherein said beam deflection means includes horizontal and vertical address registers each coupled through respective digitalto-analog converters to respective horizontal and vertical deflection coils.

3. A system as defined in claim 2 wherein each said beam indexing electrode includes a horizontal deflection reference edge and a vertical deflection reference edge.

4. A system as defined in claim 3 wherein said deflection correction feedback means includes means to address the beam to a given indexing electrode and to sense which side of the horizontal deflection reference edge is impinged by the beam.

5. A system as defined in claim 4 wherein said deflection correction feedback means includes means to shift the beam deflection horizontally toward the horizontal deflection reference edge of the indexing electrode, and to repeat the process a predetermined number of times.

6 A system as defined in claim 5 wherein said means to shift the beam deflection includes means to make each successive shift of the beam toward the horizontal deflection reference edge in an amount equal to onehalf of the previous amount.

7. A system as defined in claim 6 wherein said means to make said successive shifts in horizontal deflection includes means to add or subtract a binary l to or from successively lowered ordered bit positions of the address in said horizontal address register.

8. A system as defined in claim 7, wherein said deflection correction feedback means also includes similar means to shift the beam deflection vertically toward said vertical deflection reference edge.

9. A system as defined in claim 8 and, in addition, means to change the contents of said horizontal and vertical address registers in amounts equal to displacements of nearby associated information storage locations.

10. In a random-access information storage system, a cathodevray tube having horizontal and vertical beam deflection means and a target, horizontal and vertical address registers for binary information identifying rectangular coordinate locations on said target, and digital-to-analog converters coupling said horizontal and vertical address registers to respective horizontal and vertical beam deflection means,

a pair of beam indexing electrodes on said target separated by a vertical gap and a horizontal gap, said pair of electrodes having a size such that when its tracted, whereby the horizontal address register is made to contain the horizontal address which re sults in deflection of the beam to the vertical gap between the electrodes,

means coupled to said electrodes to similarly make address is supplied to said address registers, the beam is certain to impinge on one of the electrodes of the pair,

means coupled to said electrodes to sense which electhe vertical address register contain the vertical address which results in deflection of the beam to the horizontal gap between the electrodes, and

means to add or subtract the address of a desired data trode is impinged by the beam, to add or subtract 10 a given amount to or from the address in the h0ri storage locat on relative to said gaps between sa d zontal register to cause the beam deflection to shift P of mdexmg electrodes m the cmtcms sald i h h i l di i toward the other l horizontal and vertical address registers, to cause trode, and to repeat the roce a given n b f the electron beam to be deflected with precision to times, each time adding or subtracting an amount the desired data storage location. equal to half the amount previously added or sub- UNITED STATES PATENT OFFICE CERTIFICATE OF CORRECTION Patent No. 3,774,116 Dat d November 20, 1973 lnventofls) FRANK JEROME MARLOWE, ET. AL.

It is certified that error appears in the above-identified patent and that said Letters Patent are hereby corrected as shown below On the cover sheet after the Abstract, insert The f invention herein described was made in the course of, or 1 under a contract or subcontract thereunder, with the Department Column 2, line 17, ""1 or 0" should; .7

of the Air Force. read "l" or "0" line 19, "1" should read "l" Column 3,

lines

32 and 34,, "'1" should read "l" Signed and sealed this 6th day of August 1974.

. (SEAL) Attest:

MCCOY M. GIBSON, JR; Attesting Officer C. MARSHALL DANN Commissioner of Patents USCOMM-DC 603764 69 FORM PC4050 (10-69) w u.s. aovznnuru'r rmmmc OFFICE nu o-us-au.

Claims

2. A system as defined in claim 1 wherein said beam deflection means includes horizontal and vertical address registers each coupled through respective digital-to-analog converters to respective horizontal and vertical deflection coils.

5. A system as defined in claim 4 wherein said deflection correction feedback means includes means to shift the beam deflection horizontally toward the horizontal deflection reference edge of the indexing electrode, and to repeat the process a predetermined number of times. 6 A system as defined in claim 5 wherein said means to shift the beam deflection includes means to make each successive shift of the beam toward the horizontal deflection reference edge in an amount equal to one-half of the previous amount.

7. A system as defined in claim 6 wherein said means to make said successive shifts in horizontal deflection includes means to add or subtract a binary ''''1'''' to or from successively lowered ordered bit positions of the address in said horizontal address register.

9. A system as defined in claim 8, and, in addition, means to change the contents of said horizontal and vertical address registers in amounts equal to displacements of nearby associated information storage locations.

10. In a random-access information storage system, a cathode ray tube having horizontal and vertical beam deflection means and a target, horizontal and vertical address registers for binary information identifying rectangular coordinate locations on said target, and digital-to-analog converters coupling said horizontal and vertical address registers to respective horizontal and vertical beam deflection means, a pair of beam indexing electrodes on said target separated by a vertical gap and a horizontal gap, said pair of electrodes having a size such that when its address is supplied to said address registers, the beam is certain to impinge on one of the electrodes of the pair, means coupled to said electrodes to sense which electrode is impinged by the beam, to add or subtract a given amount to or from the address in the horizontal register to cause the beam deflection to shift in the horizontal direction toward the other electrode, and to repeat the process a given number of times, each time adding or subtracting an amount equal to half the amount previously added or subtracted, whereby the horizontal address register is made to contain the horizontal address which results in deflection of the beam to the vertical gap between the electrodes, means coupled to said electrodes to similarly make the vertical address register contain the vertical address which results in deflection of the beam to the horizontal gap between the electrodes, and means to add or subtract the address of a desired data storage location relative to said gaps between said pair of indexing electrodes, to the contents of said horizontal and vertical address registers, to cause the electron beam to be deflected with precision to the desired data storage location.