US3924156A

US3924156A - Method and system for correcting an aberration of a beam of charged particles

Info

Publication number: US3924156A
Application number: US483266A
Authority: US
Inventors: Samuel K Doran; Merlyn H Perkins
Original assignee: International Business Machines Corp
Current assignee: International Business Machines Corp
Priority date: 1974-06-26
Filing date: 1974-06-26
Publication date: 1975-12-02
Anticipated expiration: 1992-12-02
Also published as: NL7507428A; JPS5420396B2; SE7507109L; DE2521591B2; DE2521591C3; JPS515694A; NL178732B; DE2521591A1; IT1038110B; AU8138075A; GB1505363A; FR2276625A1; NL178732C; FR2276625B1; SE398415B

Abstract

A beam of charged particles is deflected in a closed path such as a square, for example, over a cross wire grid at a constant velocity by an X Y deflection system. A small high frequency jitter is added at both axes of deflection to cause oscillation of the beam at 45* to the X and Y axes. From the time that the leading edge of the oscillating beam passes over the wire until the trailing edge of the beam passes over the wire, an envelope of the oscillations produced by the jitter is obtained. A second envelope is obtained when the leading edge of the beam exits from being over the wire until the trailing edge of the beam ceases to be over the wire. Thus, a pair of envelopes is produced as the beam passes over each wire of the grid. The number of pulses exceeding ten per cent of the peak voltage in the eight envelopes produced by the beam completing a cycle in its closed path around the grid are counted and compared with those counted during the previous cycle of the beam moving in its closed path over the grid. As the number of pulses decreases, the quality of the focus of the beam increases so that correction signals are applied to the focus coil in accordance with whether the number of pulses is increasing or decreasing.

Description

United States Patent 11 1 1 1 3,924,156 Doran et al. 1 Dec. 2, 1975 METHOD AND SYSTEM FOR CORRECTING [57] ABSTRACT AN ABERRATION OF A BEAM OF A beam of charged particles is deflected in a closed CHARGED PARTICLES path such as a square. for example, over a cross wire Inventors: Samuel K. Doran, Wappingers Falls; grid at a constant velocity by an X Y deflection sys- Merlyn H. Perkins, Hopewell tem. A small high frequency jitter is added at both Junction, both of NY. axes of deflection to cause oscillation of the beam at 45 to the X and Y axes. From the time that the leading edge of the oscillating beam passes over the wire until the trailing edge of the beam passes over the [22] Filed: June 26, 1974 wire, an envelope of the oscillations produced by the jitter is obtained. A second envelope is obtained when [21] Appl NO" 483,266 the leading edge of the beam exits from being over the wire until the trailing edge of the beam ceases to be [73] Assignee: International Business Machines Corporation, Armonk, NY.

[52] US. Cl. i. 315/382 over the wire. Thus, a pair of envelopes is produced as [51] Int. Cl. H01J 29/70 the beam passes over each wire of the grid. The num- [58] Field of Search .1 315/382, 30, 31, 364 ber of pulses exceeding ten per cent of the peak voltage in the eight envelopes produced by the beam com- [56] References Cited pleting a cycle in its closed path around the grid are UNITED STATES P TENTS counted and compared with those counted during the 3,588,586 6/1971 Yanaka 315/379 Previous cycle of the beam moving in its closed Pam 3,753,035 8/1973 Verth 315/370 ever the grid- AS the "umber of PulseS deereeeee the quality of the focus of the beam increases so that corp Examiner Maynard Wilbur rection signals are applied to the focus coil in accor- Assistam potenza dance with whether the number of pulses is increasing Attorney, Agent, or FirmFrank C. Leach, Jr.; or decreasing- Theodore E. Galanthay 28 Claims, 16 Drawing Figures 772- OSCILLOSCOPE CURRENT 10 1311111111155 75 7e 89 97 9a 100 111 11? 42 VOLTAGE 1111111111111 f 7 f 2% CONVERTER AMPLIFIER 1u7o11111c l 11515011111 UP 1101111 FOCUS 11111 11111 0011111110 couurmc c011711o1 71 L12 150 00111101 11E111s 99/7 115115 T 11111 couunzn 15s 1151mm;

7 1511s fm H I III voumzm 154 152 81111110 '6 7 D t 0 155 T 10m 153 5 193 US. Patent Dec. 2, 1975 Sheet 1 of 6 3,924,156

9 1 Dn LL F MM P N M II D w c 8 7 1 n0 NO G WWW u m WU LN AN 00 U CC A US. Patent Dec. 2, 1975 Sheet 2 of6 3,924 156 [m H4 [175 180 +5V181 182 8-12BIT SQUARER COMPARATOR COUNTER g m (L COMPARATOR INVERTER 93 a BIT DAC REF /185 FEG. 13 j m +5v iv 191 192 186 185 187 f NAND INVERTER STATUS A COMPARATOR DECODER K INVERTER L 189 NUMBER OF PULSES FOCUS COIL CURRENT Patant Dec. 2, 1975 Sheet 4 of 6 MULTIPLIER T k POSITIVE VOLTAGE if INVERTER COMPARATOR 97 L 9o 92 95 +01v NAND W 95 7 96 7 94 NEGATIVE SINGLE VOLTAGE SHOT INVERTER COMPARATOR US. Patant Dec. 2, 1975 Sheet of6 324j5 101 98 1 FEG. 11

K 0 111 101;; FLIP 12 BIT UP 1 I C FLOP 0011111 COUNTER J 6 104 L F- 115 SINGLE 3|NGLE $1101 1 119 W 1 INVERTER 11a 5 f 1115 111 156 S FLIP Q I L 155 5STATE SINGLE A COUNTER $1101 151 121 g 1 11111110 127 126 m FLIP I 8 BIT UP FLOP 124 DOWN COUNTER 129 8BIT 01c -1 128 METHOD AND SYSTEM FOR CORRECTING AN ABERRATION OF A BEAM OF CHARGED PARTICLES In U.S. Pat. No. 3,644,700 to Kruppa et al, there is shown an apparatus for stepping a square-shaped beam of charged particles in a substantially raster fashion over a predetermined area. An improved apparatus for stepping a square-shaped beam of charged particles over a predetermined area is shown and described in the copending patent application of Michel S. Michail et al for Method And Apparatus For Positioning A Beam Of Charged Particles, Ser. No. 437,585, filed Jan. 28, 1974, now U.S. Pat. No. 3,900,736 and assigned to the same assignee as the assignee of this application.

In the aforesaid Kruppa et al patent and the aforesaid Michail et al application, it is necessary for the beam to be properly focused to obtain the desired precise application of the beam in various portions of the predetermined area. As described in the aforesaid Kruppa et al patent and the aforesaid Michail et al application, the beam is moved through A, B, and C cycles.

The present invention provides a method and system for measuring the quality of the focus of the beam of charged particles at various intervals and automatically correcting the focus. These measurements and the corrections occur during certain of the C cycles.

The present invention also is capable of ascertaining the illumination of the beam of charged particles, the quality of the focus of the beam of charged particles, and the astigmatism of the beam of charged particles so that manual correction of these aberrations can occur. These corrections would occur when the electron gun is set up to produce the beam.

The present invention measures the quality of the focus of the beam by directing the beam in a predetermined closed path across a target which will interrupt the beam of charged particles whenever the beam is completely engaging the target. Since the beam is deflected in orthogonal directions, the quality of the focus of the beam in both of its deflection directions is ascertained.

The present invention preferably accomplishes this by utilizing a cross wire grid as the target and moving the beam in a square over the grid at a constant velocity. Thus, the beam will pass over a portion of the wire grid extending in the X direction, for example, next over a portion of the wire grid extending in the Y direction, then over another portion of the wire grid extending in the X direction, and finally over another portion of the wire grid extending in the Y direction during each cycle of movement of the beam over the grid.

By applying a high frequency jitter to oscillate the beam in a single direction at an angle to each of X and Y axes, the oscillations produced by the jitter form two envelopes as the beam passes over a wire. A first envelope of the oscillations occurs from the time that the leading edge of the beam engages the wire until the trailing edge of the beam is disposed over the wire. A second envelope occurs when the leading edge of the beam exits from the wire until the trailing edge of the beam exits from the wire.

Each of the envelopes is indicative of the current density of the beam in the direction in which the beam is being deflected. The envelopes represents a voltage proportional to the jitter component of the current of the current collector. Since the jitter is sweeping a constant two percent slice of beam area onto and off of the wire, the peak amplitude of the envelope voltage is proportional to the average current density of the two percent slice. Thus, each of the two envelopes produced by the beam crossing a wire extending in the Y direction when the beam is deflected in the X direction is a profile of the current density distribution of the beam in the X direction.

If the beam were free of aberrations, each of these 0 envelopes would be square shaped. However, when the beam is out of focus, the edge definition of the beam is smeared so that the current density profile also is smeared. Additionally, if the beam is astigmatic, the envelopes or profiles of the current density in one direction may be sharp while the profiles in the other direction may be smeared or vice versa. Furthermore, if the beam is not properly aligned so that it does not have uniform illumination, the top of the current density profile will not be flat but will be slanted.

Therefore, with the profile of the current density not being square shaped when the beam has an aberration with the beam being square shaped, the quality of the current density of the beam can be ascertained by the use of the envelopes. The envelopes indicate whether the beam is properly focused, has no astigmatism, and is illuminated properly.

Accordingly, as the focus becomes worse, the sides of the profile of theenvelope tend to become more inclined or curved. Since the area underneath the envelope must always be the same irrespective of whether the beam has aberrations or not, an indication of the quality of the focus of the beam in comparison with the focus of the beam during the prior scan can be ascertained through determining whether the envelope sides are becoming more straight to indicate increasing quality of the focus of the beam or more inclined to indicate decreasing quality of the focus of the beam. If the beam has astigmatism, this will affect the focus quality.

Similarly, the illumination of the beam also can be ascertained through determining whether the envelope has a flat top or not. If the top ceases to be flat, then the area near the top of the envelope tends to become smaller to indicate poor illumination.

Accordingly, the quality of the focus of a beam can be compared to that during the previous pass of the beam over the wire by counting the number of times that the voltage within the. envelope exceeds a certain limit, which is preferably ten percent of the peak voltage produced by the jitter. At ten percent of the peak voltage produced by the jitter, the inclination of the sides of the profile of the envelope is readily comparable with the inclination of the sides of the profiles of the envelope produced during the prior pass of the beam over the wire. Of course, other suitable percentages, except fifty percent, of the peak voltage could be employed.

When initially starting to measure the quality of the focus, the comparison is with the information from the satisfactory focus of the beam from the previous C cycle in which focusing occurred. It should be understood that other operations concerning the quality of the beam occur during other C cycles so that a number of A, B, and C cycles normally occur between focus measurements and noise could have eliminated the prior information. Thus, the signal produced by the first cycle of movement of the beam over the target could produce a correction signal in the wrong direction.

If the number of times the voltage exceeds the selected limit increases beyond the number from the prior cycle, than the direction of change of the magnitude of current to the focus coil is reversed. If the count of the pulses is less than that produced by the prior pass of the beam over the grid, the direction of change of the magnitude of current to the focus coil is continued.

There are a total of sixteen complete cycles of the beam along its closed path over the cross wire grid during each C cycle in which automatic focus correction is occurring. This is sufficient to satisfactorily focus the beam. Whenever the direction of change of the magnitude of the current is reversed three consecutive times when the correction signal is supplied to the focus coil, the best focus value has been obtained.

To ascertain whether the illumination of the beam is satisfactory, the counting of the number of times that the voltages within the envelope exceeds another voltage limit, which is preferably ninety percent of the peak voltage produced by the jitter, enables ascertainment of the quality of the illumination of the beam. Thus, when the number of times that the voltage within the envelope exceeds the selected limit increases from the prior pass of the beam over the grid, this is an indication that the illumination of the beam is being improved. This is because an increase in the number of times that the voltage exceeds ninety percent of the maximum voltage produced by the jitter is an indication of an increase in area in the top portion of the profile so that the top of the profile of the envelope is becoming more flat.

An object of this invention is to provide a method and system for correcting an aberration of a beam of charged particles.

Another object of this invention is to provide a method and system for focusing a beam of charged particles.

A further object of this invention is to measure the quality of the focus of the beam of charged particles and adjust the focus in accordance with the quality.

Still another object of this invention is to automatically correct the focus of a beam of charged particles at selected intervals of time.

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

In the drawings:

FIG. 1 is a schematic view showing a beam of charged particles and the apparatus for controlling the beam.

FIG. 2 is a schematic top plan view ofa portion of the wire grid over which the beam of charged particles moves during focusing.

FIG. 3 is a schematic plan view of the beam showing the beam in its position when not moved by the jitter frequency and the position to which it is moved by the jitter frequency.

FIG. 4 is a schematic view showing a pulse produced by the beam passing over a portion of the wire grid with the oscillations from the jitter frequency appearing on the pulse as the beam enters and leaves the portion of the wire grid.

FIG. 5 is a view showing two different shaped envelopes produced by the oscillations from the high frequency jitter resulting from the beam entering or leav- 4 ing a portion of the wire grid with one of the envelopes representing a good focus condition of the beam and the other of the envelopes representing a poor focus condition of the beam.

FIG. 6 is a view of the shape of an envelope produced by the oscillations from the high frequency jitter resulting from the beam entering or leaving a portion of the wire grid when illumination of the beam is poor.

FIG. 7 is a block diagram showing the relationship of circuitry used for correcting aberrations of the beam.

FIG. 8 is a schematic wiring diagram of the bandpass filter and amplifier of the circuitry of FIG. 7.

FIG. 9 is a schematic wiring diagram of the automatic gain control of the circuitry of FIG. 7.

FIG. 10 is a block diagram of the detecting and counting means of the circuitry of FIG. 7.

FIG. 11 is a block diagram of an up down counting means of the circuitry of FIG. 7.

FIG. 12 is a block diagram of the focus control unit of the circuitry of FIG. 7.

FIG. 13 is a block diagram of an illumination detecting means of the circuitry of FIG. 7.

FIG. 14 is a block diagram of the apparatus for producing the jitter frequency.

FIG. 15 is a timing diagram showing the relationship of various signals during focusing of the beam.

FIG. 16 is a schematic diagram showing the relationship of the focus coil current and the number of pulses counted during cycles of the beam passing over the wire grid.

Referring to the drawings and particularly FIG. 1, there is shown an electron gun 10 for producing a beam 11 of charged particles in the well-known manner. As shown and described in the aforesaid Michail et al application, the electron beam 11 is passed through an aperture 12 in a plate 14 to shape the beam 11. The beam 11 is preferably square shaped and has a size equal to the minimum line width of the' pattern that is to be formed.

The beam 11 passes between a pair of blanking plates 15, which determine when the beam 11 is applied to the material and when the beam. 11 is blanked. The blanking plates 15 are controlled by circuits of an analog unit 17, which has a column control unit 16 connected thereto.

The analog unit 17 is controlled by a digital control unit 18 in the manner more particularly shown and described in the copending patent application of Phillip M. Ryan for Method And Apparatus For Controlling Movable Means Such As An Electron Beam, Ser. No. 398,734, filed Sept. 19, 1973, now U.S. Pat. No. 3,866,013, and assigned to the same assignee as the assignee of this application. The digital control unit 18 is connected to a computer 19, which is preferably an IBM 370 computer.

The beam 11 then passes through a circular aperture 20 in a plate 21. This controls the beam 11 so that only the charged particles passing through the centers of the lenses (not shown) are used so that a square-shaped spot without any distortion is produced.

The beam 11 is next directed through

stigmator coils

21A and 21B and then through a focus coil 22. The stigmator coils 21A and 21B and the focus coil 22 are connected to the column control unit 16.

After passing through the focus coil 22, the beam is directed through magnetic deflection coils 23, 24, 25, and 26. The magnetic deflection coils 23 and 24 control the deflection of the beam 11 in a horizontal or X direction while the magnetic deflection coils 25 and 26 control the deflection of the beam 11 in a vertical or Y direction. Accordingly, the coils 23-26 cooperate to move the beam 11 in a horizontal scan by appropriately deflecting the beam 11.

While the beam 11 could be moved in a substantially raster fashion as shown and described in the aforesaid Kruppa et al patent, it is preferably moved in a back and forth scan so that the beam 11 moves in opposite directions along adjacent lines as shown and described in the aforesaid Ryan and Michail et a]. applications. Thus, the negative bucking sawtooth of the type shown in FIG. 3b of the aforesaid Kruppa et al patent is supplied to the

coils

23 and 24 during forward scan while a positive bucking sawtooth, which is of opposite polarity to the sawtooth shown in FIG. 3b of the aforesaid Kruppa et al patent, is supplied to the

coils

23 and 24 during the backward scan.

As shown and described in the aforesaid Michail et al application, the beam 11 then passes between a first set of

electrostatic deflection plates

27, 28, 29, and 30. The

electrostatic deflection plates

27 and 28 cooperate to deflect the beam in a horizontal or X direction while the

electrostatic deflection plates

29 and 30 cooperate to move the beam 11 in the vertical or Y direction. The plates 27-30 are employed to provide any desired offset of the beam 11 at each of the predetermined positions or spots to which it is moved. In the aforesaid Kruppa et al patent, the plates 27-30 corrected for linearity, but these correction signals are supplied to the coils 23-26 in the aforesaid Michail et al application and in this application.

After passing between the electrostatic deflection plates 27-30, the beam 11 then passes between a second set of

electrostatic deflection plates

31, 32, 33, and 34. The

electrostatic deflection plates

31 and 32 cooperate to deflect the beam 11 in the horizontal or X direction while the

electrostatic deflection plates

33 and 34 cooperate to deflect the beam 11 in the vertical or y direction. The

plates

31 and 32 deflect the beam 11 in the X direction and the

plates

33 and 34 deflect the beam 11 in the Y direction from each of the predetermined positions to which it is moved in accordance with its predetermined pattern so that the beam 11 is applied to its actual position based on the deviation of the area from its designed position, both shape and location, in which the beam 11 is to write as more particularly shown and described in the aforesaid Michail et al application.

The beam 11 is then applied to a target, which is supported on a table 35 and can be a semiconductor wafer, for example. The table 35 is movable in the X and Y directions as more particularly shown and described in the aforesaid Kruppa et al patent.

The beam 11 is moved through A, B, and C cycles as shown and described in the aforesaid Kruppa et al patent and the aforesaid Michail et al application. The present invention is particularly concerned with supplying signals to automatically correct the focus of the beam 11 during some of the C cycles.

The focus of the beam 11 is controlled by the focus coil 22. A current flows through the focus coil 22 to regulate the focus of the beam 11.

When a program in the computer 19 determines that it is time for the focus of the beam 11 to be automatically corrected, if necessary, during a C cycle, the table 35 is moved to bring a cross wire grid 41 (see FIG. 2), which is in the same plane as the semiconductor wafer 6 supported on the table 35, beneath the undeflected beam 11. A current collector 42 (see FIG. .7), which is preferably a photodiode, is disposed beneath the grid 41, which is preferably formed of. tungsten and comprises a plurality of orthogonal wires extending in the X and Y directions.

After the grid 41 has been properly positioned beneath the beam 11, the digital control unit 18 causes a FOCUS SERVO signal, which is a positive pulse as shown in FIG. 15, to be supplied from the analog unit 17. The analog unit 17 suplies the FOCUS SERVO signal over a line 43 (see FIG. 14) and through an optical isolator 44 to an OR gate 45. The optical isolator 44 has a return line 45 to ground. One-suitable example of the otical isolator 44 is sold by Hewlett-Packard as model 5082-4350.

The OR gate 45 has its output connected as an input to a NAND gate 46. The other input to the NAND gate 46 is from a 300 KHZ oscillator 47.

When the FOCUS SERVO pluse is supplied to the optical isolator 44, the NAND gate 46 allows the oscillator 47 to supply its frequency through a dither control 48, which controls the amplitude of the high frequency jitter supplied through a line 49 to the analog unit 17 to each of the X'and Y

electrostatic deflection plates

31, 32, 33, and 34 (see FIG. 1 The jitter is preferably two percent of the width of the beam 11. Since it is applied to both the X and Y electrostatic deflection plates 31-34 in phase, the jitter is at an angle of 45 with respect to the X and Y axes so that the beam 11 oscillates in a single direction. Thus, in FIG. 3, the beam 11 is shown in solid lines in its position when jitter is not applied and in phantom lines when jitter is applied.

The current, which is produced from the current collector 42, has a jitter component due to the high frequency jitter. This jitter component is equal to the product of the current density of the beam 11 and the area of the jitter of the beam 11.

When the FOCUS SERVO signal is supplied to the optical isolator 44, the beam 11 is caused to move along a square-shaped path 60 (see FIG. 2) so as to cross

portions

61, 62, 63, and 64 of the wire grid 41 in this order. Thus, the beam 11 is moved in a predetermined closed path around the wire grid 41.

The beam 11 is initially moved in a +X direction as indicated by an arrow 65 so as to cross the vertical portion 61 of the wire grid 41 during this scan. This initial movement of the beam 11 is caused by a +X SCAN signal being supplied from the analog unit 17 due to the digital control unit 18 supplying a digital signal thereto. This +X SCAN pulse starts at the same time that the FOCUS SERVO signal begins as shown in FIG. 15.

Thus, as shown in thetiming diagram of FIG. 15, an increasing X magnetic deflection voltage is initially supplied to the X magnetic deflection coils 23 and 24 to move the beam 11 in the +X direction. At the end of the increase in the X magnetic deflection voltage, a decreasing Y magnetic deflection voltage is supplied to the Y magnetic deflection coils 25 and 26 so that the beam 11 moves in a Y direction as indicated by an arrow 66 in FIG.- 2 to pass over the horizontal portion 62 of the wire grid 41. When the Y magnetic deflection voltageceases to fall to stop movement in the Y direction, a decreasing X magnetic deflection voltage is supplied to the X magnetic deflection coils 23 and 24 to move the beam 11 in a X direction as indicated by an arrow 67 inFIG. 2 to pass over the vertical portion 63 of the wire grid 41. When the X magnetic deflection voltage stops falling so that movement in the X direction stops, an increasing Y magnetic deflection voltage is supplied to the Y magnetic deflection coils 25 and 26 to move the beam 11 in a +Y direction as indicated by an arrow 68 in FIG. 2 to pass over the horizontal portion 64 of the wire grid 41. As a result, the beam 11 is moved along the square-shaped predetermined path 60.

As shown in FIG. 4, the output of the current collector 42 is at a maximum when the beam 1 1 is completely off the wire grid 41, decreases from its maximum to its minimum as the beam 11 moves over the wire grid 41, is at a minimum when the beam 11 is completely disposed above one of the portions 61-64 of the wire grid 41, and increases from its minimum to its maximum as the beam 11 moves off of one of the portions 61-64 of the wire grid 41.

During the time that the beam 11 is moving onto or off of one of the portions 61-64 of the wire grid 41, the current at the current collector 42 includes the oscillations produced by a high frequency jitter from the dither control 48. The high frequency jitter produced by the beam moving onto one of the portions 61-64 of the wire grid 41 is indicated on the pulse of FIG. 4 by 68A and the high frequency jitter produced by the beam 11 moving off of one of the portions 61-64 of the wire grid 41 is indicated on the pulse of FIG. 4 by 68B. This high frequency jitter produces an envelope which is indicative of the quality of the focus of the beam 11.

Two

different envelopes

69 and 70 of the high frequency jitter are shown in FIG. 5. Each of the

envelopes

69 and 70 has the same area thereunder since the beam 11 produces the same current.

If the beam 11 were free of aberrations, it would have a square-shaped envelope since the current density would be the same in all parts of the beam 11. However, when the beam 11 goes out of focus, the edge definitions thereof become smeared so that the current density profile, which is represented by each of the

envelopes

69 and 70, for example, of the beam 11 becomes smeared. If the beam 11 is astigmatic, the envelope produced by the beam 11 moving in one of the X and Y scans may be sharp but movement of the beam 1 1 in the other scan direction would produce a smeared envelope.

It should be understood that each of the

envelopes

69 and 70 is formed by voltages which are directly related to the jitter component of the current from the current collector 42. This is because the output of the current collector 42 is supplied to a current to voltage converter 71 (see FIG. 7), which can be an operational amplifier, for example, and the

envelopes

69 and 70 represent the output of the voltage converter 71, which supplies an inverted output.

By comparing the number of times that the voltage reaches some level in each of the

envelopes

69 and 70, a determination can be made as to which of the two

envelopes

69 and 70 produces the best focus. Thus, for example, if a count is made each time that the voltage in the

envelope

69 or 70 exceeds ten percent of the peak voltage, a comparison can be made as to which of the two

envelopes

69 and 70 is producing the better focus since the envelope having the lowest number of times that the voltage exceeds ten percent of the maximum is a better envelope for focus.

A total of eight envelopes is produced during each movement of the beam 11 along the entire path 60 since an envelope is produced each time that the beam 11 moves on or off of the wire grid 41. By comparing the total number of times that the voltage exceeds ten percent of the peak voltage in two consecutive movements of the beam 1 1 around the wire grid 41 along the path 60, a comparison between the two cycles of movement of the beam 11 along the entire path 60 is obtained so that it can be ascertained whether a correction signal to the focus coil 22 in the time between the two cycles improved or lowered the quality of the focus of the beam 11.

When the beam 11 starts to cross the vertical portion 61 of the wire grid 41, for example, the current collector 42 produces a varying current with the dither thereon. The signal from the current collector 42 is supplied to the current to voltage converter 71, which supplies an inverted output voltage to a bandpass filter and amplifier 72 (see FIG. 7). The bandpass filter and amplifier 72 allows only the envelope voltage, which is produced when the beam 11 is crossingone of the portions 61-64 of the wire grid 41, to pass therethrough to form the envelopes.

As shown in FIG. 8, the bandpass filter and amplifier 72 includes a pair of

differential amplifiers

73 and 74 cooperating to produce a Q of ten and a gain of 100 at 300 KHz. The output of the bandpass filter and amplifier 72 is supplied over a line 75 to an automatic gain control 76 (see FIG. 7). The output of the bandpass filter and amplifier 72 also is supplied over a line 77 to an oscilloscope 78 located within the analog unit 17. The oscilloscope 78 enables any of the eight envelopes produced by movement of the beam 11 over the wire grid 41 to be viewed.

The automatic gain control 76 maintains the amplitude of the envelope at fourteen volts peak to peak to compensate for slow drifts in the brightness of the beam 1 1. Thus, the maximum peak to peak voltage produced on an output line 79 of the automatic gain control 76 is fourteen volts.

The focus envelope wave form can be described as a function of time by an expression of the form envelope V,,E(t)sin 2IIft where sin 2" represents a unity amplitude sine wave of frequencyf. E(t) is a voltage that is a function of time that follows the positive peaks of the envelope and is normalized to a maximum peak amplitude of one, and V is the actual positive peak amplitude of the envelope.

As shown in FIG. 9, the automatic gain control 76 has the line '/5 connected to

lines

80 and 81. The line 80 supplies one of the two inputs to a multiplier 82. The line 81 is connected to a negative peak and hold circuit through a diode 83. The output of the negative peak and hold circuit at the output of a differential amplifier 84 is the negative peak voltage, V,,. This output is supplied to a divider 85, which receives a constant input voltage on a line 86. The divider 85 inverts the negative output of the differential amplifier 84 and converts the signal on its output line 87 to a positive output of lOC/v, where C is the constant voltage input on the line 86 and V, is the positive peak voltage.

The multiplier 82 multiplies the input on the line 80 from the bandpass filter and amplifier 72 and the input from the divider 85 on the line 87 to supply an output voltage on the line 79- The output voltage on the line 79 is the envelope produced by the beam 1 l entering or leaving one of the portions 61-64 of the wire grid 41 and is one-tenth of the product of the two multiplicands 10C/V,, and V,,E(t) sin Zn-ft; thus, the output voltage on the line 79 is C[E(t) sin 21rft]. This means that each 9 of the focus envelopes at the line 79 has a constant amplitude determined by C while its shape, determined by E(r) sin 2IIit, is not distorted.

The output voltage of the automatic gain control 76 is connected through the output line 79 and a line 88 (see FIG. 7) to a detecting and counting means 89. The detecting and counting means 89 includes comparators 90 and 91 (see FIG. with each of the comparators producing an output when its threshold signal is crossed.

The threshold signal for the positive comparator 90 is ten percent of the positive peak voltage on the output line 79. Since the positive peak voltage is 7 volts, the positive comparator 90 produces an output signal each time that the voltage within the envelope exceeds 0.7 volt. Similarly, the negative comparator 91 produces an output each time that the voltage is more negative than ten percent of the negative peak voltage. Thus, each time that the negative voltage within the envelope exceeds ().7 volt, the comparator 91 produces an output pulse.

The output of the positive comparator 90 is connected to a single shot 92 which produces a positive pulse whenever the comparator 90 produces an output, which is a positive pulse. The positive output of the single shot 92 is inverted by an inverter 93 and supplied as a negative input pulse to a NAND gate 94.

Similarly, the output of the negative comparator 91 causes a single shot 95 to have a positive output pulse whenever the negative comparator 91 has its threshold voltage crossed to produce a positive output. The positive pulse of the single shot 95 is inverted by an inverter 96 and supplied as a negative input pulse to the NAND gate 94.

Accordingly, whenever either of the

comparators

90 and 91 has its threshold voltage crossed to supply a negative input to the NAND gate 94, a positive pulse occurs at the output of the NAND gate 94 since it produces a negative pulse only when both of its inputs are positive. Thus, each output pulse from the

comparators

90 and 91 appears as a positive pulse at the output of the NAND gate 94. Therefore, all of the positive output pulses at the NAND gate 94 are a total of the number of times that the voltage within the envelope exceeds ten percent of the peak positive or negative voltage when the beam 11 enters or leaves one of the portions 61-64 of the wire grid 41.

The output of the NAND gate 94 is supplied through a line 97 to input lines 98 and 99 (see FIG. 7) of an up down counting means 100. The input line 98 supplies one of the inputs to a NAND gate 101 (see FIG. 11), which has its other input supplied from Q output of a clocked JK flip flop 102. The Q output of the flip flop 102 is tied to K input of the flip flop 102.

Thus, when the flip flop 102 is supplying a positive signal at the Q output, the NAND gate 101 produces a negative pulse on its output line 103 each time that a positive pulse appears on the input line 98 to the NAND gate 101 from the output of the NAND gate 94. Accordingly, each pulse counted by the NAND gate 94 from each of the

comparators

90 and 91 is supplied on the output line 103 when the Q output of the flip flop 102 is positive.

The input line 99 of the up down counting means 100 comprises one of the inputs to a NAND gate 104, which has its other input supplied from Q output of the flip flop 102. The Q output is tied to J input of the flip flop 102.

Thus, when the Q output of the flip flop 102 is positive, the NAND gate 104 produces a negative pulse on its output line 105 each time that the NAND gate 94 has a positive pulse on the output line 97. Therefore, if the flip flop 102 is set in the state in which the Q output is positive, then the NAND gate 104 counts the pulses from the NAND gate 94 and supplies these on the line 105.

The flip flop 102 has its clock input C connected by a line 106 to a single shot 107 (see FIG: 7). The single shot 107 is connected through a line 108 to receive the +X SCAN signal from the analog unit 17 (see FIG. 1). Thus, each time that the beam 11 starts to move along the path 60 (see FIG. 2), the single shot 107 (see FIG. 7) is fired to supply a pulse to the C input of the flip flop 102 (see FIG. 11).

The negative going portion of the pulse from the single shot 107 trips the flip flop 102 to change its state. Therefore, each time that the beam 11 is to start movement along the path 60, the state of the flip flop 102 is changed to cause the opposite of the

NAND gates

101 and 104 from that in the prior cycle of movement of the beam 11 along the path 60 to be responsive to the positive pulses on the line 97.

The

lines

103 and 105 are connected to a twelve bit up down counter 110. When the NAND gate 101 is responsive to the positive pulses from the NAND gate 94 due to the Q output of the flip flop 102 being positive, the counter 110 counts up. When the NAND gate 104 is responsive to the positive pulses from the NAND gate 94 due to the Q output of the flip flop 102 being positive, the counter 110 counts down.

If the down count is greater than the up count, the counter 110 supplies a negative pulse on its output line 111. No signal is supplied on the output line 111 if the down count does not exceed the up count. One suitable example of the counter 110 is sold by Fairchild Semiconductor as model 9366.

When the flip flop 102 has its state changed so that the NAND gate 101 causes the counter 110 to count up, the signal from the Q output of the flip flop 102 is supplied not only to the NAND gate 101 but also to a single shot 112, which produces a positive pulse on its output line 113 when the Q output of the flip flop 102 goes positive. The output line 113 is connected to the master reset of the counter 110 to reset the counter 110 to zero prior to counting up.

When the flip flop 102 changes state to cause count down in the counter 110, the Q output, which enables the NAND gate 104 to be activated by any positive pulse on the line 99, also causes a single shot 114 to produce a positive pulse on its output line 1 15. The single shot 11 4 produces the positive pulse when the Q output of the flip flop 102 goes positive to enable the NAND gate 104 for the pulses from the NAND gate 94 to cause down count in the counter 110.

The output line 111 of the counter 110 is connected as one input to a NAND gate 116 (see FIG. 12) of a focus control unit 117. The other input to the NAND gate 116 is from a flip flop 118, which is the same type as the flip flop 102. The flip flop 118 receives a signal from the analog unit 17 at the start of the FOCUS SERVO signal to change its state. The pulse, which is identified as FOCUS SERVO INITIATE in the timing diagram of FIG. 15, is applied through a line 119 from the analog unit 17. An inverter 120 causes a negative pulse to be supplied to input S of the flip flop 118 whereby output Q of the flip flop 118 goes positive when the FOCUS SERVO INITIATE pulse is supplied on the line 119. Thus, a positive input is supplied to the NAND gate 116 from the Q output of the flip flop 118 at the start of the cycle for automatically focusing the beam 11.

The Q output of the flip flop 118 also is connected as one input to a NAND gate 121, which receives its other input from the output line 115 of the single shot 114. Thus, after the up count is completed and down count is about to start so that the single shot 114 provides a positive pulse, the NAND gate 121 has its output change state to supply a negative pulse on its output line 122.

The line 122 is connected as one of the inputs to each of

NAND gates

123 and 124. The other input to the NAND gate 123 is from a flip flop 125, which is the same type as the flip flop 102. Thus, the NAND gate 123 is connected to output Q of the flip flop 125. The other input to the NAND gate 124 is from output O of the flip flop 125.

The flip flop 125 has its clock input C connected by a line 126 to the output of the NAND gate 116. Thus, each time that there is a positive pulse from the NAND gate 116, the flip flop 125 changes state to cause the other of the

NAND gates

123 and 124 to have its output responsive to signals supplied from the NAND gate 121.

Each time that the NAND gate 121 provides a negative pulse on the output line 122, the counter 110 has counted up. Thus, at the completion of each up count, which is a counting of all the times that the voltage exceeds ten percent of the peak voltage in the eight envelopes produced during movement of the beam 11 one time around the path 60, one of the

NAND gates

123 and 124, depending on the state of the flip flop 125, produces a pulse to an eight bit up down counter 127. One suitable example of the eight bit up down counter 127 is sold by Fairchild Semiconductor as model 9366.

The output of the counter 127 is connected to an eight bit digital to analog converter (DAC) 128. One suitable example of the DAC 128 is sold by Beckman Instruments as model 845. The DAC 128 supplies a voltage through its output line 129 and a'coil driver 130 (see FIG. 7) to the focus coil 22.

If the counter 110 (see FIG. 11) produces a signal on the output line 111 at the completion of the down count, the NAND gate 116 (see FIG. 12) provides a positive pulse on the output line 126 to change the state of the flip flop 125. As a result, the other of the

NAND gates

123 and 124 produces an output in response to the output of the NAND gate 121 at the completion of the next cycle of the beam 11 along the path 60.

Each of the positive output pulses from the NAND gate 123 causes an up count in the counter 127. Each of the positive output pulses from the NAND gate 124 produces a down count in the counter 127. Thus, if the pulse supplied to the counter 127 is from the NAND gate 123, the voltage from the eight bit DAC 128 in creases to cause an increase in the current to the focus coil 22. If the NAND gate 124 supplies the pulse to the counter 127, then the eight bit DAC 128 decreases the output voltage on the line 129 to cause the current to the focus coil 22 to decrease.

The counter 110 (see FIG. 11) produces a negative pulse on the output line 111 only when the number of pulses in the down count exceeds that during the up count. This means that a worse focus exists during the second scan, which is when the down count is occur- 12 ring, than existed prior to change in the current in the focus coil 22 at the end of the first scan (an up count). Therefore, it is necessary to change the direction in which the counter 127 (see FIG. 12) is being stepped or counted so that the direction of change of the current to the focus coil 22 is reversed.

Thus, when the next scan occurs after the down count so that the counter is counting up, the output on the line 122 of the NAND gate 121 at the completion of the up count would now be supplied to the opposite of the

NAND gates

123 and 124 to shift the direction in which the counter 127 counts. This reverses the direction of change of the current to the focus coil 22. That is, if the current had been increasing, it will be decreased. Similarly, if the current had been decreasing, it will be increased.

Of course, if there is no output on the line 111 at the completion of the down count of the counter 110, then this means that the focus has been improved by the change in the current to,the focus coil 22 at the end of the prior scan (up count). Therefore, the state of the flip flop 125 is not altered, and the counter 127 continues to count in the same direction.

Since it is desired for a minimum number of the pulses exceeding ten percent of the negative or positive peak voltage to exist during movement of the beam 11 through a cycle along its path 60 to indicate the best quality of the focus of the beam 1 1, there is a focus coil current at which the number of pulses will be at a minimum. Referring to FIG. 16, there is shown a relationship between the number of pulses and the current in the focus coil 22. It will be assumed that the number of pulses is being reduced by each increase in the focus coil current at the end of each up count by the counter 110 (see FIG. 11) with the counter 127 (see FIG. 12) counting up because of pulses from the NAND gate 123 to increase the voltage from the DAC 128.

When the minimum number of pulses occurs, the next scan (a down count) will result in the number of pulses increasing as indicated by step 131 in FIG. 16. When this occurs, the counter 110 (see FIG. 11) produces a negative pulse on the output line 111 because the number of pulses counted during the down count exceeds those counted during the up count whereby the flip flop 125 (see FIG. 12) has its state changed. As a result, the current to the focus coil 22 will be decreased by the DAC 128 producing a lower voltage since the counter 127 is now counted down at the end of the up count in the counter 110 by the NAND gate 124.

With the focus coil current now decreased during the next down count in the counter 110, the number of pulses will be a minimum during this cycle of movement of the beam 11 along the path 60. This is during a down count in the counter 110. Thus, there is no change in the state of the flip flop 125 at the end of the down count in the counter 110.

As a result, the current to the focus coil 22 is decreased at the end of thescan producing the next up count in the counter 110. This increases the number of pulses beyond the minimum during the next down count in the counter 110 so that the counter 110 again produces a negative pulse on the output line 111 at the end of the down count in the counter 110. This increase in the number of the pulses during the down count is indicated by step 132 in FIG. 16.

As a result of the negative pulse on the output line 111, the flip flop 125 again changes state. This causes 13 the current to the focus coil 22 to again be increased until the step 131 is again reached. At this time, the counter 110 again produces a pulse on the line 111 to again shift the state of the flip flop 125.

When three of these pulses have occurred on the line 111, the minimum number of pulses within the envelope will have been produced since these three changes indicate that the valley of the number of pulses has been reached as indicated by curved line 133 in FIG. 16. When this occurs, it is desired to stop the focus control unit from functioning.

Accordingly, the output line 126 of the NAND gate 116 is not only connected to the flip flop 125 but also to a three state counter 135 (see FIG. 12). Thus, the counter 135 counts each time that the flip flop 125 changes state. The counter 135 is actually a four bit counter with its third output being employed to activate a single shot 136. One suitable example of the counter 135 is sold by Fairchild Semiconductor as model 9366.

The single shot 136 produces a negative output pulse on its line 137 to reset input R of the flip flop 118.. This causes the Q output of the flip flop 118 to become negative whereby the NAND gate 121 can no longer change state when a positive pulse is supplied from the line 115. Therefore, when the flip flop 125 has changed state three times so as to indicate that the minimum number of pulses are within the eight envelopes, the counter 135 prevents further activation of the NAND gate 121 so that there is no further change to the DAC 128. The DAC 128 holds its final output until the next time a focus servo cycle is activated to supply the FOCUS SERVO INITIATE pulse to the flip flop 118.

The single shot 136 also provides the negative pulse on its output line 137 by a line 138 to the master reset of the counter 135. This negative pulse sets the counter 135 to zero.

As previously mentioned, there are eight envelopes produced by the beam 11 crossing the four portions 61-64 (see FIG. 2) of the wire grid 41. The total times that the voltage in these eight envelopes exceeds ten percent of the positive or negative peak voltage is what is counted in the counter 110.

Furthermore, the beam 1 1 passes completely around the wire grid 41 sixteen times. Even though two of these cycles are required for the up and down count to the counter 110, the eight times that there can be control of the current to the focus coil 22 are sufficient to focus the beam 11 satisfactorily.

In addition to automatically correcting the focus of the beam 11, the present invention also allows manual correction for quality of the focus of the beam 11. Accordingly, the output line 97 (see FIGS. 7 and of the NAND gate 94 (see FIG. 10) also supplies the positive output pulses of the NAND gate 94 through a line 150 (see FIG. 7) to an eight-twelve bit counter 151. One suitable example of the eight-twelve bit counter 151 is sold by Fairchild Semiconductor as model 9366.

The eight-twelve bit counter 151 counts all of the pulses from the output of the NAND gate 94 and obtains an average of the eight envelopes produced during one cycle of movement of the beam 11 around the path 60 by dividing by two or four depending on the size of the beam 11. By dividing by two or four, the average quality of the focus of the beam 11 in both the X and Y axes is obtained.

The output of the eight-twelve bit counter 151 is supplied to an eight-bit digital to analog converter (DAC) 14 152. One suitable example of the DAC 152 is sold by Analog Devices Inc. as model DAC SQM.

When the scan of the beam 11 in the +X direction starts as indicated by the +X SCAN pulse on the line 108 to cause the single shot 107 to produce a positive pulse, the positive pulse from the single shot 107 is not only transmitted over the line 106 to the flip flop 102 (see FIG. 11) but also is supplied through a line 153 (see FIG. 7) to the eight bit DAC 152 to strobe it. When this occurs, the eight bit DAC 152 stores the value of the counter 151 until the next strobe and supplies an output voltage, which is representative of the average quality of the focus of the beam 11 in both the X and Y axes at the completion ofa cycle of movement of the beam 11 along its entire path 60, on its output line 154.

A line 155 transmits this signal from the line 154 to a voltmeter 156. The output voltage on the line 154 is proportional to the average width of the eight envelopes. Through applying the voltage from the eight bit DAC 152 to the voltmeter 156, the quality of the focus supplied to a single shot 157 to produce a positive pulse on its output line 158. The positive pulse on the output line 158 is supplied by a line 159 to master reset of the eight-twelve bit counter 151 to reset it to zero. Since this occurs after the eight bit DAC 152 has stored the value of the counter 151 and from this supplies the output voltage over the line 154, the counter 151 is ready to begin to count again. Since the output pulse of each of the

single shots

107 and 157 is very short, the eighttwelve bit counter 151 is set to zero before the beam 1 1 reaches the vertical portion 61 (see FIG. 2) of the wire grid 41.

To cause the jitter frequency to be supplied over the line 49 (see FIG. 14) when there is not to be automatic adjustment of the quality of the focus of the beam 11, a positive input voltage must be supplied to the NAND gate 46. This can be supplied from closing a manual switch 160 to supply a positive input voltage to the OR gate 45. Similarly, a MONITOR ENABLE signal can be supplied to the OR gate 45 through an optical isolator 161, which is the same as the optical isolator 44, to the OR gate 45. The input of the MONITOR ENABLE signal from the analog unit 17 is over a line 162 to the optical isolator 161. A line 163 from the optical isolator 161 is grounded.

Thus, by either closing the manual switch 160 or producing the MONITOR ENABLE signal from the analog unit 17, a positive input voltage is supplied to the NAND gate 46 whereby the jitter frequency appears on the line 49. Of course, it should be understood that the voltmeter 156 also visually presents the quality of the beam during the automatic adjustment.

In addition to manual correction for quality of the focus of the beam 11, the present invention also allows a determination of whether the brightness of the beam 11 is sufficient to enable manual correction of the focus of the beam 11.

As shown in FIG. 6, an envelope is produced when the focus of the beam 11 is satisfactory but there is poor illumination. The envelope 170 has a non-flat or slanted top. Accordingly, the area beneath the top and above ninety percent of the peak voltage is smaller, for

example, in comparison with that shown in the envelope 69, which has good focus and illumination in FIG. 5. Thus, the number of times that the voltage exceeds ninety percent of the peak voltage in the envelope 170 is less than in the envelope 69; this indicates that the envelope 69 is produced by the beam 11 having better illumination than when the envelope 170 is produced by the beam 11.

The poor illumination is due to the current density not being the same in all portions of the beam 11. Correction for illumination is obtained through correcting the alignment of the beam 11 as more particularly shown and described in the copending patent application of Hans C. Pfeiffer et al for Method And Apparatus For Aligning Electron Beams, Ser. No. 393,365, filed Aug. 31, 1973, now U.S. Pat. No. 3,894,271, and assigned to the same assignee as the assignee of this application.

To ascertain whether the illumination of the beam 11 is poor, an illumination detecting means 171 (see FIG. 7) is connected to the output line 79 of the automatic gain control 76 by a line 172. The line 172 is connected to a squarer 173 (see FIG. 13), which is a multiplier used as a squarer, of the illumination detecting means 171. The squarer 173 makes each of the envelopes produced by the beam 11 passing over the wire grid 41 entirely positive. I

To ascertain each time that the voltage exceeds ninety percent of the peak voltage as the beam 11 passes over the wire grid 41, the output of the squarer 173 is connected to a comparator 174, which produces an output each time that its threshold voltage is crossed by the output voltage of the squarer 173 exceeding eighty-one percent of the peak voltage. This is because the squarer 173 reduces the voltage of ninety percent of the peak voltage to eighty-one percent of the peak voltage. The output of the comparator 174 is connected to an eight-twelve bit counter 175, which is preferably the same as the eight-twelve bit counter 151. The counter 175 counts each time that the comparator 174 produces an output due to its threshold being crossed.

The eight-twelve bit counter 175 counts all of the pulses from the output of the comparator 174 and obtains an average of the eight envelopes produced during one cycle of movement of the beam 11 around the path 60 (see FIG. 2) by dividing by two or four depending on the size of the beam 11. By dividing by two or four, the average quality of illumination of the beam 11 in both the X and Y axes is obtained.

The output of the eight-twelve bit counter 175 (see FIG. 13) is supplied to an eight bit digital to analog converter (DAC) 176. When the beam 11 starts to move in the +X direction as indicated by the +X scan signal on the line 108 (see FIG. 7) to cause the single shot 107 to produce a positive output pulse, the positive pulse is not only transmitted to the flip flop 102 and the eight bit DAC 152 but also is transmitted by a line 177 to the eight bit DAC 176 (see FIG. 13) to strobe it. When this occurs, the eight bit DAC 176 supplies an output voltage, which is representative of the illumination of the beam 11 in both the X and Y axes at the completion of a cycle of movement of the beam 11 around its closed path 60, on its output line 178.

The output voltage on the line 178 is transmitted to a comparator 180. The output voltage on the line 178 is 16 proportional to the average width of the eight envelopes at the ninety percent level of the peak voltage.

If the output voltage from the eight bit DAC 176 is high enough to indicate that the illumination of the beam 11 is satisfactory, then the threshold voltage of the comparator 180 is crossed and a positive output pulse occurs on a line 181. However, if the output voltage from the eight bit DAC 176 is not sufficient to indicate satisfactory illumination, then the threshold voltage of the comparator 180 is not crossed and a positive output voltage does not appear on the output line 181 from the comparator 180.

The output line 181 of the comparator 180 is connected to an inverter 182, which is connected by a line 183 as one input to a NAND gate 184. The other input to the NAND gate 184 is from an inverter 185, which receives an input from a comparator 186.

The comparator 186 is connected to the line 154 (see FIG. 7) by a line 187 to receive the output voltage of the eight bit DAC 152. As previously mentioned, this voltage is proportional to the average width of the eight envelopes at the ten percent level of the peak voltage of the envelope.

If the voltage from the eight bit DAC 152 exceeds the threshold voltage of the comparator 186 (see FIG. 13), then a pulse appears on an output line 189, which connects the comparator 186 to the inverter 185. This is indicative of the quality of the focus of the beam 11 being poor.

If the output voltage from the eight bit DAC 152 does not exceed the threshold voltage of the comparator 186, then there is no output from the comparator 186 so that the output line 189 does not have a positive pulse thereon. This indicates that the focus of the beam 11 is satisfactory.

Accordingly, when the focus of the beam 11 is satisfactory, the inverter provides a positive pulse as the other input to the NAND gate 184.If the illumination is poor, then a positive pulse also appears on the line 183 from the inverter 182 because the threshold voltage of the comparator 180 was not crossed by the output of the eight bit DAC 176. As a result of two positive or high signals as the inputs to the NAND gate 184, a low output occurs on output line 190 of the NAND gate 184. An inverter 191 changes this signal to a high for supply to a status decoder 192. This high to the status decoder 192 is indicative of poor illumination and causes the status decoder 192 to produce a signal to illuminate a warning light. As a result, the alignment of the beam 11 must be corrected for the focus and astigmatism of the beam 11 to be manually corrected.

The output line 158 (see FIG. 7) of the single shot 157 also is connected by a line 193 to the master reset of the counter 175 (see FIG. 13) to reset it to zero. Since this occurs after the eight bit DAC 176 has supplied the output voltage over the line 178, the counter 175 is ready to begin to count again at the same time that the counter 151 is. Since the output pulse of each of the

single shots

107 and 157 is very short, the counter 175 is set to zero before the beam 11 reaches the vertical portion 61 (see FIG. 2) of the wire grid 41 in the same manner as discussed with respect to the counter 151.

Considering the operation of the present invention, the beam 11 is disposed at the center of its deflection field for manual correction for focus and astigmatism of the beam 11. Either the MONITOR ENABLE signal is supplied to the optical isolator 161 (see FIG. 14) or the manual switch 160 is closed to cause a jitter frequency to be supplied to the X and Y electrostatic deflection plates 3134 (see FIG. 1) in phase from the oscillator 47 (see FIG. 14). Manual controls on the column control unit 16 are turned to adjust the focus of the beam 11 and its astigmatism in accordance with the voltage on the voltmeter 156. The manual controls change the resistances of potentiometers in the column control unit 16.

After the beam 11 has been satisfactorily manually adjusted at the center of the field. the beam 11 is deflected to each of the corners of the field with the focus coil 22 being adjusted initially and then the stigmator coils 21A and 21B being adjusted. Thereafter, the beam 11 is deflected to the middle of each of the sides of the field and the stigmator coils 21A and 21B are manually adjusted to further correct for astigmatism. This manual adjustment is not necessary for about a year unless some portion of the electron beam assembly is disassembled. 7

Whenever there is to be automatic adjustment of the focus of the beam during one of the C cycles, the FOCUS SERVO signal, which is a positive pulse as shown in the timing diagram of FIG. 15, is supplied through the optical isolator 44 (see FIG. 14) to the OR gate 45 to cause a jitter frequency to appear on the output line 49 of the dither control 48. The output line 49 supplies this voltage through the analog unit l7 (see FIG. 1) to the X and Y electrostatic deflection plates 31-34 in phase.

As shown in the timing diagram of FIG. 15, the X magnetic deflection voltage to the X magnetic deflection coils 23 and 24 increases from its minimum to its maximum to move the beam 11 in the +X direction as indicated by the arrow 65 in FIG. 2. The Y magnetic deflection voltage to the Y magnetic deflection coils 25 and 26 starts to decrease from its maximum to its minimum as soon as the X magnetic deflection voltage reaches its maximum. Thus, this causes the beam 11 to move in the Y direction as indicated by the arrow 66 in FIG. 2.

At the time that the FOCUS SERVO signal is supplied, the +X SCAN signal, which is a positive pulse, is supplied from the analog unit 17 (see FIG. 1) on the line 108 (see FIG. 7). As shown in the timing diagram of FIG. 15, the +X SCAN pulse remains up for the entire time that the beam 11 moves in the +X direction as indicated by the arrow 65 in FIG. 2.

The +X SCAN pulse causes the timing pulse from the single shot 107 (see FIG. 7) to be supplied through the line 106 to change the state of the flip flop 102 (see FIG. 11). It also Strobes the eight bit DAC 152 (see FIG. 7) and the eight bit DAC 176 (see FIG. 13) to cause their outputs to be supplied at the same time to the voltmeter 156 (see FIG. 7) and the comparator 186 (see FIG. 13) from the eight bit DAC 152 and to the comparator 180 (see FIG. 13) from the eight bit DAC 176. However, illumination is not utilized during the automatic adjustment. Furthermore, the voltmeter 156 is normally not utilized during automatic adjustment of the focus of the beam 11.

It should be understood that the timing pulse from the single shot 107 and the timing pulse from the single shot 157 are very short in time in comparison with the +X SCAN as shown in FIG. 15. Thus, the flip flop 102 changes state almost as soon as the +X SCAN signal is supplied. 3

As shown in the timing diagram of FIG. 15, the state of the flip flop 102 (see FIG. 11) is set so that the counter 110 will count up during the deflection of the beam 11 around its path 60 (see FIG. 2). Thus, the NAND gate 101 (see FIG. 11) will supply pulses from the output of the NAND gatee 94 (see FIG. 10) to the counter 110 (see FIG. 11).

As shown in the timing diagram of FIG. 15, the pulse from the output of the single shot 112 (see FIG. 11) to the master reset of the counter 110 is supplied when the flip flop 102 changes state. Thus, the counter 110 is set at zero just after the flip flop 102 changes state to allow pulses from the NAND gate 94 to be supplied through the NAND gate 101 to the counter 110.

While the beam 11 starts to scan in the +X direction as soon as the +X SCAN signal is supplied to increase the magnetic deflection voltage to the coils 23 and 24 (see FIG. 1), the flip flop 102 (see FIG. 11) does not change state until the negative going portion of the output of the single shot 107. When the Q output of the flip flop 102 goes positive, the single shot 112 supplies a positive pulse on the line 113 to reset the counter 110 to zero. All of this occurs before the beam 11 reaches the vertical portion 61 (see FIG. 2) of the wire grid 41 during its movement in the +X direction.

At the completion of the first deflection cycle of the beam 11 around its path 60, another +X SCAN pulse is supplied from theanalog unit 17. The counter 110 has counted each time that the voltage in each of the eight envelopes has exceeded ten percent of the positive or negative peak voltage. These pulses are stored in the counter 110.

When the +X SCAN pulse occurs to cause the next deflection cycle of the beam 11 around its closed path 60, the state of the flip flop 102 is changed by the pulse from the single shot 107 (see FIG. 7) so that the 6 output of the flip flop 102 goes positive. This results in the output pulses from the NAND gate 94 (see FIG. 10) being supplied through the NAND gate 104 (see FIG. 11) to the counter 110. This causes the counter 110 to count down from the count stored in the counter 110 during the prior deflection cycle of the beam 11 around its closed path 60.

If the down count exceeds the up count, then a negative pulse appears on the line 111 as indicated in the timing diagram of FIG. 15 as BORROW WRONG DI- RECTION ON LINE 111. This negative signal occurs during or at the end of the down count depending on when the down count exceeds the up count. If the down count does not exceed the up count, then no signal appears on the line 111.

When the FOCUS SERVO signal starts, the FOCUS SERVO INITIATE signal (see FIG. 15) is supplied from the analog unit 17 (see FIG. 1) to the line 119 (see FIG. 12) to set the flip flop 118 in the condition in which the Q output of the flip flop 118 is positive. The Q output of the flip flop 118 is supplied as one input to each of the

NAND gates

116 and 121.

When a positive pulse from the single shot 114 (see FIG. 11) occurs due to the flip flop 102 changing state so that the Q output goes positive to start the down count, the NAND gate 121 (see FIG. 12) supplies a negative pulse on the line 122 since both of its inputs are positive. This causes one of the

NAND gates

123 and 124, depending on the state of the flip flop 125, to have its output supply a positive pulse to the counter 127 whereby the eight bit DAC 128 supplies a change in voltage to the focus coil driver 130 (see FIG. 7) to change the current to the focus coil 22.

If the down count in the counter 110 (see FIG. 11) exceeded the up count, then the negative pulse on the line 111 causes the NAND gate 116 (see FIG. 12) to produce a positive pulse on the line 126 to change the state of the flip flop 125. Accordingly, when the counter 110 completes the up count during the next cycle of deflection of the beam 11 around its closed path 60 (see FIG. 2) over the wire grid 41, the other of the NAND gates 123 (see FIG. 12) and 124 will supply the signal to the counter 127 to reverse the direction of count in the counter 127 whereby the eight bit DAC 128 changes the direction in which the magnitude of the output voltage is being altered. If this change in voltage results in the down count in the counter 110 (see FIG. 11) again exceeding the up count during the 7 next down count cycle, then the flip flop 125 (see FIG.

12) again changes state. This is shown in FIG. by a second negative pulse on the line 111 (see FIG. 11) no later than completion of the down count. If at the end of the third down count cycle, another negative pulse appears on the line 111 to indicate that the down count exceeded the up count in the counter 110, then the focus is at its best adjustment as previously mentioned. As a result, the three state counter 135 (see FIG. 12), which has counted each time that there was a positive pulse on the line 126, causes the flip flop l 18 to change state through a negative pulse from the single shot 136 to the R input of the flip flop 118.

The changing of the state of the flip flop 118 prevents any further negative pulse from being supplied from the NAND gate 121. It also prevents the NAND gate 116 from supplying positive pulses as an output since the output of the NAND gate 116 remains up. This is because the input to the

NAND gates

116 and 121 from the O output of the flip flop 118 remains negative.

The FOCUS SERVO signal is supplied to the optical isolator 44 (see FIG. 14) throughout the C cycle. During the time that the optical isolator 44 is supplying an input to the OR gate 45 to activate the dither control 48, this signal also is supplied through an inverter 194 and a line 195 to the analog unit 17 (see FIG. 1) to inhibit any other operations which occur during the C cycle such as correcting for alignment of the beam 11, for example, as shown and described in the aforesaid Pfeiffer et al application.

While the present invention has shown and described a change in the voltage to the focus coil driver 130 occurring only after the end of an up count to the counter 110 and a reversal in the direction in which the magnitude is being changed only during the down count or at the end thereof, it should be understood that such is not a requisite of the present invention. Thus, if desired, each cycle of deflection of the beam 1 1 along its closed path 60 could be compared with the prior cycle and correction made to the focus of the beam 11 after each cycle rather than after every other cycle. Of course, this would require a different circuit arrangement.

While the present invention has shown and described the beam 11 as being moved in the square path 60, it should be understood that such is not a requisite for satisfactory operation. Thus, the beam 11 could be deflected in any closed path such as a circle or a rectangle, for example. Furthermore, it is not even necessary for the beam 11 to be deflected or moved in a closed path but only for the beam 11 to move along the same predetermined path during each cycle of movement.

Anadvantage of this invention is that the focus of a beam of charged particles can be automatically adjusted. Another advantage of this invention is that the quality of the beam focus is visually obtainable. A further advantage of this invention is that the quality of illumination of the beam is visually obtainable.

While the invention has been particularly shown and described with reference to a preferred embodiment 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.

What is claimed is:

l. A method for correcting an aberration of a beam of charged particles including:

directing the beam at a constant velocity in a predetermined path across a target during each cycle; ascertaining the lengths of time for the beam to enter and leave at least one portion of the target;

and adjusting the aberration of the beam in accordance with the lengths of time for the beam to enter and leave the portion of the target in comparison with the lengths of time for the beam to enter and leave the portion 'of the target during the prior cycle of movement of the beam in its predetermined path.

2. The method according to claim 1 including:

oscillating the beam in a single predetermined direction during its movement in the predetermined path with the single predetermined direction being other than any direction of movement of the beam in its predetermined path;

and ascertaining the length of time for the beam to enter the portion of the target by voltage signals produced by the oscillations when the beam enters the portion of the target and the length of time for the beam to leave the portion of the target by voltage signals produced by the oscillations when the beam leaves the portion of the target.

3. The method according to claim 2 including:

directing the beam in a closed path to provide the predetermined path;

and forming the target in the shape of a cross so that four portions of the target are crossed during each movement of the beam in the closed path.

4. The method according to claim 3 including:

counting the number of times that the voltages produced by the oscillations pass a predetermined voltage limit during each entrance or exit of the beam over the portion of the target to ascertain the length of time for the beam to enter or leave the portion of the target;

and adjusting the aberration of the beam in accordance with the count during a cycle of movement of the beam in its predetermined path in comparison with the count during the prior cycle of movement of the beam in its predetermined path.

5. The method according to claim 4 in which:

the aberration of the beam is the quality of the focus of the beam;

and the focus is adjusted by changing the current flow through the focus coil.

6. The method according to claim 5 including automatically reversing the direction of change of the current flow through the focus coil if the focus quality decreases.

7. The method according to claim 5 including:

21 adjusting the focus only every other cycle of movement of the beam in its predetermined path; and adjusting the focus only after completion of a cycle of movement of the beam in its predeter- 22 means to move the beam at a constant velocity in a predetermined path across a target during each cycle; means to ascertain the lengths of time for the beam to mined path f ll Wiflg C mPlCIi H f th cycl of enter and leave at least one portion of the target movement Of the beam in its predetermined path in during each cycle of movement in the predeterwhich comparison of the count with the count from i d h; the Prior Cycle of movement of the beam in its P and means to correct the aberration of the beam i detemhhed P l0 accordance with the lengths of time for the beam to e method aeeordmgto elalm Zmehldmg? enter and leave the portion of the target in compareouhtlhg the humheh of thhes that the Voltages P ison with the lengths of time for the beam to enter dueed y t oeelhatlohs P a Predetemhhed and leave the portion of the target during the prior voltage limit during each entrance or exit of the cycle f movement f the beam i its Predetflbeam over the portion of the target to ascertain the mined path length of time for the beam to enter or leave the The System according to Claim 16 including; P e the target; means to oscillate the beam in a single predetermined and adlustmg the aberranon of the b m accor direction during its movement in the predeterdance wlth h F durmgfl cycle movemeht mined path by' said moving means with the single of theibeam predtermmeq path m compan' 20 predetermined direction being other than any di- Son wlth the count durmg pnor Cycle of move rection of movement of the beam in its predeterment of the beam in its predetermined path. mined path. 5 method i il i 9 clhalm 8 It F f said ascertaining means including: t f h g t 6 cam is t e qua 0 t e Ocus means to obtain the voltages produced by said oso t e cillating means when the beam enters and leaves and the focus lS ad usted by changing the current i the portion of the target, flow through the focus coil.

and means to count the number of times that the 10. The method according to claim 9 including autovoltages produced by the beam entering and matrcally reversing the direction of change of the curleaving the portion of the target pass a predeterrent flow through the focus COll if the focus quality decreases mined voltage limit during each cycle of movement of the beam in its predetermined path; 11. The method according to claim 9 including:

. and said correcting means includes means to correct ad usting the focus only every other cycle of movethe aberration of the beam in accordance with the ment of the beam in its predetermined path;

count obtained by said counting means during a and ad usting the focus only after completion of a cycle of movement of the beam in its predetermined path in comparison with the count obtained by said counting means during the prior cycle of movement of the beam in its predetermined path.

18. The system according to claim 17 in which said oscillating means comprises means to apply a single high frequency jitter to said moving means.

19. The system according to claim 18 in which:

said moving means includes first and second means to deflect the beam in orthogonal directions;

and said first and second deflecting means moves the beam in a closed path across portions of the target extending in the orthogonal directions.

20. The system according to claim 19 in which:

the aberration of the beam is the quality of the focus of the beam;

and said correcting means includes means to change the focus of the beam in accordance with the count obtained by said counting means during a cycle of movement of the beam in its predetermined path in comparison with the count obtained by said counting means during the prior cycle of movement of the beam in its predetermined path.

21. The system according to claim 20 in which said focus change means includes means to change the focus only every other cycle of movement of the beam in its predetermined path with the change occurring only after completion of a cycle of movement of the beam in its predetermined path following completion of the cycle of movement of the beam in its predetermined path in which comparison of the count with the count from the prior cycle of movement of the beam in its predetermined path occurred.

cycle of movement of the beam in its predetermined path following completion of the cycle of movement of the beam in its predetermined path in which comparison of the count with the count from the prior cycle of movement of the beam in its pre- 4 determined path occurred.

12. The method according to claim 1 in which:

the aberration of the beam is the quality of the focus of the beam;

and the focus is adjusted by changing the current flow through the focus coil.

13. The method according to claim 12 including automatically reversing the direction of change of the current flow through the focus coil if the focus quality decreases.

14. The method according to claim 13 including:

directing the beam in a closed path to provide the predetermined path;

15. The method according to claim 12 including:

adjusting the focus only every other cycle of movement of the beam in its predetermined path;

and adjusting the focus only after completion of a cycle of movement of the beam in its predetermined path following completion of the cycle of movement of the beam in its predetermined path in which comparison of the count with the count from the prior cycle of movement of the beam in its predetermined path occurred.

16. A system for correcting an aberration of a beam of charged particles including:

22. The system according to claim 17 in which said counting means counts the number of times that the voltages exceed a predetermined minimum voltage.

23. The system according to claim 22 in which the predetermined minimum voltage is ten percent of the peak voltage.

24. The system according to claim 23 in which:

the aberration of the beam is the quality of the focus of the beam;

25. The system according to claim 24 in which said focus change means includes means to change the focus only every other cycle of movement of the beam in its predetermined path with the change occurring only after completion of a cycle of movement of the beam in its predetermined path following completion of the cycle of movement of the beam in its predetermined path in which comparison of the count with the 24 count from the prior cycle of movement of the beam in its predetermined path occurred.

26. The system according to claim 17 in which:

the aberration of the beam is the quality of the focus of the beam;

and said correcting means includes means to change the focus of the beam in accordance with the count obtained by said counting means during a cycle of movement ofthe beam in its predetermined path in comparison with the count obtained by said counting means during the prior cycle of movement of the beam in its predetermined path.

27. The system according to claim 26 in which said focus change means includes means to change the focus only every other cycle of movement of the beam in its predetermined path with the change occurring only after completion of a cycle of movement of the beam in its predetermined path following completion of the cycle of movement of the beam in its predetermined path in which comparison of the count with the count from the prior cycle of movement of the beam in its predetermined path occurred.

28. The system according to claim 16 in which said oscillating means comprises means to apply a single high frequency jitter to said moving means.

Claims

1. A method for correcting an aberration of a beam of charged particles including: directing the beam at a constant velocity in a predetermined path across a target during each cycle; ascertaining the lengths of time for the beam to enter and leave at least one portion of the target; and adjusting the aberration of the beam in accordance with the lengths of time for the beam to enter and leave the portion of the target in comparison with the lengths of time for the beam to enter and leave the portion of the target during the prior cycle of movement of the beam in its predetermined path.

2. The method according to claim 1 including: oscillating the beam in a single predetermined direction during its movement in the predetermined path with the single predetermined direction being other than any direction of movement of the beam in its predetermined path; and ascertaining the length of time for the beam to enter the portion of the target by voltage signals produced by the oscillations when the beam enters the portion of the target and the length of time for the beam to leave the portion of the target by voltage signals produced by the oscillations when the beam leaves the portion of the target.

3. The method according to claim 2 including: directing the beam in a closed path to provide the predetermined path; and forming the target in the shape of a cross so that four portions of the target are crossed during each movement of the beam in the closed path.

4. The method according to claim 3 including: counting the number of times that the voltages produced by the oscillations pass a predetermined voltage limit during each entrance or exit of the beam over the portion of the target to ascertain the length of time for the beam to enter or leave the portion of the target; and adjusting the aberration of the beam in accordance with the count during a cycle of movement of the beam in its predetermined path in comparison with the count during the prior cycle of movement of the beam in its predetermined path.

5. The method according to claim 4 in which: the aberration of the beam is the quality of the focus of the beam; and the focus is adjusted by changing the current flow through the focus coil.

7. The method accorDing to claim 5 including: adjusting the focus only every other cycle of movement of the beam in its predetermined path; and adjusting the focus only after completion of a cycle of movement of the beam in its predetermined path following completion of the cycle of movement of the beam in its predetermined path in which comparison of the count with the count from the prior cycle of movement of the beam in its predetermined path occurred.

8. The method according to claim 2 including: counting the number of times that the voltages produced by the oscillations pass a predetermined voltage limit during each entrance or exit of the beam over the portion of the target to ascertain the length of time for the beam to enter or leave the portion of the target; and adjusting the aberration of the beam in accordance with the count during a cycle of movement of the beam in its predetermined path in comparison with the count during the prior cycle of movement of the beam in its predetermined path.

9. The method according to claim 8 in which: the aberration of the beam is the quality of the focus of the beam; and the focus is adjusted by changing the current flow through the focus coil.

10. The method according to claim 9 including automatically reversing the direction of change of the current flow through the focus coil if the focus quality decreases.

11. The method according to claim 9 including: adjusting the focus only every other cycle of movement of the beam in its predetermined path; and adjusting the focus only after completion of a cycle of movement of the beam in its predetermined path following completion of the cycle of movement of the beam in its predetermined path in which comparison of the count with the count from the prior cycle of movement of the beam in its predetermined path occurred.

12. The method according to claim 1 in which: the aberration of the beam is the quality of the focus of the beam; and the focus is adjusted by changing the current flow through the focus coil.

14. The method according to claim 13 including: directing the beam in a closed path to provide the predetermined path; and forming the target in the shape of a cross so that four portions of the target are crossed during each movement of the beam in the closed path.

15. The method according to claim 12 including: adjusting the focus only every other cycle of movement of the beam in its predetermined path; and adjusting the focus only after completion of a cycle of movement of the beam in its predetermined path following completion of the cycle of movement of the beam in its predetermined path in which comparison of the count with the count from the prior cycle of movement of the beam in its predetermined path occurred.

16. A system for correcting an aberration of a beam of charged particles including: means to move the beam at a constant velocity in a predetermined path across a target during each cycle; means to ascertain the lengths of time for the beam to enter and leave at least one portion of the target during each cycle of movement in the predetermined path; and means to correct the aberration of the beam in accordance with the lengths of time for the beam to enter and leave the portion of the target in comparison with the lengths of time for the beam to enter and leave the portion of the target during the prior cycle of movement of the beam in its predetermined path.

17. The system according to claim 16 including: means to oscillate the beam in a single predetermined direction during its movement in the predetermined path by said moving means with the single predetermined direction being other than any direction of movement of the beam in its predetermined path; said ascertaining means including: means to obtain the voltages produced by said oscillating means when the beam enters and leaves the portion of the target; and means to count the number of times that the voltages produced by the beam entering and leaving the portion of the target pass a predetermined voltage limit during each cycle of movement of the beam in its predetermined path; and said correcting means includes means to correct the aberration of the beam in accordance with the count obtained by said counting means during a cycle of movement of the beam in its predetermined path in comparison with the count obtained by said counting means during the prior cycle of movement of the beam in its predetermined path.

19. The system according to claim 18 in which: said moving means includes first and second means to deflect the beam in orthogonal directions; and said first and second deflecting means moves the beam in a closed path across portions of the target extending in the orthogonal directions.

20. The system according to claim 19 in which: the aberration of the beam is the quality of the focus of the beam; and said correcting means includes means to change the focus of the beam in accordance with the count obtained by said counting means during a cycle of movement of the beam in its predetermined path in comparison with the count obtained by said counting means during the prior cycle of movement of the beam in its predetermined path.

24. The system according to claim 23 in which: the aberration of the beam is the quality of the focus of the beam; and said correcting means includes means to change the focus of the beam in accordance with the count obtained by said counting means during a cycle of movement of the beam in its predetermined path in comparison with the count obtained by said counting means during the prior cycle of movement of the beam in its predetermined path.

25. The system according to claim 24 in which said focus change means includes means to change the focus only every other cycle of movement of the beam in its predetermined path with the change occurring only after completion of a cycle of movement of the beam in its predetermined path following completion of the cycle of movement of the beam in its predetermined path in which comparison of the count with the count from the prior cycle of movement of the beam in its predetermined path occurred.

26. The system according to claim 17 in which: the aberration of the beam is the quality of the focus of the beam; and said correcting means includes means to change the focus of the beam in accordance with the count obtained by said counting means during a cycle of movement of the beam in its predetermined path in comparison with the count obtained by said counting means during the prior cycle of movement of the beam in its predetermined path.