JP2005077295A - Measuring apparatus for optical three-dimensional position and measuring method for position - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To realize a measuring apparatus for an optical three-dimensional position capable of detecting a position mark with position accuracy of the order of submicron. <P>SOLUTION: A camera 16 is held with a Z slide 13 that is supported on an X slide 11 and a Y slide 12, and then the X, Y, and Z slides 11-13 are moved so that the sphere center position of a position mark sphere is matched with the focal position of the camera 16 by irradiating the convergent light from an optical lever optical system 19 to the position mark sphere of an object W<SB>1</SB>to be measured. Under this condition, the three-dimensional coordinate of position mark sphere is obtained from measurement values by measuring the X, Y, and Z positions of the Z slide 13 with a laser length measuring machines 61-65. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention accurately measures the relative position and orientation of an assembly part using a position mark, for example, with submicron accuracy, and assembles precision equipment such as an electron lens of an electron beam exposure apparatus with high precision and efficiency. The present invention relates to an optical three-dimensional position measuring apparatus and a position measuring method.

Conventionally, a method using a camera is known as a method for measuring the position of an object to be measured in a non-contact manner. Such as those shown in FIG. 16, there is provided an apparatus for measuring the shape and dimensions of such elongated hole S ₁ of the object W ₀ optically, from the position where the position of the object W ₀ which is carried by the conveyor 100 is predetermined In order to be able to measure the dimension and shape of the measurement target such as the long hole S ₁ with high accuracy even if they are displaced, the television camera 102 attached to the working end of the multi-axis industrial robot 101 is used. imaging the target object, and image processing in the processing circuit 103 an image thereof captured, first detects the relative position of the TV camera 102 with respect to the measured object W _0, adjust the position. For example, the position and inclination of the long axis S ₁ of the long hole formed in the workpiece W ₀ are measured by image processing, and the slit light L ₀ (sheet-shaped light beam) from the light source 104 integrated with the TV camera 102 is measured. ) Is moved and positioned so as to pass through the center of gravity of the long hole S ₁ and in the long axis direction, and then the long hole S ₁ due to the slit light L ₀ and other An image of the hole is picked up by the TV camera 102, and the opening size and the like are measured.

  As described above, a technique for detecting the position of an object to be measured by image processing is introduced in known patent documents.

  However, in the above apparatus, since the camera is mounted on the robot arm, the positioning accuracy of the object to be measured is affected by the positioning error of the robot arm. In particular, the positioning accuracy of an articulated robot arm as shown in FIG. 16 is generally not high. Even if this robot arm is changed to one with very high accuracy, it is difficult to reduce the influence of mechanical errors, for example, to make the positional accuracy several μm in the measurement region of 300 mm, for example, on the order of submicron. It is technically very difficult to realize the high accuracy of this, and even if it is possible, an ultra-high precision machine element such as an air bearing is required, and the manufacturing cost of the device becomes high.

  In addition, it is difficult to accurately measure the three-dimensional position of the object to be measured by processing an image obtained by imaging a measurement target such as a long hole with a camera. In more detail, the method of capturing the measurement object with the camera, performing image processing, measuring the position and orientation of the object to be measured, adjusting the camera position, and then measuring the shape of the measurement object is simple. Although it is possible to perform highly accurate shape measurement compared to using an image obtained by capturing the measurement object with a camera, the measurement accuracy depends on the camera's position and orientation because it relies on image processing. Limited by imaging performance. However, a high-capacity imaging performance of a camera using light cannot be expected, and since the resolution of the camera is at most several μm, it is difficult to achieve high accuracy on the order of submicrons.

In addition, it is difficult to measure the distance in the optical axis direction of the camera by a method in which a measurement target is imaged with a camera, image processing is performed, and the position and orientation are measured. Although it is conceivable to use the autofocus mounted on the camera, it is difficult to improve the accuracy with autofocus based on image processing.
JP-A-5-045117

  The present invention has been made in view of the above-mentioned unsolved problems of the prior art. For example, an optical three-dimensional position measurement capable of three-dimensionally reading a position mark with high accuracy on the order of submicrons. An object is to provide an apparatus and a position measurement method.

  In order to achieve the above object, an optical three-dimensional position measuring apparatus of the present invention includes a base having support means for supporting an object to be measured having a position mark, and an XY that moves two-dimensionally on the base. A stage, a Z stage that is vertically movable on the XY stage, an optical position mark detection means that is held by the Z stage and moves together with the Z stage to detect the position mark, and the Z stage An X laser length measuring device for measuring the position in the X direction, a Y laser length measuring device for measuring the position in the Y direction of the Z stage, and a Z value for measuring the position in the Z direction of the Z stage. Based on the measured values of the laser length measuring device and the X, Y and Z laser length measuring devices when the position mark is detected by the optical position mark detection means, the three-dimensional position mark It characterized by having a calculating means for calculating the location.

  In addition, the position measuring method of the present invention is a position measuring method for capturing an object to be measured having a position mark with a camera held on a three-dimensional stage on a base, and the X constituting the three-dimensional stage. , Y, and Z stages are moved to align the image of the position mark of the measured object with the optical axis position of the camera held on the Z stage, and with the position mark image aligned with the optical axis position of the camera. Measuring the X, Y, and Z positions of the Z stage with X, Y, and Z laser length measuring devices, respectively, and calculating the three-dimensional position of the position mark based on the respective measured values. .

  Since the measurement value of the position and orientation of the object to be measured is obtained using the laser length measuring device, the measurement accuracy is high and the sub-nanometer resolution is possible. For example, the conventional method that relies on the movement accuracy of the robot arm can improve the accuracy of several μm at the most by three orders of magnitude or more.

  In addition, in the configuration in which the reflected light of the reference mirror is detected by the X, Y, and Z laser length measuring devices and the three-dimensional coordinate position is measured, instead of using three mirrors, one mirror is mounted in the biaxial direction. By using two mirrors facing the laser length measuring device, the reference mirror placed above the measurement point is unnecessary, and in particular in application fields where large cameras are mounted, the mirror arrangement is not important. This makes it possible to mount a camera without having to make a significant contribution to downsizing the apparatus and improving workability.

FIG. 1 shows an optical three-dimensional position measuring apparatus according to an embodiment, and a base having support means for supporting an object W ₁ provided with a position mark ball 4 (see FIG. 2) as a position mark. A support table 10, an XY stage including an X slide 11 and a Y slide 12 that move two-dimensionally on the support table 10, and a Z slide 13 that is a Z stage that is vertically movable on the XY stage; The XYZ controller 14, the computer 15, the camera 16 held on the Z slide 13, the monitor 17 that outputs an image captured by the camera 16, the image processing device 18, and the camera 16 constitute an optical position mark detection means. The optical lever optical system 19 to be measured, X laser length measuring devices 61 and 62 for measuring the position of the Z slide 13 in the X direction, and the position of the Z slide 13 in the Y direction are measured. And Y laser length measuring device 63, 64 for, and a Z laser length measuring machine 65 for measuring the position in the Z direction of the Z slide 13 (see FIG. 8).

The computer 15 detects the position mark based on the measured values of the X, Y, and Z laser length measuring devices 61 to 65 when the position mark sphere 4 of the workpiece W ₁ is detected by the optical position mark detecting means. Calculation means for calculating the three-dimensional position of the sphere 4 is provided.

FIG. 2 shows the configuration of the workpiece W ₁ according to the first embodiment. This is a part of an electron beam exposure apparatus called an electron lens, and a lens action is generated by lens openings 3a and 3b provided in the first and second plates 2a and 2b, respectively. In the assembly of the electron lens, two position mark balls 4 for aligning the lens openings 3a and 3b are attached to the first and second base plates 5a and 5b, respectively. The plates 5a and 5b are fixed with the spacer 6 therebetween.

  The first plate 2a is fixed to the first base plate 5a by a plate spring 7a. Similarly, the second plate 2b is fixed to the second base plate 5b by a plate spring 7b. As described above, the lens openings 3a and 3b through which electrons pass are opened. A control electrode (not shown) is formed on the electron lens, and exhibits a lens action when the electron beam passes through the lens openings 3a and 3b.

  What is important at this time is that the first lens opening 3a and the second lens opening 3b are exactly on the vertical line. If this is not the case, the electron beam will not pass through the center of the lens, so that precise lens action cannot be achieved. Therefore, it is necessary to precisely align and fix the horizontal direction of the first and second base plates 5a and 5b.

  The number of base plates is not limited to two, but may be three or more. Here, for the sake of simplicity, description will be made with only two sheets.

  A position mark sphere 4 that is a measurement target for precisely measuring these positions is fixed to each of the base plates 5a and 5b. For example, as shown in FIG. 2, a cylindrical hole is formed, and the ball is bonded and fixed with an adhesive that can be used in a vacuum. The upper surface of the position mark sphere 4 is a spherical surface. For example, if a steel ball used for a bearing is used, a high accuracy of 50 nanometers is currently available. Further, an optical element called a ball lens provided with a reflective film can be used. As shown in FIG. 1, the three-dimensional position of the position mark sphere 4 is determined by the camera 16 and the optical lever optical system 19 and the X, Y, Z laser length measuring devices 61 to 65 held on the Z slide 13 on the XY stage. It measures using the optical three-dimensional measuring apparatus which consists of. FIG. 2 shows a state where the measurement point M at the focal position of the camera 16 and the sphere center position of the position mark sphere 4 coincide with each other.

  One or more position mark balls 4 of the base plates 5a and 5b are provided. For example, if two are provided, not only the position in the horizontal direction but also the posture can be measured, so that the accuracy can be improved. If three more are provided, all six axes of position and orientation can be measured. Here, only two position mark spheres are shown for simplicity of explanation, but the configuration and operation flow are almost the same even when three or more position mark spheres are used.

  Each position mark sphere 4 measures the relative position with the lens openings 3a and 3b in advance. A microscope combined with an XY table can be used for this measurement. Since the relative positions of the lens openings 3a and 3b and the position mark sphere 4 are measured in this way, the position of the lens openings 3a and 3b can be determined by measuring the position of the position mark sphere 4.

  A position adjustment piece 9 is fixed to the first base plate 5a, and the position of the second base plate 5b can be finely adjusted in the horizontal direction. This fine adjustment can be automatically adjusted by a method such as controlling this screw with a pulse motor in addition to the method using a push screw as shown.

  A viewing hole 8 is formed in the second base plate 5b, and the position mark sphere 4 provided on the first base plate 5a can be measured from above.

The object to be measured W ₁ is set on the support base 10 of the apparatus of FIG. An X slide 11 is provided on the support base 10 so as to be movable in the X direction. A Y slide 12 is provided on the X slide 11 and is movable in the Y direction. The Y slide 12 is movable on the Y slide 12 in the Z direction. A Z slide 13 is provided. The three-dimensional stage including the X, Y, and Z slides 11 to 13 does not need to be highly accurate. Therefore, it can be manufactured at a low cost by a combination of a normal rolling guide and a ball screw.

  A three-axis motor that drives the X, Y, and Z slides 11 to 13 is connected to an XYZ controller 14, and the XYZ controller 14 is connected to a computer 15. The computer 15 is provided with a manual operation panel 15a so that the XYZ motors can be moved manually.

  A camera 16 is fixed to the Z slide 13, and an image output from the camera 16 is connected to a monitor 17 for confirmation and an image processing device 18. The image processing device 18 can process the image of the position mark sphere 4 photographed by the camera 16 and calculate the center position (ball center) of the position mark sphere 4. Since the position is subjected to image processing, the measurement error is relatively large as described above.

  An optical lever optical system 19 is provided fixed to the Z slide 13. Further, the half mirror 20 is adjusted and fixed so that the optical axis of the camera 16 and the optical axis of the optical lever optical system 19 coincide with each other.

  FIG. 3 illustrates the camera 16 and the optical lever optical system 19. The camera 16 photographs the position mark sphere 4 via the half mirror 20 and the quarter-wave plate 25. This image is used to measure the approximate position of the position mark sphere 4. The light emitted from the semiconductor laser 21 is condensed at one point by the condenser lens 21 a and introduced into the optical fiber 22. The light emitted from the optical fiber 22 reflects only the linearly polarized P wave at the polarization beam splitter 23 and is converted into convergent light by the condenser lens 24. Then, the light is reflected by the half mirror 20, passes through the quarter-wave plate 25, changes to circularly polarized light, is reflected by the position mark sphere 4, passes through the quarter-wave plate 25 again, and this time, Changes the polarization direction by 90 degrees. Then, it is reflected by the half mirror 20 and enters the polarization beam splitter 23 again, but is polarized in a direction different by 90 degrees from the previous one, so that it passes without being reflected by the polarization beam splitter 23.

Next, the light incident on the half mirror 26 and reflected is incident on the position sensor 27. The position sensor 27 is a kind of a photodiode which is a part made of a silicon semiconductor, and the current flowing through the four terminals varies depending on the location where the light hits. At this time, the position sensor 27 is arranged at the focal position. This focal position is determined by the focal length of the condensing lens 24, the position of the optical fiber exit end, and the position of the position mark sphere 4. As shown in FIG. 4, the exit position P ₁ of the optical fiber 22 and the position of the condensing lens 24 are obtained. L, a focal length f1 of the condenser lens 24, the focal position P ₂ is assuming that matches the sphere center position mark ball 4, the distance d1, d2 between the respective components from the official lens is expressed by the following formula The

1 / d1 + 1 / d2 = 1 / f1
When the focal point P ₂ and the sphere center of the position mark sphere 4 coincide with each other, the light once condensed at the focal point P ₂ returns to the optical path as it is. Then, the focal point is again focused at the emission position P ₁ of the optical fiber 22, but a position sensor 27 is arranged at the condensing position via the half mirror 26. However, the position of the position sensor 27 does not need to be strictly adjusted. Since the position sensor 27 outputs a current signal according to the position of the center of gravity of the light amount of the incident light spot, the position signal of the light spot is output if the size of the light spot is smaller than the position sensor 27 even if it is not strictly focused. be able to. Therefore, instead of intentionally shifting the position sensor 27 from the focal position in the optical axis direction and increasing the size of the light spot, the sensitivity to the position of the position mark sphere can be adjusted.

  In FIG. 3, when the difference between the current signals in the horizontal direction is calculated by the subtraction circuit 28, a signal representing the horizontal position of the place where the light hits is obtained, and when the difference between the current signals in the vertical direction is calculated by the subtraction circuit 29, A signal representing the vertical position of the place where the light strikes is obtained, and when all four current signals are calculated by the addition circuit 30, a signal representing the amount of light striking the position sensor 27 and the total amount of light is obtained. The output signals of the subtraction circuit 29 are divided by the division circuits 31 and 32 to obtain a horizontal position signal φ1 and a vertical position signal φ2.

  When dividing by the total light quantity in this way, even if the light quantity incident on the position sensor 27 changes, no measurement error occurs. The total light quantity signal is guided to the comparison circuit 33 and compared with a predetermined voltage. When the voltage does not reach, that is, when the amount of light incident on the position sensor 27 does not reach a predetermined value, an error signal e12 is output.

On the other hand, the light that has traveled straight without being reflected by the half mirror 26 passes through the cylindrical lens 34 and enters the quadrant photodiode 35. This part is autofocus using so-called astigmatism. FIG. 5 illustrates the operation of this portion. Since the cylindrical lens 34, that is, the cylindrical lens does not have a curvature in the axial direction, the direction is focused on the same place as described for the position sensor 27. That is, it is the focal position P ₂ in FIGS. However, in the direction having the curvature of the cylindrical lens, the focal length is shortened accordingly. That is, the focal position P ₄ in FIG. At these two focal positions P ₂ and P ₄ , the light spot is focused only in one direction, so that it becomes a linear spot shape as shown in the figure. The spot shape between them is an intermediate shape between the two, and a balanced circular spot should be obtained at some position as shown in the figure. This position is P ₃ as shown in FIG. The shape of the spot can be measured by the four-divided photodiode 35 as follows.

  The quadrant photodiode 35 is a kind of photodiode that is a part made of a silicon semiconductor, and the current that flows changes according to the amount of light at each of the quadrants. That is, if the difference between the current output of the photodiode divided in the vertical direction and the current output of the photodiode divided in the horizontal direction is calculated, a signal representative of the spot shape can be obtained. Further, the position of the center of the position mark sphere 4 when this balance is achieved is defined as a measurement point M. When the position mark sphere 4 approaches or moves away from the measurement point M, the focal position moves accordingly. This is a mechanism for changing the shape of the spot on the quadrant photodiode.

Returning to FIG. 3, the lateral light amount of the four-divided photodiode 35 is calculated by the adding circuit 36, and the vertical light amount of the four-divided photodiode 35 is calculated by the adding circuit 37. When the difference between the two is calculated by the subtraction circuit 39, a signal corresponding to the spot shape described above is obtained. A zero signal in which the horizontal and vertical signals are balanced corresponds to P ₃ in FIG. Further, when all four current signals are calculated by the addition circuit 38, a signal indicating the amount of light hitting the four-division photodiode 35 and the total light amount is obtained, and the division circuit 40 divides the output signal of the subtraction circuit 39. Then, a focus signal φ3 is obtained. When dividing by the total light quantity in this way, even if the light quantity incident on the quadrant photodiode 35 changes, no measurement error occurs. Further, the signal of the total light quantity is guided to the comparison circuit 41 and compared with a predetermined voltage. When the voltage does not reach, that is, when the amount of light incident on the four-divided photodiode 35 does not reach a predetermined value, an error signal e3 is output.

  The signals φ1, φ2, e12, φ3, and e3 output from the optical lever optical system 19 described above are input to the computer 15, and the image center of the position mark sphere 4 coincides with the optical axis position of the camera 16 with high accuracy. The X, Y, Z slides 11 to 13 are moved until the three-dimensional coordinates of the position mark sphere 4 are obtained from the measured values of the X, Y, Z laser length measuring devices 61 to 65 when the state shown in FIG. obtain.

  As shown in FIGS. 1 and 6, a column 42 is fixed to the support base 10, and a measurement box 46 is attached to the column 42 via a fixed spacer 43, a linear guide 44, and a ball 45. That is, the first point of the measurement box 46 is firmly fastened via the fixed spacer 43, and the second point is attached via the linear guide 44. At this time, the linear guide 44 is moved in the direction of the fixed spacer 43. The third point is fixed via the ball 45. Here, the ball 45 receives a load in the vertical direction but acts as a bearing that moves freely in the horizontal direction. That is, this point is fixed in the vertical direction but can move freely in the horizontal direction.

  When supported at three points in this way, the measurement box 46 is supported so as not to be deformed no matter how the column 42 is deformed. This is because even if the relative distance between the first, second, and third points changes, the linear guide 44 and the ball 45 move, and the force is not transmitted to the measurement box 46.

  An X mirror 47 and a Y mirror 48 are fixed to the measurement box 46 by the method shown in FIG. What is important in fixing the mirrors 47 and 48 is to support the mirror so that the plane of the mirror mirror surface, which is a reflection surface, is maintained. For this purpose, it is important to take care not to apply an unnecessary bending moment to each of the mirrors 47 and 48. Therefore, the abutment piece 50 is provided at three locations for each mirror. FIG. 7 shows two of these. The plunger 49 is pressed directly behind the butting piece 50. The plunger 49 is a part in which a spring is charged, and can press the article with a constant force. What is important at this time is that the plunger 49 is arranged directly behind the abutment piece 50. By doing so, an unnecessary bending moment is not applied to each of the mirrors 47 and 48, so that the accuracy of the mirrors 47 and 48 can be prevented from deteriorating. For fixing the X mirror 47 in the vertical direction, the butting pieces 53 are also provided at both ends. However, the plunger is not necessary because it is pressed against the counterpiece by its own weight.

  FIG. 8 illustrates the arrangement of the five laser length measuring devices 61 to 65 and the X and Y mirrors 47 and 48. Each of the laser length measuring devices 61 to 65 can measure the distance with high accuracy such as sub-nanometers. It is a precision measuring part. The X direction is measured by X1 and X2 laser length measuring devices 61 and 62. The Y direction is measured by Y1, Y2 laser length measuring instruments 63, 64. The Z direction is measured by the Z1 laser length measuring device 65. Therefore, for the X mirror 47, the surface in the X direction and the surface in the Z direction are mirror surfaces. The Y mirror 48 is mirrored only in the Y direction.

  The X1, X2 measuring axis and the Z1 measuring axis are arranged in the same vertical plane, and the Y1, Y2 measuring axes are also arranged in the same vertical plane. The optical lever optical system 19 is disposed so as to coincide with the optical axis. Further, as shown in the figure, the X1 and Y1 length measurement axes are also in the same horizontal plane, and similarly, the X2 and Y2 length measurement axes are also arranged in the same horizontal plane.

  The distance L1 is the distance in the Z direction of the X2 measurement axis and the Y2 measurement axis, and the X1 measurement axis and the Y1 measurement axis, and the distance L2 is the Z2 to the measurement point M. When the distance in the direction is set, L3 is the distance from the optical axis of the camera 16 to the Z1 laser length measuring device 65, and the sign output from each laser length measuring device 61-65 is a positive length, the three-dimensional measurement point M The XYZ coordinates that are the coordinates are calculated by the following formula.

-X = X1 + (X2-X1) * (L1 + L2) / L1
-Y = Y1 + (Y2-Y1) * (L1 + L2) / L1
Z = Z1 + (X2-X1) * L3 / L1

  When the Z slide 13 moves, a large position / posture error occurs due to the influence of the X and Y slides 11 and 12 below, but the influence can be eliminated by calculating the above equation. For example, if a rotation error around the X axis occurs, the rotation angle is (Y2−Y1) / L1. The deviation in the Y direction of the measurement point M when this rotation occurs is (Y2−Y1) * (L1 + L2) / L1. This is the meaning of the second term of the above formula in the Y coordinate. In the above description, the X, Y, and Z coordinate offsets are omitted for the sake of simplicity. It is obvious that an arbitrary position can be made the origin, that is, X = Y = Z = 0 by adding an offset constant to the above equation.

  Next, the operation of the optical three-dimensional position measuring apparatus of FIG. 1 will be described based on the flowchart of FIG. All these operations are controlled by a program of the computer 15.

First, the counter representing the number of repetitions is reset to zero, and the object to be measured W ₁ is set in the apparatus (step 100). Next, the approximate position of the first position mark sphere 4 is taught to the apparatus (step 101). The details of this operation are as follows. First, the XYZ stage position is adjusted using the manual operation panel 15a while checking the image of the camera 16 on the monitor 17. At this time, the position mark sphere 4 is at the center of the screen of the monitor 17 and is in focus. That is, since a sphere can be seen at the center of the screen of the monitor 17, the position mark sphere 4 is on the camera optical axis. In addition, since the focus is almost the same, the position mark sphere 4 is almost at the measurement point M in the vertical direction.

  Next, it is confirmed that the error signals e12 and e3 of the optical lever optical system 19 are not generated (step 102). If there is an error, the optical axis of the camera 16 and the optical axis of the optical system 19 are not aligned, or a failure of a component such as a laser diode is considered.

  Subsequently, the X and Y axes (X, Y slides 11 and 12) are moved so that the signals φ1 and φ2 become zero, and the Z axis (Z slide 13) is moved so that the signal φ3 becomes zero ( Step 103). By this operation, the light spot reflected from the position mark sphere 4 is automatically adjusted at the center of the position sensor 27 to the center of the autofocus position by the four-division photodiode.

The position of the Z slide 13 at this time indicates that the center of the position mark sphere 4 coincides with the measurement point M, that is, the focal position on the optical axis of the camera 16. The sensitivity of the optical lever 19 is high, and in particular, high accuracy such as submicron can be obtained relatively easily in the horizontal direction that is the output of the position sensor 27. This is because the sensitivity can be easily increased by reducing the size of the position mark sphere 4 or increasing the length of the light lever. That is, in the arrangement shown in FIG. 4, the change in position of the focal position P ₁ is multiplied by d2 / d1 and appears at the focal position P ₂ . Increasing this magnification will increase sensitivity. For example, when the size of the position sensor 27 is 4 mm and the signal / noise ratio of the analog amplifier is 1000, a resolution of 4 μm can be obtained. If the magnification of the light lever is 10, a resolution of 0.4 μm can be obtained.

  Next, the values of the laser length measuring devices 61 to 65 are read (step 104), the XYZ coordinates are calculated by the method described above, displayed on the monitor 17, and recorded in the memory of the image processing device 18 (step 105).

  It is confirmed whether all the position mark spheres 4 on the base plates 5a and 5b have been measured (step 106). If not, since there is a position mark sphere that has not been measured yet, the process returns to step 101. Then, the position and orientation of all the parts on which the position mark sphere is installed, that is, the base plates 5a and 5b are calculated.

  As described above, when one position mark sphere is attached to one part, only the XYZ position of the part can be measured. In this case, even if a rotation error occurs around the vertical axis, it is not known. When two position mark balls are attached to one part, the rotation angle around the vertical axis can be measured together with the XYZ position of the part. When three position mark spheres are attached to one part, six degrees of freedom of the position and orientation of the part can be determined.

  The amount of correction of the position of the base plates 5a and 5b is determined from the parts thus measured, that is, the positions and orientations of the base plates 5a and 5b (step 108). If the correction amount is within the predetermined target accuracy, the process ends (step 109). Otherwise, 1 is added to the counter of the number of position adjustments (step 110). If the predetermined number of repetitions has been reached, it is determined that it has not converged, an error is displayed, and the process is stopped (step 111).

  The bolts connecting the two base plates 5a and 5b are loosened, the position adjusting piece 9 is rotated to adjust the position according to the correction amount, and the bolts are tightened again to fix the base plates 5a and 5b. (Step 112). Then, the process returns to step 101 again.

  Further, teaching can be omitted after the measurement operation described above has been performed once or more. That is, it is possible to omit the step 101 and automatically move the XYZ slides 11 to 13 to a position where the teaching is completed to accurately measure the position of the position mark sphere 4. In addition, although the signal processing from each sensor was demonstrated using the symbol of the analog arithmetic circuit, it is the same even if this is directly analog-digital converted and calculated inside the computer.

  Further, it is known that when measuring the position with very high accuracy, the wavelength of the laser changes under the influence of atmospheric pressure or temperature change. For this, it is effective to add a wavelength tracker and correct the wavelength. This wavelength tracker is a well-known technology that is already available for purchase.

  FIG. 10 shows the second embodiment. Only the five laser length measuring devices 71 to 75 fixed to the Z slide 13 are different from the first embodiment, and the remaining members are the same, and are therefore denoted by the same reference numerals. Is omitted.

  In this embodiment, the X mirror 47 is a mirror whose surfaces in the X direction and the Z direction are mirror surfaces, and the Y mirror 48 is a mirror whose mirror surface is in the Y direction. The X direction is measured by X1 and X2 laser length measuring instruments 71 and 72. The Y direction is measured by the Y1 laser length measuring device 73. The Z direction is measured by Z1 and Z2 laser length measuring instruments 74 and 75. Also, the X1, X2 measurement axis and the Z1 measurement axis are arranged in the same vertical plane, the X1, Y1 measurement axis is arranged in the same horizontal plane, and the Z1, Z2 measurement axes are arranged apart from each other in the Y direction. To do.

  The distance L1 is the distance in the Z direction of the X2 measurement axis and the X1 measurement axis, the distance L2 is the distance in the Z direction to the X2 measurement axis and the measurement point M, the optical axis of the camera 16 and the Z1 laser measurement XYZ coordinates of the measurement point M are given as a distance L3 up to the measuring device 74, a distance L4 between the Z1 measuring axis and the Z2 measuring axis, and the sign output from each laser measuring device 71-75 is a positive length. It is calculated by the following formula.

-X = X1 + (X2-X1) * (L1 + L2) / L1
-Y = Y1 + (Z2-Z1) * (L1 + L2) / L4
Z = Z1 + (X2-X1) * L3 / L1

  When the Z slide 13 moves, a large position / posture error occurs due to the influence of the X and Y slides 11 and 12 below, but the influence can be eliminated by calculating the above equation. For example, if a rotation error around the X axis occurs, the rotation angle is (Z2−Z1) / L4. The deviation in the Y direction of the measurement point M when this rotation occurs is (Z2−Z1) * (L1 + L2) / L4. This is the meaning of the second term of the above formula in the Y coordinate.

  In the above description, the X, Y, and Z coordinate offsets are omitted for the sake of simplicity. It is obvious that an arbitrary position can be made the origin, that is, X = Y = Z = 0 by adding an offset constant to the above equation. In this embodiment, there is an advantage that the Y mirror 48 can be made smaller than in the first embodiment.

  FIG. 11 shows the third embodiment. Only the five laser length measuring devices 81 to 85 fixed to the Z slide 13 are different from the first embodiment, and the remaining members are the same, and are therefore denoted by the same reference numerals. Is omitted.

  In this embodiment, the X mirror 47 is a mirror whose surfaces in the X direction and the Z direction are mirror surfaces, and the Y mirror 48 is a mirror whose mirror surface is in the Y direction. The X direction is measured by the X1 laser length measuring device 81. The Y direction is measured by the Y1 laser length measuring device 82. And about Z direction, it measures with Z1, Z2, Z3 laser length measuring devices 83-85. Also, the X1 measuring axis and the Z1 and Z3 measuring axes are arranged in the same vertical plane, the X1 and Y1 measuring axes are arranged in the same horizontal plane, and the Z1 and Z2 measuring axes are arranged apart from each other in the Y direction. To do.

  The distance L1 is the distance in the Z direction from the X1 measuring axis to the measuring point M, the distance L3 is the distance from the optical axis of the camera 16 to the Z3 laser measuring instrument 83, and between the Z1 measuring axis and the Z2 measuring axis. Is the distance L4, the distance between the Z2 measuring axis and the Z3 measuring axis is the distance L5, and the sign output from the laser length measuring devices 81-85 is a positive length, the XYZ coordinates of the measuring point M are as follows: Calculated.

-X = X1 + (Z3-Z1) * L1 / L5
-Y = Y1 + (Z2-Z1) * L1 / L4
Z = Z1 + (Z3-Z1) * (L3 + L5) / L5

  When the Z slide 13 moves, a large position / posture error occurs due to the influence of the X and Y slides 11 and 12 below, but the influence can be eliminated by calculating the above equation. For example, if a rotation error around the X axis occurs, the rotation angle is (Z2−Z1) / L4. And the deviation of the measurement point M in the Y direction when this rotation occurs is (Z2-Z1) * L1 / L4. This is the meaning of the second term of the above formula in the Y coordinate.

  In the above description, the X, Y, and Z coordinate offsets are omitted for the sake of simplicity. It is obvious that an arbitrary position can be made the origin, that is, X = Y = Z = 0 by adding an offset constant to the above equation. Also in this embodiment, there is an advantage that the Y mirror 48 can be made smaller than in the first embodiment.

  FIG. 12 shows an electron lens that is an object to be measured in Example 4. Instead of the position mark sphere 4 of the electron lens in Example 1, a concentric Fresnel pattern 94 that is a position mark shown in FIG. 13 is used. It was. By irradiating such a concentric pattern with light of a single wavelength, such as a semiconductor laser, and diffracting it, the same action as a position mark sphere is possible.

  The electron lens shown in FIG. 12 has the same configuration as that shown in FIG. That is, the second base plate 5b is fixed with the spacer 6 sandwiched between the first base plate 5a. A first plate 2a is fixed to the first base plate 5a by a leaf spring 7a, and a lens opening 3a through which electrons pass is formed in the plate 2a. Similarly, a second plate 2b is fixed to the second base plate 5b by a leaf spring 7b, and a lens opening 3b through which electrons pass is formed in the plate 2b. A control electrode (not shown) is formed on the electron lens, and exhibits a lens action when the electron beam passes through the lens openings 3a and 3b.

  What is important at this time is that the first-stage lens opening 3a and the second-stage lens opening 3b are exactly on the vertical line. If this is not the case, the electron beam will not pass through the center of this hole, so the precise lens action cannot be achieved. Therefore, it is necessary to precisely align and fix the horizontal direction of the first and second base plates 5a and 5b.

  Therefore, a Fresnel pattern 94 is formed on each plate 2a, 2b. This has the same effect as the spherical surface of the position mark sphere 4 described above. At this time, mainly two arrangements are conceivable depending on the order of diffraction. As shown in FIGS. 14A and 14B for the Fresnel pattern 94, the measurement point M is detected on the upper side and the lower side of the Fresnel pattern 94, so that the position on the upper side, that is, FIG. The measurement point M is measured by the arrangement, and then the measurement point M is measured by the lower position, that is, the arrangement of FIG. . In this way, the center of the Fresnel pattern 94 can be measured accurately even if the plates 2a and 2b forming the Fresnel pattern 94 are inclined. In this embodiment, compared to the first embodiment, since the Fresnel pattern 94, which is a position mark, can be arranged closer to the lens openings 3a and 3b of the electron lens, an improvement in accuracy can be expected.

  Further, since the Fresnel pattern can be manufactured by applying a semiconductor lithography process simultaneously with the manufacturing of the lens aperture of the electron lens, the relative position of both can be manufactured with high accuracy. That is, in Example 1, it was necessary to measure the relative positions of the position mark spheres 4 and the lens openings 3a and 3b in advance using a microscope or the like, but this is not necessary in this example.

  Instead of the concentric Fresnel pattern 94 shown in FIG. 13, strip-like Fresnel patterns 95 and 96 in the horizontal and vertical directions as shown in FIG. 15 may be used. By irradiating these two strip-shaped Fresnel patterns 95 and 96 with light of a single wavelength such as a semiconductor laser and diffracting them, the same action as a cylindrical surface can be achieved. The patterns 95 and 96 may be considered as cylindrical surfaces in the vertical and horizontal directions, respectively.

  That is, three such Fresnel patterns and three reflection planes are used for the position mark. When the Fresnel pattern in the X direction is used, the position in the X direction can be measured, and when the Fresnel pattern in the Y direction is used, the position in the Y direction can be measured. The Z direction is measured using a reflection plane having no Fresnel pattern. Further, since each Fresnel pattern 95, 96 can be measured at the upper and lower measurement positions in the same manner as the concentric Fresnel pattern 94, its midpoint is calculated and set as a measurement point M of the position mark.

  Since the strip-shaped Fresnel pattern is made of a straight line, it is easy to manufacture a mask used for semiconductor lithography. Therefore, there are advantages that the manufacturing cost can be reduced and the manufacturing accuracy can be improved.

  In each of the above embodiments, since the position measurement is performed using the laser length measuring device with the two mirrors as the position reference, it is possible to perform the position measurement with high accuracy. Further, since the laser length measuring device is responsible for the accuracy of position measurement, the X slide, Y slide, and Z slide that are moving members do not need to be highly accurate. Therefore, the device manufacturing cost can be suppressed.

  Also, by changing the arrangement of the laser length measuring device, the vertical size of one mirror, for example, the Y-side mirror, can be reduced, or the vertical size of the two mirrors can be reduced.

  Further, the position mark can be a spherical surface of the position mark sphere, and the position of the sphere center in the X and Y directions can be read with an optical lever optical system, so that the position can be measured very precisely. The optical axis direction, that is, the Z position can also be measured with high accuracy.

  Alternatively, by forming the position mark with a Fresnel pattern, the position mark can be manufactured by a semiconductor lithography process, and a part manufactured by the semiconductor process can be manufactured simultaneously with the part and the position mark. Therefore, the relative positional accuracy between the part and the position mark can be improved, and the manufacturing cost can be reduced.

  Further, the use of a linear Fresnel pattern makes it easier to manufacture the position mark, and the manufacturing cost of the object to be measured and the position mark can be reduced.

  Also, the mirror holding member and the base are fixed at three locations, one point is fixed with a fixed spacer, and one point is fixed with a linear guide with the moving axis along the direction of the spacer. At one point, by fixing the sphere with a flat plate, an optical three-dimensional position measurement device is realized that does not cause the measurement accuracy to deteriorate because the relative relationship of the mirror, which is the measurement reference, remains unchanged even if the support base is deformed. it can.

1 is a schematic diagram showing an optical three-dimensional position measuring apparatus according to Example 1. FIG. It is a figure explaining the structure of the electron lens which is a to-be-measured object. It is a figure explaining the structure of the optical lever optical system of the apparatus of FIG. It is a figure explaining the optical principle of an optical lever optical system. It is a figure explaining the autofocus part of an optical lever optical system. It is a figure explaining the position measurement part of the apparatus of FIG. It is a figure explaining attachment of the mirror of the apparatus of FIG. It is a figure explaining arrangement | positioning of the laser length measuring device of the apparatus of FIG. 3 is a flowchart illustrating an operation flow according to the first embodiment. It is a figure explaining arrangement | positioning of the laser length measuring device by Example 2. FIG. It is a figure explaining arrangement | positioning of the laser length measuring device by Example 3. FIG. FIG. 6 is a diagram illustrating a configuration of an electron lens according to Example 4. It is a figure which shows a concentric circular Fresnel pattern. It is a figure explaining the relationship between a Fresnel pattern and a to-be-measured object. It is a figure which shows two strip-shaped Fresnel patterns. It is a figure explaining a prior art example.

Explanation of symbols

2a, 2b Plate 3a, 3b Lens opening 4 Position mark ball 5a, 5b Base plate 6 Spacer 7a, 7b Plate spring 8 Peep hole 9 Position adjustment piece 10 Support base 11 X slide 12 Y slide 13 Z slide 14 XYZ controller 15 Computer 15a Manual operation panel 16 Camera 17 Monitor 19 Optical lever optical system 20, 26 Half mirror 21 Semiconductor laser 21a, 24 Condensing lens 22 Optical fiber 23 Polarizing beam splitter 25 Quarter wave plate 27 Position sensor 28, 29, 39 Subtraction circuit 30 , 36, 37, 38 Addition circuit 31, 32, 40 Division circuit 33, 41 Comparison circuit 34 Cylindrical lens 35 4-division photodiode 42 Column 43 Fixed spacer 44 Linear guide 45 Ball 46 Measurement box 47 X mirror 48 Y mirror 49 Plunger 50, 53 Butting piece 61, 71, 81 X1 laser length measuring device 62, 72 X2 laser length measuring device 63, 73, 82 Y1 laser length measuring device 64 Y2 laser length measuring device 65, 74, 83 Z1 laser length measuring device 75, 84 Z2 laser length measuring device 85 Z3 laser length measuring device 94, 95, 96 Fresnel pattern

Claims (12)

  A base having support means for supporting an object to be measured having a position mark; an XY stage that moves two-dimensionally on the base; a Z stage that is vertically movable on the XY stage; Optical position mark detection means that is held by a Z stage and moves with the Z stage to detect the position mark, an X laser length measuring device for measuring the position of the Z stage in the X direction, and the Z stage The position mark is detected by the Y laser length measuring device for measuring the position in the Y direction, the Z laser length measuring device for measuring the position in the Z direction of the Z stage, and the optical position mark detecting means. And an arithmetic means for calculating a three-dimensional position of the position mark on the basis of the measured values of the X, Y, and Z laser length measuring devices. Location measurement device.   A first mirror having a reflecting surface facing the X laser length measuring device and a second mirror having a reflecting surface facing the Y laser length measuring device are provided, and the first mirror and the second mirror are provided. 2. The optical three-dimensional position measuring apparatus according to claim 1, wherein a reflective surface facing the Z laser length measuring device is formed on one of the side surfaces.   3. The optical three-dimensional position measurement according to claim 1, wherein two X laser length measuring devices and two Y laser length measuring devices are provided for measuring at two positions facing each mirror. apparatus.   2. One of the X laser length measuring device and the Y laser length measuring device and two Z laser length measuring devices are provided for measuring at two positions facing each mirror. Or the optical three-dimensional position measuring apparatus according to 2;   3. An optical three-dimensional position measuring apparatus according to claim 1, wherein three Z laser length measuring devices are provided for measuring at three positions facing the mirror.   6. The optical three-dimensional position measuring apparatus according to claim 1, wherein the optical position mark detecting means includes a camera for obtaining an image of the position mark.   The optical position mark detection means converts the illumination light emitted from the point light source into the first convergent light by the condensing lens, reflects it by the position mark to convert it into the diffused light, and converts the illumination light into the second converging lens. The optical three-dimensional position measurement according to any one of claims 1 to 6, further comprising an optical lever optical system that converts the light into convergent light and detects an optical axis position of the second convergent light by a position sensor. apparatus.   8. The optical position mark detecting means includes an autofocus optical system for branching the second convergent light by a half mirror and detecting a condensing position by a cylindrical lens and a quadrant diode. Optical three-dimensional position measuring device. A position measurement method for imaging an object to be measured having a position mark with a camera held on a three-dimensional stage on a base,
Moving each of the X, Y, and Z stages constituting the three-dimensional stage to align the image of the position mark of the object to be measured with the optical axis position of the camera held on the Z stage;
With the position mark image aligned with the optical axis position of the camera, the X, Y, and Z positions of the Z stage are measured by the X, Y, and Z laser length measuring devices, respectively. Calculating a dimension position;
A position measuring method characterized by comprising:   The position measuring method according to claim 9, wherein a position mark is aligned with a focal position of the camera using an optical lever optical system and an autofocus optical system.   11. The position measuring method according to claim 10, wherein the position mark is a position mark sphere whose sphere center is detected by an optical lever optical system and an autofocus optical system.   11. The position measuring method according to claim 10, wherein the position mark is a Fresnel pattern whose center position is detected by an optical lever optical system and an autofocus optical system.

Images