JP4238402B2 - Length measuring device that eliminates errors due to attitude error of moving table - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a length measuring device having a moving table that moves in at least two directions perpendicular to each other, in which an error due to a posture error of the moving table is removed from a reference position of the moving table.
[0002]
[Prior art]
A stage apparatus having a movable table movable in two perpendicular directions and a position detecting means for detecting the position of the movable table in the two perpendicular directions is used as, for example, a stage for a microscope or a projection inspector. By the means, the position observed with the objective lens of the microscope or projection inspection instrument, that is, the two-dimensional coordinate value of the observation position is obtained. Based on the two-dimensional coordinate value, the dimension of the test object placed on the stage device is obtained.
[0003]
A two-dimensional coordinate value of an arbitrary observation position is determined as a coordinate value in a coordinate having the observation position when the moving table is at a predetermined position as an origin and the two perpendicular directions as coordinate axis directions.
Now, it is difficult to move the moving table of the stage apparatus completely straight without any posture errors such as yawing, pitching, and rolling. Therefore, when the position at which the object is observed, that is, the positional relationship between the position of the objective lens and the position detection means does not satisfy the Abbe condition, in other words, the position of the objective lens and the position detection means. When the detection direction is not on a straight line in the movement direction to be measured by the moving table, the length measurement result includes an Abbe primary error.
[0004]
In particular, a moving mechanism for moving in one direction (hereinafter sometimes referred to as the Y direction) out of two perpendicular directions of the mounting table is provided on a moving mechanism for moving in the other direction (hereinafter sometimes referred to as the X direction). In the stage apparatus configured so that the entire moving mechanism in the Y direction moves in the X direction, the distance from the position detecting unit to the objective lens in the X direction may change, so that the distance may increase. As a result, a large error is likely to occur.
[0005]
In order to remove the Abbe primary error, position detection means for detecting the position in the Y direction are provided at two positions separated by a predetermined distance in the X direction, and the values Y1 detected by the two position detection means, A length measuring machine that obtains a value for the observation position by interpolation or extrapolation based on Y2 and the distance from the position detecting means to the objective lens in the X direction at that time is known (Japanese Patent Publication No. 50). -23304).
[0006]
[Problems to be solved by the invention]
In the apparatus in which the Abbe primary error is removed, the coordinate axis in the X direction, that is, the X axis, is based on the detection values Y1 and Y2 of the two position detection means when the moving base is at a predetermined reference position. The two position detection means are determined as straight lines connecting the positions of the position detection means that have detected Y1 and Y2. For this reason, if an attitude error has occurred in the moving table when determining the X axis, the determined X axis is inclined with respect to the original X axis, and a coordinate system having a squareness error is set. End up. As a result, there is a problem that the coordinate value indicating an arbitrary observation position also becomes a value including the squareness error of this coordinate system. Details thereof will be described in the description of the embodiment of the invention.
[0007]
An object of the present invention is to provide a high-precision length measuring device that eliminates the squareness error in the device that removes the Abbe primary error.
[0008]
[Means for Solving the Problems]
The apparatus of the present invention that achieves the above object has a moving table that moves in at least two directions at right angles, and first and second position detecting means for determining the position of the moving table in the first direction are perpendicular to the first direction. Each of the first and second position detecting means includes a third position detecting means for obtaining a position of the movable base in the second direction, and a third position detecting means for obtaining the position of the movable table in the second direction. In the length measuring device having a calculating means for obtaining the position in the first direction at the observation position of the moving table based on the value and the detection value of the third position detecting means, the calculating means includes the moving table The first and second position detection means detect the position of the moving base in the first direction at a plurality of positions moved in the vicinity of a reference position that is a reference for position detection in the first direction, and the plurality of positions At each position of the position Obtains a difference between the detected value of the detection value and the second detection means detecting means, a length measuring device, characterized in that the correction of the reference by the average value of the difference.
[0009]
In the above apparatus, the apparatus has a reference position signal generating means for generating a reference position signal when the moving table reaches the reference position, and the first and second position detecting means are reset by the reference position signal. Can do.
In the above apparatus, the correction can be performed as the power supply to itself is turned on.
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described, but the technical scope of the present invention is not limited thereto.
FIG. 1 is a perspective view of a measuring apparatus provided with a stage apparatus as one embodiment of the present invention. The measuring apparatus 100 includes a stage device 10 that moves a placed test object (not shown) in two directions at right angles within a horizontal plane, and an observation unit 12 that optically observes the test object. The observation unit 12 observes a test object located on the optical axis of the optical system. The stage apparatus 10 is provided with a linear encoder and a counter as position detecting means for obtaining the position of the object under observation on the optical axis, that is, the position of the measurement point, and the dimension of the object to be obtained is obtained based on the obtained position. It is done.
[0011]
FIG. 2 shows the stage apparatus 10. The stage apparatus 10 will be described in detail using this figure. The lower plate 10X of the stage device 10 is supported and guided by a base 10A (see also FIG. 1) of the measuring device 10 and is movable in the left-right direction (X direction). The upper plate 10Y of the stage device 10 is supported and guided by the lower plate 10X and is movable in the front-rear direction (Y direction). Therefore, the upper plate 10Y is movable in the X and Y directions.
[0012]
A non-illustrated linear encoder (hereinafter referred to as linear encoder X) is provided between the lower plate 10X and the base 10A. The movement signal of the lower plate 10X is detected, and the counter that receives the movement signal is the lower plate 10X. A coordinate value representing the position in the X direction is obtained. Linear encoders 22 and 24 are provided between the lower plate 10X and the upper plate 10Y. The linear encoder 22 is disposed on the left end surface of the upper plate 10Y and the linear encoder 24 is disposed on the right end surface at a distance W from each other. The linear encoders 22 and 24 detect the movement amounts of the left and right end surfaces of the upper plate 10Y. Then, the counter that receives the movement signal from the linear encoders 22 and 24 obtains a coordinate value representing the position of the upper plate in the Y direction.
[0013]
Since the upper plate 10Y moves in the X direction together with the lower plate 10X, the counter detects the two-dimensional position of the upper plate 10Y as a two-dimensional coordinate value. This two-dimensional coordinate value indicates the two-dimensional coordinate of the measurement point.
The two-dimensional coordinate values are based on the predetermined position of the linear encoder X and the position of the reference position signal generating means provided at one of the predetermined positions of the linear encoders 22 and 24, respectively.
[0014]
The counter is provided in the processing unit 30 described later. Details of the counter will be described later.
FIG. 3 is a diagram illustrating an outline of the configuration of the processing unit 30 that obtains the position of the measurement point based on the movement signals of the linear encoder X and the linear encoders 22 and 24. The processing unit 30 has the counter 21 described above. The counter 21 includes a counter 21A that receives the signal Lx of the linear encoder X, a counter 21B that receives the signal Ly1 of the linear encoder 22, and a counter 21C that receives the signal Ly2 of the linear encoder 24.
[0015]
The counter 21A is reset to zero at a predetermined position in the detection range of the linear encoder X, for example, a reference position signal from a reference position signal generator provided at the left end, and obtains an X coordinate value of a measurement point based on that position. .
The contents described above are the same for the conventional stage apparatus.
In the conventional stage apparatus, the counters 21B and 21C are simultaneously reset to zero by the reference position signal Ref from the reference position signal generating means of the linear encoder 22 or 24, and the Y coordinate value of the measurement point with the position as a reference. I was looking for. That is, the Y coordinate value of the measurement point obtained by the counters 21B and 21C is a coordinate value based on the generation position of the reference position signal, and both values are always consistent with each other.
[0016]
However, the amount of movement detected by the linear encoders 22 and 24 may be slightly different due to a posture error (yawing) accompanying the movement of the upper plate 10Y. That is, two types of Y coordinate values Y1 and Y2 are obtained for one measurement point.
Therefore, in order to obtain the Y coordinate value from which the influence of the posture error has been removed, the following processing, which is the technique described in the patent publication, is performed by the processing unit 30. That is, the coordinate values X, Y1, and Y2 obtained by the counters 21A, 21B, and 21C are respectively input to the calculation means 32, and based on the coordinate values X, Y1, and Y2 input by the calculation means 32,
Y = Y1 + X * (Y2−Y1) / W Equation 1
To obtain a coordinate value representing the position in the Y direction at the observation position of the upper plate 10Y.
[0017]
FIG. 4 is a diagram for explaining the calculation principle of the Y coordinate value of the observation position according to Equation 1. In the figure, RefA and RefB are detection positions (hereinafter referred to as reset positions) of the linear encoders 22 and 24 when the reference position signal Ref is detected. Ob is the observation position of the observation device 12. The other components and the like are the same as those in FIG. Then, the two-dimensional coordinates (X, Y) of the observation position Ob are represented by the Y-coordinate value obtained by Expression 1 and the X-coordinate value. This two-dimensional coordinate value is obtained by removing an error caused by yawing.
[0018]
Note that the counters 21A, 21B, and 21C are simultaneously reset to zero when the power of the stage apparatus is turned on, and thus the same as described above even when the zero reset based on the reference position signal is not performed.
However, as described above, there is an error due to an attitude error that cannot be removed even by the processing of Expression 1. That is, as can be seen from Equation 1, on the straight line connecting the positions where the values Y1 and Y2 of the counters 21B and 21C are equal, the Y value obtained by Equation 1 is constant regardless of the X value. In other words, the straight line is a straight line parallel to the X axis, and the straight line connecting the reset positions RefA and RefB is the X axis. When the reference position signal Ref is input, if the yawing has not occurred, the straight line is parallel to the moving direction of the lower plate 10X.
[0019]
However, the upper plate 10Y moves with yawing even at a position where the reference position signal Ref is input. This is the cause of the error that cannot be removed even by the equation 1. FIG. 5 is a diagram illustrating the detection positions of the linear encoders 22 and 24 when yawing occurs at a position where the reference position signal Ref is input.
In the figure, when the yaw of the angle Δ occurs when the reference position signal Ref is input, the reset position RefB of the linear encoder 24 by the reference position signal Ref becomes RefB ′. As a result, the straight line L1 connecting the reset positions RefA and RefB ′ of the linear encoders 22 and 24 based on the reference position signal Ref is an angle Δ with respect to the straight line L2 connecting the reset positions RefA and RefB when yawing is not occurring. Make. That is, when yawing of the angle Δ occurs when the reference position signal Ref is input, a two-dimensional coordinate system is set in which the X-axis direction is inclined by the angle Δ with respect to the left and right movement directions and the squareness error Δ is set. End up. As a result, as shown in FIG. 5, the Y coordinate value Y of the coordinate value (X, Y) of an arbitrary point whose X coordinate value is x includes an error of (Δ * x).
[0020]
Hereinafter, in the present invention, details of processing performed in addition to the processing of Expression 1 in order to remove this error will be described. In the present invention, the reference position signal Ref from the linear encoder 22 or 24 is input to the computing means 32 of the processing unit 30 as shown in FIG.
It is known that the yawing of the upper plate 10Y has a substantially normal distribution centering on yawing error = 0 from the observation of the occurrence state. Therefore, the difference y1n−y2n between the detection values y11 to y1n (value of the counter 21B) and y21 to y2n (value of the counter 21C) of the linear encoders 22 and 24 in a predetermined range near the generation position of the reference signal Ref. The average value y is obtained, and -y is added to the value of the counter 21C. That is, it is preset to (current value−y). As a result, the value of the counter 21B and the value of the counter 21C at the position where the reference position signal Ref is generated coincide on average. Note that the expected value of y is 0.
[0021]
In addition, when adding -y, it is preferable to preset so that the values of the counter 21B and the counter 21C at the position where the reference position signal Ref is generated become zero. As a result, the straight line connecting the position where the counter 21B is “0” and the position where 21C is “0”, that is, the X axis is a straight line connecting RefA and RefB in FIG. Since the expected value of y is 0, the straight line (X axis) is expected to be parallel to the left and right movement directions.
[0022]
Therefore, if the values of the counters 21B and 21C including the preset value are used for the calculation of Expression 1, the Y coordinate value of the obtained measurement point is a value from which the squareness error of the coordinate system is also removed.
The process for obtaining the average value y of the detected value differences used for the preset includes a first process and a second process, and is performed by the calculation means 32.
[0023]
In the first process, the calculation means 32 calculates the value y1 of the counters 21B and 21C every time the value of the counter corresponding to the linear encoder on the side where the reference position detection means of the linear encoder 22 or 24 is provided changes by a predetermined amount. , Y2 is read to obtain the difference y1-y2 between the two, and the past predetermined quantity, n, is stored as the difference value from the latest value. When calculating means 32 obtains the latest difference value from counters 21B and 21C, it abandons the oldest stored value and adds the latest value to the storage. The first process is started when the power of the stage apparatus is turned on and the counters 21A to 21C are reset to zero.
[0024]
The second process is started by an external interrupt signal, counts the number of the difference y1-y2 stored after the interrupt signal, m, and stores it when m = n / 2. An average value y of the n differences y1−y2 is obtained, and the average value y is output. When n is an odd number, the fraction of m is rounded down or rounded up.
The arithmetic processing means 32 further performs a process for removing an error caused by the squareness error of the coordinate system as a third process, that is, a process for removing an error caused by an attitude error at the preset and measurement points. I do. FIG. 6 is a flowchart for explaining the procedure of the third process by the computing means 32.
[0025]
When the power of the stage apparatus is turned on and the first process is started, the calculation means 32 determines whether or not the reference position signal Ref from the linear encoder has already been detected (S11). For example, the process proceeds to a process for removing an error caused by an attitude error at a measurement point after S21. If not detected, the process proceeds to S12 and waits for detection of the reference position signal Ref (S12).
[0026]
Since the reference position signal Ref has not been detected immediately after the stage apparatus is turned on, the process proceeds to S12. If the reference position signal Ref is detected, the process proceeds to S13. In S13, the values y1ref and y2ref of the counters 21B and 21C when the reference position signal Ref is detected are stored, and the process proceeds to S14. In S14, an interrupt signal that causes the second process to calculate and output an average value is output, and the average value y output from the second process is received. Then, the process proceeds to S15.
[0027]
In S15, the values y1 and y2 of the current counter 21B and counter 21C are preset to y1-y1ref and y2-y2ref-y, respectively, and the process returns to S11. The values of the counters 21B and 21C corresponding to the detection positions RefA and RefB of the linear encoders 22 and 24 when the reference position signal Ref is generated in the state where no attitude error has occurred by the process of S15 are both 0. .
[0028]
In S11 after returning from S15, since the reference position signal Ref has been detected, the process proceeds to S21, and X is read from the counters 21A, and Y1 and Y2 are read from the counters 21B and 21C (S21). The read Y1 and Y2 are values including the preset value. Then, these values are applied to Equation 1 to calculate the Y coordinate value at the observation position (S22). Then, the two-dimensional coordinate value (X, Y) is output to, for example, a coordinate display unit or an external device (S23), and the process returns to S21.
[0029]
The average y of the differences obtained during the above process is stored, and this value can be used when the stage apparatus starts to be used next time. Since the first and second processes can be performed only when the stage apparatus is used for the first time, the startup time of the daily apparatus is shortened.
Next, the number of differences (y1−y2) used in the process of obtaining the average y of the detected value differences is examined using a t-test. An average value D of n differences (y1−y2) when n times of obtaining a difference (y1−y2) between the detected positions of the linear encoders 22 and 24 when the reference position signal Ref is detected is , Δ = 0, the formula is 2.
[0030]
D = Σ (y1-y2) k / n Equation 2
However, k is a subscript for identifying each of the n differences (y1-y2), 1 to n.
Here, when the width of the variation of n differences (y1−y2) is 2σ and 1 μm (σ is the standard deviation of the variation), and the reproducibility of the required reference position is 0.01 μm, the average value D The distribution is a t-distribution, and the standard deviation σ (D) of the variation is Equation 3.
[0031]
σ (D) = σ / √n = 0.5 / √n Equation 3
The 95% confidence value t of the t-distribution is
± t = t0.05 · σ (D) = t0.05 · σ / √n Equation 4
The value of t0.05 is a value in the t distribution table giving a 95% confidence value, and is 1.96 when n is infinite. Therefore, t0.05 = 2 and t is 0.01 μm from the reproducibility of the required reference position, and the number n of data used for averaging is obtained.
[0032]
0.01 = 2 · 0.5 / √n Equation 5
n = 10,000 Equation 6
That is, even if the difference (y1−y2) between the positions of the linear encoders 22 and 24 detected by the reference position signal Ref varies by 2σ = 1 μm due to yawing when the upper plate 10Y moves in the vicinity thereof, the present invention. A reproducibility of 0.01 μm can be obtained by averaging the values of 10,000 points in the vicinity of the detection position by the calculation unit 32. As a result, the X coordinate axis is inclined and set for the yawing, and it is possible to prevent an error due to the squareness error from occurring in the obtained two-dimensional coordinate value.
[0033]
When the linear encoders 22 and 24 both have reference position signal generating means, the above-described processing is performed by the reference position signal Ref1 detected earlier, and the reference position signal Ref2 detected later is invalidated.
The position of the reference position signal Ref described above may be replaced with the position when the power of the stage apparatus is turned on. In this case, since the position of the reference position signal cannot be reproduced, it is necessary to perform the first process and the second process of the calculation means 32 every time the stage apparatus is used.
[0034]
Note that the present invention is not limited to a length measuring device, and is a general device having a moving base for placing an object to move in two directions at right angles and detecting the amount of movement in the two directions, for example, predetermined The present invention can also be applied to a processing apparatus that performs various processing such as cutting at one place and a marking apparatus that performs marking at a predetermined place.
[0035]
【The invention's effect】
In a device that removes Abbe's primary error, the difference between the measurement values of the two position detection means is obtained at a plurality of positions near the reference position, and the measurement reference yaw error is corrected by the average value of the differences, thereby achieving high accuracy. A length measuring device is realized.
[Brief description of the drawings]
FIG. 1 is a perspective view of a measuring apparatus including a stage apparatus according to a position embodiment of the present invention.
FIG. 2 is a perspective view of a stage device of the apparatus of FIG.
FIG. 3 is a diagram showing an outline of a configuration of a processing unit.
FIG. 4 is a view for explaining the principle of calculation by calculation means of a processing unit.
FIG. 5 is a diagram for explaining how a squareness error occurs.
FIG. 6 is a flowchart of a calculation procedure performed by a calculation unit.
[Explanation of symbols]
10 ......... Stage device 10X ......... Lower plate 10Y ......... Upper plates 21A to C ... Counters 22, 24 ... Linear encoder 32 ......... Calculation means 33 ......... Store means

Claims (3)

At least it has a moving base which moves perpendicular two directions, a second direction away position of the first and second position wherein the detecting means a first direction perpendicular to the direction determining the position in the first direction of the moving base And a third position detecting means for obtaining a position of the movable table in the second direction, and a detection value of each of the first and second position detecting means and the third position detecting means. In the length measuring device having a calculating means for obtaining the position in the first direction at the observation position of the moving table based on the detected value of
The calculating means is configured to move the moving base in the vicinity of a reference position that is a reference for position detection in the first direction, and the first and second position detecting means are configured to move the moving base in the first direction of the moving base. And detecting the difference between the detection value of the first detection means and the detection value of the second detection means at each of the plurality of positions, and correcting the reference based on an average value of the differences A length measuring device characterized by In claim 1,
Reference position signal generation means for generating a reference position signal when the movable table reaches the reference position, and the first and second position detection means are reset by the reference position signal. Measuring device. In claim 1 or claim 2,
The length measuring device is characterized in that the correction is performed as the power to the device is turned on.

Images