JP4510520B2 - Shape measuring method and shape measuring apparatus - Google Patents

Description

  The present invention mainly relates to a technique for measuring an optical element and an optical element molding die with high accuracy, and in particular, measuring a shape of an object with high accuracy using a contact-type probe scanned by computer control. The present invention relates to a shape measuring method and a shape measuring apparatus for doing so.

  In recent years, with the improvement in performance of various optical devices such as imaging cameras, laser beam printers, copiers, and semiconductor exposure devices, the quality of optical elements such as lenses, mirrors, and prisms incorporated in these optical devices, especially shape accuracy, has been improved. The demand for this is becoming more sophisticated. Specifically, the shape of the optical element is complicated from a conventional flat surface, spherical surface, or axisymmetric aspheric surface to a free-form surface, and the required shape accuracy of the optical surface becomes severer as the shape becomes more complex. Yes.

  Under such circumstances, in the process of measuring the shape of an optical element, which is indispensable for manufacturing a high-precision optical element, or measuring the shape of a mold for molding an optical element, a complicated shape such as a free-form surface is used. A shape measuring apparatus using a contact type probe is widely used because it can be measured and a surface shape having a steep gradient can be measured.

  In order to achieve particularly high measurement accuracy, a shape measuring apparatus having a function of performing measurement by scanning a contact probe with respect to the shape of the surface to be measured is used. At this time, since the probe tip portion is required to realize continuous scanning scanning with respect to the shape of the surface to be measured, a probe having a spherical tip portion is generally used. Such a conventional shape measuring apparatus and a method for measuring the shape of an optical element or an optical effective surface of an optical element molding die using the apparatus are disclosed in, for example, Patent Document 1.

  FIG. 10 shows a shape measuring method according to the above prior art. First, when the surface shape of the surface to be measured 104 has no shape error with respect to the design shape defined by z = f (x, y), that is, Aim based on the design shape of the surface 104 to be measured, assuming that a surface to be measured having a surface shape that completely matches the design shape and scanning the contact-type probe 101 with respect to the surface to be measured The measurement path 105 is set. The probe 101 has a probe tip sphere 102. By calculating the coordinate points obtained by offsetting the coordinate points on the target measurement path 105 by the radius of the probe tip sphere 102 in the normal direction at each coordinate point, these are obtained. The probe scanning path 106 obtained by continuously arranging the coordinate points is set, and the center position of the probe tip sphere 102 is scanned along the probe scanning path 106 in the shape measurement step.

ここで、（２）式のＲはプローブ先端球１０２の半径を表す。

Here, R in equation (2) represents the radius of the probe tip sphere 102.

  As described above, in the shape measurement of the optical element or the optically effective surface of the optical element molding die according to the prior art, as shown in FIG. 10, the probe scanning in which the center position of the probe tip sphere 102 is derived by the above procedure. The probe tip sphere 102 is scanned along the path 106 while the probe tip sphere 102 is in contact with the surface to be measured 104, and the center coordinates of the probe tip sphere 102 are measured at a predetermined sampling interval. By repeating this operation, the optically effective surface shape of the surface to be measured 104 is measured as a surface shape constituted by a point group offset in the normal direction at each measurement point on the surface to be measured 104.

After the measurement is completed, the probe tip sphere shown in FIG. 10 is reversed from the measurement surface shape expressed as the point cloud data offset in the normal direction according to the shape analysis program installed in the analysis computer of the shape measuring apparatus. The contact point coordinates of 102 are estimated and calculated, and the measurement shape of the measured surface 104 is calculated.
JP 2003-97939 A

  When the shape measuring device and the shape measuring method according to the prior art are employed, when the shape error of the actual measured surface is large with respect to the design shape of the optically effective surface such as an optical element of a plastic mold product, the above procedure is used in advance. There is an unresolved problem that the deviation between the target contact position of the probe with respect to the surface to be measured and the actual probe contact position becomes large, and as a result, highly accurate shape measurement is difficult.

  More specifically, in the shape measurement according to the conventional technique, as described above, the normal direction at the target contact position with respect to the measurement surface of the probe tip sphere, that is, the probe tip sphere is obtained based on the design shape of the measurement surface. The probe scanning path, which is the scanning path of the center position of the probe tip sphere, is derived by continuously arranging coordinate points that are offset by the radius of the probe tip sphere in the same direction. When is large, the deviation from the target measurement path on the surface to be measured increases, and the measurement accuracy is lowered.

  For example, the surface shape of the surface 104 to be measured shown in FIG. 10 does not actually coincide completely with the design shape represented by z = f (x, y) in FIG. To do. In particular, when the object to be measured is an optical element of a plastic mold product or the like, since the shape correction of the optical element molding die is insufficient during the manufacturing process, the shape error with respect to the design shape of the surface to be measured is large. There is a case. When the shape error is large as described above, the reason why it is difficult to measure the shape with high accuracy by the conventional technique is as follows.

FIG. 12 is a diagram showing a state of contact between the measurement target surface 104 shown in FIG. 10 and the probe tip sphere 102 of the probe 101 in the XZ section of the XYZ coordinate system. Here, the actual measured surface shape 108 has a shape error as shown by a solid line with respect to the measured surface design shape 107 indicated by a broken line. As described above, the probe scanning path 106 according to the conventional technique obtains the center position L _Md of the probe tip sphere 102 based on the calculation method described above for the coordinate point P _Wd on the target measurement path 105 in FIG. . In the shape measuring apparatus according to the prior art, the center position L _Md has only to be obtained as XY coordinates defining two degrees of freedom in order to follow and control the contact type probe with respect to the surface to be measured.

Here, assuming that the change rate of the shape is negligibly small with respect to the Y-axis direction orthogonal to the illustrated cross section of the measured surface design shape 107 and the actual measured surface shape 108 shown in FIG. It is assumed that scanning is performed in a direction that approximately matches the axial direction. At this time, the probe 101 has an XY coordinate positioned at the center position L _Md of the probe tip sphere 102 calculated in advance, and as a result, the probe tip sphere 102 has a coordinate point P _{Wm with} respect to the measured surface shape 108 as shown in FIG. Will come in contact. That is, this contact point (coordinate point P _Wm ) is an actual measurement path represented as a point group of actual contact points of the probe tip sphere 102. As is apparent from FIG. 12, the target measurement path and the actual measurement path in contact with the probe tip sphere 102 are deviated by d ₁ in the XZ section shown.

  This deviation becomes particularly large when measuring a surface to be measured that is steeply inclined and has a large shape error as shown in FIG. As described above, in the measurement according to the conventional technique, the contact point of the probe tip sphere cannot be scanned along the target measurement path depending on the shape error of the object to be measured, and as a result, the measurement accuracy may deteriorate.

  Moreover, since there is always a setting error of the object to be measured with respect to the shape measuring device to be used, even if the shape of the surface to be measured is a shape that completely matches the design shape, the setting error is For this reason, it is difficult to measure the shape with high accuracy by the conventional technology. The reason will be described below.

FIG. 13 is a diagram showing a state of contact between the measured surface 104 shown in FIG. 10 and the probe tip sphere 102 of the probe 101 in the XZ cross section of the XYZ coordinate system. Here, the measured surface shown by a broken line The actual surface shape 108 to be measured is located at a position shifted by the setting error as shown by the solid line with respect to the design shape 107. In FIG. 13, the probe 101, the probe tip sphere 102, the coordinate point P _Wd on the target measurement path, the center position L _Md of the probe tip sphere 102 to be positioned, and the contact point P _Wm with respect to the actual measured surface shape This is the same as FIG.

Even when there is a setting error as shown in FIG. 13, the probe scanning path deviation d ₂ is reduced in the conventional method for the same reason as when there is a shape error with respect to the design shape described with reference to FIG. 12. The probe tip sphere 102 cannot be contacted and scanned along the target measurement path. As a result, the measurement accuracy may deteriorate.

  The present invention has been made in view of the above-mentioned unsolved problems of the prior art, and in measuring the shape of an optical element or optical element molding die that requires high-precision shape measurement accuracy, the surface to be measured To provide a shape measuring method and a shape measuring apparatus capable of scanning a probe along a target measurement path even when the shape error of the object is large or when there is a setting error of an object to be measured with respect to the shape measuring apparatus It is intended.

In order to achieve the above object, the shape measuring method of the present invention is a shape measuring method for measuring the surface shape of the surface to be measured by scanning a probe having a spherical tip while making contact with the surface to be measured, a first calculation processing step of calculating a temporary measuring probe scanning route of radius of offset of the spherical tip of the probe in a normal direction from the measurement path of the aim based on the design shape of the surface to be measured, the provisional measuring probe A first measurement step of measuring the measurement surface by scanning the probe along the scanning path to obtain temporary measurement data; and a second step of calculating a setting error and an estimated shape of the measurement surface based on the temporary measurement data an arithmetic processing step, a reference plane normal vector calculated based on the reference shape which is the estimated shape and setting error, measurement production based on the radius of the spherical tip of the reference surface normal vector and the probe A third calculation process for calculating the probe scanning path for the measurement, and a second measurement process for obtaining the actual measurement data by scanning the probe along the probe scanning path for the actual measurement to measure the surface to be measured. And measuring the surface shape of the surface to be measured based on the actual measurement data.

The shape measuring apparatus of the present invention includes a holding base for supporting an object to be measured, a probe having a spherical tip, an XY axis stage for scanning the probe with respect to the object to be measured, and the probe to the object to be measured. A Z-axis stage that contacts the surface to be measured of the object to be measured and maintains a constant contact load, detection means for detecting the three-dimensional position of the spherical tip of the probe, and a design shape or an estimated shape of the surface to be measured A first computing means for computing a probe scanning path that is offset by a radius of the spherical tip of the probe in a normal direction from the target measurement path, and based on the output of the detection means, calculates a setting error of the estimated shape, the setting error and based on the reference shape which is the estimated shape, the reference surface normal vector is calculated, the reference surface normal vector and the professional A shape measurement device comprising a second calculating means for calculating a probe scanning path for production measured based on the radius of the spherical tip of the probe, and to scan the probe along the production measurement probe scanning path The surface shape of the surface to be measured is calculated based on the actual measurement data obtained by measuring the surface to be measured.

  First, provisional measurement is performed using the probe scanning path for temporary measurement based on the design shape, and the estimated shape and setting error of the object to be measured are obtained from the measured values, and a new probe scanning path for actual measurement is set. Therefore, in the actual measurement, the probe tip sphere does not greatly deviate from the target measurement path.

  The probe position data can be obtained on the intended measurement path even when the shape error relative to the design shape of the measurement surface is large, when measuring a steep measurement surface, or when the setting error is large. Highly accurate measurement is possible.

  As shown in FIG. 1, in the shape measuring apparatus that measures the surface shape of the surface 4 to be measured by bringing the probe tip sphere 2 of the probe 1 into contact with the object 3 to be measured, the analyzing computer 10 performs the first calculation. The center position of the probe tip sphere 2 with respect to the target probe contact position on the measured surface 4 is set to the normal direction of the measured surface 4 at the same position based on the design shape data of the measured object 3 by means. A temporary measurement probe scanning path is calculated as a point group obtained as a result of repeating the calculation with respect to all target probe contact positions on the measurement target surface 4 by calculating the points offset by the radius of the sphere 2. Then, the surface to be measured 4 temporarily measured as a shape defined by a continuous function by the second calculation means for the temporary measurement data obtained by scanning the surface to be measured 4 along the probe scanning path. And a calculation process for estimating a setting error of the object 3 to be measured with respect to the shape measuring device, and a new measurement coordinate system and the estimated shape are used to calculate the actual measurement probe scanning path. The actual measurement of scanning the probe 1 in accordance with the new probe scanning path calculated as described above is performed, and the obtained measurement surface shape data (actual measurement data) is subjected to the same calculation processing as the above-described shape analysis.

  In addition, when the shape error of the measured surface with respect to the design shape is relatively small, only the setting error may be separated and the actual measurement may be performed.

  As shown in FIG. 1, a probe 1 that is a contact probe has a spherically shaped portion at the tip, and this spherical tip is called a probe tip sphere 2. The probe tip sphere 2 may have a structure in which a sphere is arranged with respect to the probe 1 by some fixing method, or may be an integral structure with the probe 1.

  The probe 1 is supported so as to be relatively movable in the X, Y, and Z axial directions while bringing the probe tip sphere 2 into contact with the object to be measured 3, and the X axis stage 11a and the Y axis stage 11b are supported. The Z-axis stage 12a moves in each axial direction. The axis stages 11a, 11b, and 12a are driven by X, Y, and Z driving motors 11c, 11d, and 12b, respectively. When shape measurement is performed on the object to be measured 3, the probe tip sphere 2 is covered. The object 3 is moved two-dimensionally in the XY directions while being in contact with the surface 4 to be measured. At this time, the movement locus of the center position of the probe tip sphere 2 is detected by a position detection sensor (not shown), and the movement locus data is transferred to the data sampling device 13 and sampled. In addition, as a position detection sensor, in order to implement | achieve a highly accurate shape measurement, a laser length measuring device, a linear scale encoder, etc. are used, for example, However, It is not limited to these position detection sensors.

  The drive motors 11c, 11d, and 12b are driven by supplying power based on drive control signals from the XY-axis control device 11 and the Z-axis control device 12 to the drive motors 11c, 11d, and 12b, respectively. Is done. The XY axis control device 11 and the Z axis control device 12 operate according to a drive control signal from the analysis computer 10. The analysis computer 10 includes a processor (not shown) mainly composed of a CPU and the like, and a hard disk drive and other storage means (not shown). The analysis computer 10 is installed with a shape measurement program having first and second calculation means for realizing the shape measurement method of the present embodiment, as will be described later, and the probe according to the shape measurement program. Various arithmetic processes for shape measurement using 1 are executed.

  Here, driving of the Z-axis stage 12a during shape measurement will be described in detail. 1 has a function of detecting the contact force of the probe 1 with respect to the object 3 to be measured, and a contact force signal corresponding to the magnitude of the detected contact force is output to the Z-axis control device 12. It is captured. The Z-axis control device 12 incorporates a processor (not shown) mainly composed of a DSP and the like and a storage means (not shown) such as a ROM, and controls so that the contact force signal is always kept constant. A control program is installed. According to this control program, the Z-axis control device 12 outputs Z-axis drive power that maintains the contact force signal at a constant value, and the output drive power is supplied to the Z-axis drive motor 12b. Further, the Z-axis stage 12a is driven by the Z-axis drive motor 12b, and the contact force of the probe 1 to the object to be measured 3 is kept constant.

  The driving of the XY axis stages 11a and 11b during shape measurement will be described in detail. The analysis computer 10 transmits a drive control signal to the XY axis control device 11 in accordance with an instruction of the shape measurement program executed at the time of shape measurement. The XY axis control device 11 supplies drive power to the XY axis drive motors 11 c and 11 d based on the drive control signal received from the analysis computer 10. Furthermore, the X-axis stage 11a and the Y-axis stage 11b are driven by the XY-axis drive motors 11c and 11d that operate according to the drive power, and the tip of the probe 1, that is, the center position of the probe tip sphere 2 is positioned on the object 3 to be measured. Positioning control is performed with respect to the surface to be measured 4 in the XY-axis direction.

  By driving the XYZ axis stages 11a, 11b, and 12a, the probe 1 scans the surface to be measured 4 of the object to be measured 3 according to the probe scanning path for temporary measurement or actual measurement set by the shape measurement program. . During probe scanning, the output signal of the position detection sensor that detects the center position of the probe tip sphere 2 is captured by the data sampling device 13 at a constant sampling interval, and the captured data is further transferred from the data sampling device 13 to the analysis computer 10. Are sequentially output at appropriate sampling intervals according to the measurement conditions set by the shape measurement program, and the measured surface shape data is acquired as X, Y, Z coordinate point group data. Finally, surface shape data is obtained by performing shape analysis processing described later after the probe scan is completed.

  FIG. 2 is a flowchart for measuring the optical surface shape of an optical element or an optical element molding die using the shape measuring apparatus shown in FIG. 1. The shape measurement is started in step S1, and in step S2, the shape measurement is started. Temporary measurement which becomes 1 measurement process is performed. In this temporary measurement, specifically, shape measurement is performed under the following measurement conditions.

  In the temporary measurement, which is the first measurement process in step S2, the shape of the surface 4 to be measured is measured using the shape measurement apparatus of FIG. 1 by a specific measurement method described later based on FIG. 4 or FIG. . After the provisional measurement in step S2, the shape analysis process is performed on the provisional measurement data in the analysis computer 10 in step S3. First, as shown in FIG. Set the coordinate system. This new coordinate system is used in the actual measurement, which is the second measurement process performed in step S7 described later.

  That is, in each of the measurement steps (steps S2 and S7) of the temporary measurement and the actual measurement, the coordinate system before setting error correction (design shape coordinate system) shown in FIG. Each system (measurement shape coordinate system) is used. Here, the coordinate system before setting error correction (design shape coordinate system) is used in the shape analysis for the temporary measurement data performed in step S3. Specifically, the center position of the probe tip sphere 2 shown in FIG. The contact point between the probe tip sphere 2 and the measured surface 4 is determined based on the radius of the probe tip sphere 2 and the design shape of the measured object 4 on the measured surface 4 with respect to the temporary measurement data acquired as the movement trajectory. This is a coordinate system used when defining the design shape z = f (x, y) in the calculation process for curve fitting the measurement shape to the design shape while performing estimation calculation. In the curve fitting calculation process, estimation calculation is performed such that the shape error of the measurement shape with respect to the design shape is minimized by, for example, the least square method. This shape analysis is performed by a shape analysis computer program which is a calculation means installed in the analysis computer 10 of FIG. 1 in step S3.

On the other hand, the coordinate system after setting error correction (measurement shape coordinate system) is a coordinate transformation matrix [T _S ] that represents the setting error of the device under test 3 calculated as a result of performing the shape analysis on the temporary measurement data. This is a coordinate system defined by the relationship shown in the figure with respect to the coordinate system before setting error correction using (setting error coordinate transformation matrix). In the curve fitting calculation process generally performed in the measurement shape analysis, a maximum of six degrees of freedom (a parallel movement component of three freedoms) is set as to how much the measurement shape is deviated from the design shape and the position (setting error). It is obtained as a coordinate transformation matrix obtained from a parameter of degree + rotational movement component 3 degrees of freedom. That is, the setting error coordinate transformation matrix [T _S ] corresponds to this. Here, the coordinates of one point on the design shape z = f (x, y) are (x, y, z), and the coordinates are transformed by the setting error coordinate transformation matrix [T _S ] estimated after the curve fitting calculation process. If the coordinates of the one point are (x _s , y _s , z _s ), the relationship between the two coordinates is expressed by the following equation in the two coordinate systems shown in FIG.

  Up to this point, in order to simplify the explanation, the design processing for the design shape z = f (x, y) of the surface to be measured that becomes the reference shape in the curve fitting calculation processing in step S3 is performed without changing the shape. We decided to carry out. However, in this curve fitting calculation process, by changing the reference shape in the estimation calculation process, a continuous function z = f (x, x, A calculation processing method for estimating a shape as a new function z = g (x, y) different from y) is generally well known. In this calculation processing method, the reference shape z = g (x, y) calculated as a result of the estimation calculation accompanied by the convergence calculation is the estimated shape of the measured surface 4.

  Also in the present embodiment, a shape analysis computer program capable of executing this measurement shape estimation calculation process is installed in the analysis computer 10, and is used in the shape analysis process in step S3, so that the setting is performed. Simultaneously with the error, the estimated shape represented by the continuous function z = g (x, y) which is the reference shape and is different from the design shape is calculated. The setting error and the estimated shape obtained in this way are used for calculation of the probe scanning path for actual measurement in the processes after step S4.

  In step S4, the reference shape represented by z = g (x, y) calculated by the shape analysis in step S3, that is, the estimated shape based on the provisional measurement data, is newly designed in the arithmetic processing in steps S5 and S6 described later. Shape. That is, in the processing of the shape measurement program installed in the analysis computer 10 of FIG. 1, the reference shape z = g (the estimated shape) is used as a new design shape used for calculating the actual measurement probe scanning path. x, y) are used.

  Step S6 is a process of setting the actual measurement probe scanning path performed in the process of step S7 according to the same procedure as in the prior art based on the unit normal vector and the probe tip sphere radius calculated in step S5. . As for the setting of the scanning path, the shape measurement program installed in the analysis computer 10 in FIG. 2 has the setting function, and the program by the first calculation means is executed in the analysis computer 10. A new probe scanning path is calculated and stored in the storage means in the analysis computer 10.

  Based on the actual measurement probe scanning path derived from the steps S1 to S6, the same measurement object 3 is removed from the shape measuring apparatus without removing the measurement object 3 from the shape measuring apparatus. While maintaining the attached state, the actual measurement, which is the second measurement process, is performed in step S7.

  In this way, the shape measurement is performed according to the probe scanning path taking into account the setting error and the shape error of the object to be measured, so that the target measurement path can be scanned with higher accuracy than the conventional technique. As a result, the shape measurement accuracy can be greatly improved.

  Finally, in step S8, the same calculation process as the shape analysis performed in step S3 is performed on the actual measurement data obtained by the actual measurement in step 7. As a result, the setting error and the shape error with respect to the shape measuring device of the object to be measured can be obtained for the actual measurement data as in the provisional measurement. After performing the measurement shape analysis process, the shape measurement is finished in step 9.

  Next, a specific method of the temporary measurement in step S2 will be described with reference to FIG. The XYZ coordinate system in the figure indicates the coordinate system before setting error correction (design shape coordinate system) described in FIG. 3, and the design shape of the surface 4 to be measured is z = f (x, y) in the same coordinate system. Is defined. Here, the actual shape of the measured surface 4 does not coincide with the design shape z = f (x, y), and is a shape including the shape error described above. Further, since there is a setting error of the object to be measured 3 with respect to the above-described shape measuring apparatus, a deviation from the design shape due to the error also occurs.

  At this time, the target measurement path 5 of the probe 1 and the probe scanning path (provisional measurement probe scanning path) 6 that is the center position of the probe tip sphere 2 corresponding to the path are calculated from the same calculation process as in the prior art. And set as measurement conditions for temporary measurement. The detailed procedure for deriving the probe scanning path is as described above, and is omitted here. In the method shown in FIG. 4, cross-sectional shape measurement is performed for each cross section in the X-axis direction of the coordinate system before setting error correction (design shape coordinate system) and the Y-axis direction orthogonal thereto.

  Such provisional measurement is performed when the shape of the surface to be measured 4 is an axisymmetric spherical shape or aspherical shape. This is because the axially symmetric optical element or the optical element molding die used as the object to be measured 3 measures the cross-sectional shape including the optical axis (axis defining the axially symmetric shape) from the characteristics of the processing method. This is because the overall shape of the measured surface 4 can be estimated. Here, only from the measurement shape of one cross section, the curve fitting calculation process for estimating the setting error with respect to the shape measuring device of the surface to be measured 4 has a total of five degrees of freedom of translation and rotation (the optical axis because of the axially symmetric shape). The rotation component with respect to cannot have a degree of freedom). On the other hand, from the measurement shape of two or more cross-sections, the estimation calculation of the five degrees of freedom for the setting error can be performed. For this reason, in consideration of shortening of the measurement tact, provisional measurement is performed on two cross-sections, which is the minimum number of cross-sections capable of estimating a setting error. Furthermore, in consideration of setting error and shape error estimation accuracy in temporary measurement, temporary measurement of two orthogonal cross-sections that enables higher accuracy estimation is performed. That is, in FIG. 4, the target measurement paths 5 in the X-axis direction and the Y-axis direction are orthogonal to each other on the optical axis where the axisymmetric design shape z = f (x, y) of the measurement surface 4 is defined.

  In step S2, provisional measurement is performed by the method shown in FIG. 4, and then the above-described provisional measurement shape analysis (step S3) and setting of the actual measurement probe scanning path are performed (steps S4 to S6). In the shape analysis on the temporary measurement data in step S3, the continuous function z = g (x, y, which is the estimated shape of the surface 4 to be measured and is different from the design shape simultaneously with the estimation of the setting error by the curve fitting calculation process described above. ) Is calculated. This reference shape z = g (x, y) is a curvature (R value) that specifically defines the shape of an axisymmetric aspherical surface when the shape of the surface 4 to be measured is an axisymmetric shape as in this embodiment. For the conic coefficient or polynomial coefficient, after the setting error correction, that is, in a coordinate system after the setting error correction shown in FIG. Using the new reference shape z = g (x, y) derived in this manner, the actual measurement probe scanning path is set in steps S4 to S6.

  After setting the actual measurement probe scanning path in steps S4 to S6, the actual measurement (step S7) and the measurement shape analysis (step S8) are sequentially performed to finish the shape measurement (step S9). As described above, in the temporary measurement, two orthogonal cross sections are measured. In the actual measurement, the number of measurement data points is increased from that in the temporary measurement, for example, by increasing the number of measurement cross sections to three or more cross sections. Details regarding the setting of the probe scanning path during the actual measurement will be described later with reference to FIGS. 6 to 9 as specific examples of the actual measurement execution method.

  As described above, by adopting the provisional measurement method of FIG. 4, it is possible to set a probe scanning path with high accuracy by a simple method for an object having an axially symmetric spherical surface or an axially symmetric aspherical shape, and efficiently. This is effective in improving the shape measurement accuracy.

  FIG. 5 shows another specific method of temporary measurement in step S2, and the XYZ coordinate system in the figure is the same as FIG. 4 before the setting error correction coordinate system (design shape coordinates) described in FIG. The design shape of the measured surface 4 is defined by z = f (x, y) in the same coordinate system. Here, the actual shape of the surface 4 to be measured does not match the design shape z = f (x, y), and includes the shape error described above. Further, since there is also a setting error of the object 3 to be measured with respect to the above-described shape measuring apparatus, a deviation from the design shape caused by the error is also the same as the specific example of FIG.

  Similarly to the specific example of the temporary measurement shown in FIG. 4, the target measurement path 5 of the probe 1 and the probe scanning path 6 which is the scanning path of the center position of the probe tip sphere 2 corresponding to the same path are also described. It is calculated from the calculation process by the same method as the conventional technique. In this specific example, as shown in FIG. 5, raster scanning is performed in which the Y-axis direction of the coordinate system before setting error correction (design shape coordinate system) is the main scanning direction of the probe 1, and the surface shape of the measured surface 4 is measured. This is different from the provisional measurement method of FIG. In FIG. 5, the probe main scanning direction in the raster scanning is the Y-axis direction, but it may be the X-axis direction. Further, raster scanning may be performed in which an arbitrary direction is not the main scanning direction but the X-axis or Y-axis direction is the main scanning direction.

  The temporary measurement according to this specific example is used when the shape of the surface to be measured 4 is mainly a non-axisymmetric shape such as a free-form surface shape. For the axially symmetric shape, it was possible to estimate the overall shape of the surface to be measured by measuring the cross-sectional shape including the optical axis as shown in FIG. 4, but the non-axisymmetric shape such as the free curved surface shape. Therefore, it is impossible to estimate the surface shape, that is, the shape error of the surface to be measured 4 and the setting error with respect to the design shape only with the representative cross-sectional shape such as the two orthogonal cross-sections. Therefore, as shown in FIG. 5, it is necessary to measure the surface shape of the measurement surface 4 by raster scanning, for example.

  In this specific example, considering measurement tact shortening, provisional measurement is performed with the number of scanning lines equal to or less than the number of raster scanning lines at the time of actual measurement, which will be described later with reference to FIG. For example, the number of low-density scanning lines is set such as one-half or one-fourth of the number of probe scanning lines set as the measurement condition in the actual measurement. Note that the method for determining the number of probe scanning lines in this temporary measurement, that is, how to determine the number of scanning lines for temporary measurement with respect to the number of scanning lines for actual measurement is not limited in any way. It may be less than the same number of lines at the time.

  In step S2, provisional measurement is performed by the probe scanning method shown in FIG. 5, and then temporary measurement shape analysis (step S3) and actual measurement probe scanning path setting (steps S4 to S6) are sequentially performed based on the flowchart of FIG. . Here, when the design shape of the surface 4 to be measured shown in FIG. 5 is a shape obtained by adding an axisymmetric aspheric shape to a free-form surface shape represented by an XY polynomial such as an XY power series polynomial, in step S3. In the curve fit calculation process for the temporary measurement data, the reference shape z = g (x, y) defined by the value obtained by fitting the curvature representing the axisymmetric aspheric shape, the conic coefficient, and the polynomial coefficient is calculated. Further, the shape (fitting error shape) obtained by subtracting the reference shape z = g (x, y) calculated here from the measurement shape is further fitted with an XY power series polynomial or the like, and the fitting shape is referred to as the reference shape z = The shape added to g (x, y) may be set again as the reference shape z = g (x, y).

  Further, when the design shape of the measured surface 4 is not a shape based on an axially symmetric aspherical shape but only a free-form surface shape represented by an XY power series polynomial, etc., Rather than calculating the fitting shape expressed by an XY power series polynomial, the curve fit calculation process estimates the setting error and simultaneously fits the polynomial coefficient of the XY power series polynomial to obtain the reference shape z = g (x, y). The method of obtaining may be used. At this time, the fitting process for the fitting error shape described above may be omitted. The analysis computer 10 shown in FIG. 1 is installed with a computer program for shape analysis, which is a calculation means that can execute any of the above functions.

  Needless to say, the polynomial used in calculating the reference shape z = g (x, y) is not limited to an XY power series polynomial but may be a polynomial expressing an axially symmetric shape such as a Zernike polynomial, for example. .

  Based on the reference shape z = g (x, y) of the surface to be measured 4 obtained according to the method described above, the actual measurement probe scanning path is set in steps S4 to S6.

  Next, a method for setting the actual measurement probe scanning path will be described in detail with reference to FIG.

The coordinate system in FIG. 6 is the coordinate system after the setting error correction (measurement shape coordinate system) shown in FIG. 3, and is the target measurement path 7 of the probe 1 and the center position of the probe tip sphere 2 corresponding to the path. When deriving a probe scanning path (production probe scanning path) 8 that is a scanning path, the reference shape calculated up to step S6 represented by z _s = g (x _s , y _s ) shown in FIG. Calculate based on The detailed calculation method here is based on the coordinate system before setting error correction (design shape coordinate system) shown in FIG. 3 using the design shape z = f (x, y) in the probe scanning path derivation of the prior art. The only difference is that the reference shape z _s = g (x _s , y _s ) is used as a new design shape on the coordinate system after the setting error correction (measurement shape coordinate system). Other calculation procedures are the same as those of the conventional technique. By adopting such a probe scanning path derivation method in actual measurement, the probe scanning path in consideration of the setting error of the object to be measured 3 with respect to the shape measuring device and the shape error of the surface to be measured 4 with respect to the design shape. Since 8 can be set, it is possible to realize highly accurate shape measurement as compared with the prior art.

  In the actual measurement (step S7) as the second measurement process, as shown in FIG. 6, the number of measurement lines larger than the number of probe scanning lines shown in FIG. 5 is set. That is, surface measurement is performed under conditions where the number of measurement points is larger than that of the temporary measurement that is the first measurement step. Thereafter, the measurement shape analysis of step S8 is performed on the measurement shape data acquired in the actual measurement, and the setting error of the surface to be measured 4 and the shape error with respect to the design shape are higher than those at the time of the shape analysis of the temporary measurement data in step S3. It can be calculated with accuracy. After the shape analysis (step S8) for the actual measurement data is finished, the shape measurement is finished (step S9).

Instead of the raster scanning shown in FIG. 6, the actual measurement may be performed by the method shown in FIGS. 7 to 9, the coordinate system in each figure represents a coordinate system after setting error correction (measurement shape coordinate system), and is represented by z _s = g (x _s , y _s ) on the measured surface 4. the aim of measurement path 17, 27, 37 is set based on the reference shape which is has a view looking down from Z _s axis plus direction.

Instead of the actual measurement by raster scanning as shown in FIG. 6, for example, a radial probe scanning as shown in FIG. 7 may be used. At this time, specifically, the probe is scanned in the order of the linear path T ₁ to the linear path T ₆ of the target measurement path 17 to measure the surface shape of the measurement target surface 4.

  Similarly, in the provisional measurement described so far, the measurement was performed on the two orthogonal cross-sectional shapes shown in FIG. 4 or the raster scan shown in FIG. 5, but in the same way as the actual measurement shown in FIG. You may employ | adopt the method of the temporary measurement which scans a probe to a surface radially. In the temporary measurement at this time, the number of lines equal to or less than the number of radial probe scanning lines shown in FIG. 7 and the surface to be measured defined on the coordinate system before setting error correction (design shape coordinate system) is used. The scanning path of the center position of the probe tip sphere, that is, the probe scanning path is set so as to scan the target measurement path set based on the design shape z = f (x, y).

  Further, both the temporary measurement and the actual measurement are performed on the concentric target measurement path 27 as shown in FIG. 8 or the spiral target measurement path 37 as shown in FIG. Also good. Further, the probe scanning path at the time of the temporary measurement and the actual measurement may be a path of a different form. For example, in the temporary measurement, the measurement may be performed with two orthogonal cross-sectional shapes as shown in FIG. 4, and in the actual measurement, the concentric measurement as shown in FIG. 8 may be performed. The combination is not limited.

  When measuring a surface to be measured that has been previously known to have a small shape error with respect to the design shape, such as an optical element molding die, a setting error in the shape analysis for the temporary measurement data in step S3 of FIG. The design shape z = f (x, y) may be used without setting the reference shape z = g (x, y) in step S4.

  As described above, when the shape error with respect to the design shape of the surface to be measured is small, it is possible to scan the target measurement path with high accuracy only by setting the actual measurement probe scanning path in consideration of the correction of the setting error. .

  It can be widely applied not only to optical elements and molds for molding them, but also to parts and stages that require highly accurate surface shapes.

It is a mimetic diagram showing the shape measuring device by one embodiment. It is a flowchart explaining the shape measuring method by the shape measuring apparatus of FIG. It is a figure explaining the relationship of two different coordinate systems used when deriving a probe scanning path | route. It is a figure explaining the specific example of the temporary measurement used as the 1st measurement process. It is a figure explaining the other specific example of temporary measurement used as the 1st measurement process. It is a figure explaining the specific example of the production measurement used as the 2nd measurement process. It is a figure explaining another specific example of the production measurement used as the 2nd measurement process. It is a figure explaining another example of the actual measurement used as the 2nd measurement process. It is a figure explaining another example of the actual measurement used as the 2nd measurement process. It is a figure explaining a prior art. It is a flowchart which shows the procedure of the shape measuring method by a prior art. It is a figure explaining the condition where a measurement path | route will shift | deviate by the shape error of the to-be-measured surface in the shape measuring method in a prior art. It is a figure explaining the condition where a measurement path | route will shift | deviate by the setting error of a to-be-measured object in the shape measuring method in a prior art.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Probe 2 Probe tip sphere 3 Object to be measured 4 Surface to be measured 5, 7, 17, 27, 37 Target measurement path 6, 8 Probe scanning path 10 Computer for analysis 11 XY axis controller 11a X axis stage 11b Y axis stage 12 Z-axis control device 12a Z-axis stage 13 Data sampling device

Claims (4)

A probe having a spherical tip is scanned while contacting the surface to be measured, and is a shape measuring method for measuring the surface shape of the surface to be measured,
A first calculation processing step of calculating a provisional measurement probe scanning path that is offset from the target measurement path in the normal direction by the radius of the spherical tip of the probe based on the design shape of the surface to be measured;
A first measurement step of scanning the probe along the temporary measurement probe scanning path to measure the surface to be measured to obtain temporary measurement data;
A second calculation processing step for calculating a setting error and an estimated shape of the surface to be measured based on the temporary measurement data;
The reference plane normal vector calculated based on the reference shape is a setting error and the estimated shape, first calculates the probe scanning path for production measured based on the radius of the spherical tip of the reference surface normal vector and the probe 3 arithmetic processing steps;
A second measurement step of scanning the probe along the production measurement probe scanning path to measure the surface to be measured and obtaining production measurement data, and
A shape measuring method comprising calculating a surface shape of a surface to be measured based on actual measurement data.   2. The shape measuring method according to claim 1, wherein the actual measurement probe scanning path has a higher density than the temporary measurement probe scanning path.   3. The shape measuring method according to claim 1, wherein at least one provisional measurement probe scanning path is provided in each of two orthogonal axes. A holding base for supporting the object to be measured, a probe having a spherical tip, an XY axis stage for scanning the probe with respect to the object to be measured, and the probe contacting the surface to be measured of the object to be measured A Z-axis stage that keeps the contact load constant, a detecting means for detecting the three-dimensional position of the spherical tip of the probe, and a method for determining a target measurement path using a design shape or an estimated shape of the surface to be measured. First calculation means for calculating a probe scanning path offset in a linear direction by the radius of the spherical tip of the probe, and setting error of the measurement surface and the estimated shape are calculated based on the output of the detection means and, wherein on the basis of the setting error and the reference shape said is estimated shape, the reference surface normal vector is calculated, the radius of the spherical tip of the reference surface normal vector and the probe A shape measuring apparatus having, a second calculating means for calculating a probe scanning path for production measured Zui,
A shape measuring device characterized in that the surface shape of the surface to be measured is calculated based on the actual measurement data obtained by measuring the surface to be measured by scanning the probe along the probe scanning path for actual measurement. .

Images