Chromatic Range Sensor System

Abstract

To provide a chromatic range sensor (CRS) system that realizes more accurate thickness measurement.SOLUTION: A chromatic range sensor (CRS) system is provided so as to determine a workpiece thickness and includes an optical pen, an illumination source, a wavelength detector and a signal processor. The optical pen indicates axial chromatic dispersion, the illumination source is configured to generate multi-wavelength light and the wavelength detector includes a plurality of pixels arranged along a measurement axis. In measurement operation, the optical pen inputs a spectral profile from the illumination source and outputs corresponding radiation to first and second workpiece surfaces of a workpiece (e.g., translucent material) and outputs reflected radiation to the wavelength detector which provides output spectral profile data. The signal processor processes the output spectral profile data to determine the workpiece thickness. In various implementations, the processing to determine the thickness may not rely on determining a distance to the workpiece and / or may utilize transform processing, etc.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to precision measurement instruments, and more particularly to chromatic range sensors and similar optical length measuring instruments and their applications. [Background technology]

[0002] Chromatic confocal technology is widely used in the field of optical distance sensors (e.g., height sensors, distance sensors, etc.). As described in U.S. Patent No. 7,876,456 (hereinafter referred to as the '456 patent), an optical element with axial chromatic aberration (also referred to herein as axial or longitudinal chromatic dispersion) is used to focus a broadband light source and vary the axial focal length with wavelength. That is, only one wavelength of the broadband light source is precisely focused on the surface, and the surface height or distance relative to the focusing element determines which wavelength is best focused. Light reflected from the surface is refocused by a small detection aperture, such as a pinhole or the end of an optical fiber. Of the light that reflects off the surface and returns through the optical system to the input / output fiber, only the wavelength well-focused at the surface is well-focused at the aperture. All other wavelengths are imperfectly focused at the aperture and therefore couple little power into the fiber. Therefore, the signal level of the light returning through the fiber will be greatest at a wavelength that corresponds to the surface height or distance to the surface. A spectroscopic detector measures the signal level at each wavelength to determine the surface height.

[0003] Several manufacturers have proposed practical, compact chromatic range sensor (CRS) systems suitable for industrial use that operate as described above, referred to as chromatic point sensors (CPS) or chromatic line sensors. The compact, chromatic-dispersive optical assemblies used in such systems are referred to as "optical pens" or "pens." The optical pen is connected to the chromatic range sensor's electronics by an optical fiber. The electronics includes a light source that transmits light down the fiber and out of the optical pen. The electronics also includes a spectrometer that detects and analyzes the returning light. The returning light is received by the spectrometer's detector array to form a wavelength-dispersive intensity profile. The pixel data corresponding to the wavelength-dispersive intensity profile is analyzed to determine a "dominant wavelength position coordinate" indicated by the peak or centroid of the intensity profile. The pixel coordinates of the peak and / or centroid, along with a lookup table, are then used to determine the distance to the surface. These pixel coordinates are determined with subpixel resolution and are referred to herein as "distance-indicating coordinates" or "distance-indicating pixel coordinates."

[0004] Also, as described in U.S. Pat. No. 8,773,757, prior art "line sensor" CRSs are known that use a slit aperture to focus light along a line rather than a single point, allowing measurement of the distance to a surface at multiple points along the line. [Prior art documents] [Patent documents]

[0005] [Patent Document 1] U.S. Patent No. 7,876,456 [Patent Document 2] U.S. Patent No. 8,773,757 Summary of the Invention [Problem to be solved by the invention]

[0006] Conventional CRS systems have presented challenges with thickness measurement, such as measurement errors. Thickness measurement typically involves determining measurement distances to two opposing surfaces of a workpiece and determining the difference between the two measurement distances. For example, when measuring the thickness of a transparent workpiece such as glass (e.g., fused silica), CRS illumination reflects off both the top and bottom surfaces of the glass, resulting in two corresponding wavelength peaks. As the workpiece thickness decreases, the two peaks tend to merge with each other, making it difficult to separate the two peaks in the intensity profile representing each measurement distance, which can result in measurement errors. For example, the widths of the two peaks may be such that the two peaks at least partially overlap or merge with each other due to insufficient separation (e.g., the measurement profile signal 510a shown by the dotted line in FIG. 5(A) and FIGS. 10(A), 10(B), 10(D), 10(E), and 10G, as described in detail below).

[0007] In particular, for the seven intensity profiles shown in Figures 10(A) through 10(G), each has two measurement peaks (e.g., a single merged peak in Figure 10(A)), which indicate the measurement distance from the top and bottom surfaces of workpieces of various thicknesses. For example, Figures 10(A) and 10(D) correspond to a 25 μm-thick workpiece, Figures 10(B) and 10(E) correspond to a 50 μm-thick workpiece, and Figures 10(C) and 10(F) correspond to a 100 μm-thick workpiece. The intensity profiles in Figures 10(A) through 10(G) have pixel coordinates or "pixel positions" of the CRS wavelength detector on the horizontal axis and intensity or "signal level" expressed as a function of pixel coordinates on the vertical axis. The pixel positions correspond to wavelengths measured by the CRS wavelength detector, and the CRS system is adjusted to convert these wavelength peaks into CRS measurement distances. Similarly, the thickness of a workpiece can be determined based on the difference in measurement distance from the two workpiece surfaces using conventional methods.

[0008] As shown in some examples in FIGS. 10A-10C, the merging of two peaks may be more severe in the blue portion of the Z range (i.e., the blue wavelength portion) of the CRS system than in the red portion of the Z range (i.e., the red wavelength portion) of the CRS system, as shown in FIGS. 10D-10F. This is because the blue portion is closer to the optical pen optics of the CRS system along the optical axis (Z axis) of distance measurement. Therefore, one technique that can provide some improvement in the merging issue is to position the workpiece to be measured in the red portion of the Z range of the CRS system (i.e., at a sufficient distance from the optical pen). Also, as clearly shown in FIG. 10G, another technique that can provide some improvement in the merging of wavelength peaks is to set a higher signal level threshold (e.g., threshold B) to separate and detect the two peaks. In contrast, a lower signal level threshold (e.g., threshold A) will not separate the two peaks and will only detect a single (merged) peak. This related concept will be explained in more detail below with reference to FIG. 5A. These methods have certain limitations (e.g., it is not always easy to place a particular part in the red part of the Z-range of the CRS system, and many signal levels may go undetected if the detection threshold is set too high).

[0009] Various aspects of the present invention aim to ameliorate these problems, in one aspect by providing more accurate thickness measurements (e.g., for thinner workpiece thicknesses where wavelength peaks merge, and for measurements over an expanded range of possible optical pen-to-workpiece distances). [Means for solving the problem]

[0010] According to an embodiment of the present invention, a chromatic range sensor (CRS) system for determining workpiece thickness is provided, the CRS system comprising: an optical pen having a confocal optical path including an optical system that exhibits axial chromatic dispersion and is configured to focus different wavelengths of light at different distances near the workpiece; an illumination source that generates multi-wavelength incident light having an input spectral profile that is input to the optical pen; and a CRS wavelength detector providing output profile data, the CRS wavelength detector having a plurality of pixels located at positions along a measurement axis of the CRS wavelength detector, wherein the CRS system is configured such that, when the optical pen is positioned to perform a measurement operation on the workpiece, the optical pen receives the input spectral profile, outputs illumination corresponding to the input spectral profile onto a first work surface and a second work surface of the workpiece, receives reflected light from the first work surface and the second work surface, and outputs the reflected light to the CRS wavelength detector. The CRS system further includes a signal processing unit that processes the output spectrum profile data to determine a work thickness according to the distance between the first work surface and the second work surface, and is characterized in that the work thickness determination process utilizes a conversion process.

[0011] According to one aspect, the transform process preferably includes at least one of a Fourier transform process and a Fourier transform-related process, and utilizing the transform process preferably includes determining a transform of the output spectral profile data and determining the workpiece thickness based at least in part on a first characteristic included in the transform of the output spectral profile data. According to another aspect, the first characteristic preferably corresponds to a first dip in the transform of the output spectral profile data.

[0012] According to another aspect, the output spectral profile data comprises distance-dependent profile components having first and second wavelength peaks corresponding to the first and second workpiece surfaces, respectively, and the workpiece thickness is determined based on a first measured distance to the first workpiece surface and a second measured distance to the second workpiece surface. The decision value of Preferably not to depend on it. According to a further aspect, the first wavelength peak and the second wavelength peak preferably appear visually as a single peak in an image of the output spectral profile data provided by the CRS wavelength detector.

[0013] According to another aspect, for a workpiece in which the distance between the first workpiece surface and the second workpiece surface is 5 μm, the use of the conversion process enables the signal processing unit to determine the workpiece thickness with an accuracy of less than 10% error.

[0014] According to another aspect, the signal processing unit is preferably further configured to process the output spectrum profile data to determine a measured distance to the workpiece. According to a further aspect, determining the measured distance to the workpiece includes processing the power spectrum profile data to determine a center of gravity indicating the measured distance to the workpiece, and determining the workpiece thickness includes processing the power spectrum profile data to determine the center of gravity indicating the measured distance to the workpiece. To the value Preferably not to depend on it. According to a further aspect, the determined measured distance is preferably a distance to a point inside the workpiece between the first workpiece surface and the second workpiece surface. According to a still further aspect, the measurement distance is preferably determined according to a distance indicating coordinate that is an average of a first wavelength peak and a second wavelength peak corresponding to the first work surface and the second work surface, respectively. According to a further aspect, the signal processing unit uses the determined measurement distance and the determined workpiece thickness to calculate the distance to the first workpiece surface. No. 1. Measurement distance and the second work surface No. Preferably, provision is made to determine at least one of the two measured distances.

[0015] According to another aspect, the conversion process is performed using: determining a Fourier transform of the output spectral profile data; dividing the Fourier transform of the output spectral profile data by a Fourier transform of a reference to obtain a Fourier transform of target spectral profile data; calculating an inverse Fourier transform of the Fourier transform of the target spectral profile data to determine first and second wavelength peaks corresponding to a first measured distance to the first workpiece surface and a second measured distance to the second workpiece surface, respectively; and Preferably, the method includes determining the workpiece thickness based at least in part on a difference between the measured distances to the first workpiece surface and the second workpiece surface.

[0016] According to yet another aspect, the CRS system is operable in multiple modes, including a thickness measurement mode, and for this purpose, the signal processor preferably performs processing to determine the thickness as part of the thickness measurement mode. According to a further aspect, the multiple modes further include a distance measurement mode for measuring the distance to the workpiece, and the CRS system is preferably configured such that, during the distance measurement mode, when the optical pen is in a position where it can perform a measurement operation on the workpiece, the optical pen inputs the input spectral profile, outputs illumination to the workpiece in accordance with the input spectral profile, receives reflected light from the workpiece, and outputs the reflected light to the CRS wavelength detector, which is capable of providing output spectral profile data, and the output spectral profile comprises distance-dependent profile components having wavelengths of light that indicate the measured distance to the workpiece. According to a further aspect, the thickness measurement mode can only operate to determine the thickness depending on the distance between the first work surface and the second work surface, and cannot operate to determine multiple distances between three or more work surfaces.

[0017] According to another aspect, the CRS system is preferably at least one of a chromatic point sensor system and a chromatic line sensor system.

[0018] According to another aspect, a method of operating a chromatic range sensor (CRS) system for determining workpiece thickness is provided. The CRS system comprises: an optical pen having a confocal optical path including an optical system that exhibits axial chromatic dispersion and is configured to focus different wavelengths of light at different distances near the workpiece; an illumination source that generates multi-wavelength incident light having an input spectral profile that is input to the optical pen; and a CRS wavelength detector having a plurality of pixels positioned at respective positions along a measurement axis of the CRS wavelength detector, the plurality of pixels capable of receiving light of each wavelength and providing output spectral profile data. The method of operation comprises: operating the CRS system with the optical pen operatively positioned relative to the workpiece to perform a measurement operation, the measurement operation including the optical pen receiving the input spectral profile, outputting illumination responsive thereto onto a first work surface and a second work surface of the workpiece, receiving reflected light from the first work surface and the second work surface, and outputting the reflected light to the CRS wavelength detector capable of providing the output spectral profile data; and The method comprises processing the output spectrum profile data to determine a work thickness according to the distance between the first work surface and the second work surface, and is characterized in that the work thickness determination process utilizes a conversion process.

[0019] According to another aspect, a method of operating a chromatic range sensor (CRS) system is provided. The CRS system comprises: an optical pen having a confocal optical path including an optical system that exhibits axial chromatic dispersion and is configured to focus different wavelengths of light at different distances near the workpiece; an illumination source that generates multi-wavelength incident light having an input spectral profile that is input to the optical pen; and a CRS wavelength detector having a plurality of pixels positioned at respective positions along a measurement axis of the CRS wavelength detector, the plurality of pixels capable of receiving light of each wavelength and providing output spectral profile data. The method of operation comprises: operating the CRS system with the optical pen operatively positioned relative to the workpiece to perform a measurement operation, wherein the measurement operation includes the optical pen receiving the input spectral profile, outputting illumination responsive to the input spectral profile onto a first work surface and a second work surface of the workpiece, receiving reflected light from the first work surface and the second work surface, and outputting the reflected light to the CRS wavelength detector capable of providing the output spectral profile data; processing the output spectral profile data to determine a measured distance to the workpiece, wherein determining the measured distance includes determining distance indicating coordinates corresponding to pixel locations of the CRS wavelength detector that represent the measured distance; and processing the power spectrum profile data to determine a workpiece thickness as a function of a distance between the first workpiece surface and the second workpiece surface; The process of determining the workpiece thickness is characterized in that it does not depend on the determined value of the measured distance to the workpiece. [Brief explanation of the drawings]

[0020] [Figure 1] FIG. 1 is a block diagram illustrating one embodiment of a representative CRS system. [Figure 2] FIG. 2 is a diagram illustrating the system noise (bias) profile of a CRS system, showing the wavelength-dependent voltage offset signal level for the pixels of the detector array when no measurement surface is present. [Figure 3] Figure 3 is a diagram of an intensity profile from a CRS system showing effective wavelength peaks caused by reflected wavelength light from the top surface of a workpiece (e.g., an opaque workpiece), with the pixel location of the peaks coinciding with the measured distance to the top surface. [Figure 4] Figure 4(A) is a diagram of CRS distance calibration data, showing the correlation between distance-indicating pixel coordinates and known measurement distances to the workpiece surface being measured. Figure 4(B) is an example of a CRS distance calibration lookup table, referencing the distance-indicating coordinates (DICs) of the CRS system to the corresponding measurement distances of the CRS system. [Figure 5] Figure 5(A) is a diagram of the intensity profile from the CRS system, showing two effective wavelength peaks caused by reflected wavelength light from the top and bottom surfaces of a transparent workpiece, with the pixel locations of the peaks corresponding to the measured distances to those surfaces. Figure 5(B) is a diagram showing the measured distances to the top and bottom surfaces of a transparent workpiece, including the actual and apparent measured distances to the bottom surface. [Figure 6] FIG. 6 is a flow diagram illustrating the method of operation of the CRS system for determining workpiece thickness. [Figure 7] 7(A) to 7(D) are diagrams showing the process of measuring the thickness of a workpiece using the conversion process. [Figure 8] FIG. 8 is a diagram illustrating a process for measuring workpiece thickness using fast Fourier transform (FFT) processing as part of a deconvolution method. [Figure 9] Figures 9(A) to 9(H) show examples in which the FFT processing method of Figure 8 can determine the location of the first dip in the Fourier transform of a signal that can be used to determine workpiece thickness, regardless of the workpiece position within the measurement range that can be operated on the optical pen. [Figure 10] Figures 10(A)-10G are graphs showing intensity profiles of workpieces of various thicknesses (e.g., 25 μm, 50 μm, 100 μm) measured by the CRS system, showing the varying degree of merging of the two wavelength peaks depending on the thickness and distance from the optical pen. [Figure 10G]Figures 10(A)-10G are graphs showing intensity profiles of workpieces of various thicknesses (e.g., 25 μm, 50 μm, 100 μm) measured by the CRS system, showing the varying degree of merging of the two wavelength peaks depending on the thickness and distance from the optical pen. DETAILED DESCRIPTION OF THE INVENTION

[0021] FIG. 1 is a block diagram illustrating one exemplary embodiment of a chromatic range sensor (CRS) system 100, which includes an optical element 120 (e.g., an optical pen), an electronics unit 160, and a user interface 171. In this embodiment, the electronics unit 160 includes a signal processor 166, a memory unit 168, and a light source / detector subsystem 161, which includes a wavelength detector 162 and a broadband light source 164. The CRS system 100 shown in FIG. 1 is a chromatic point sensor (CPS) system that measures a single measurement point at a time; however, it will be understood that other CRS systems (e.g., chromatic line sensors) can be used in other embodiments. In various embodiments, the wavelength detector 162 includes a spectrometer detector array 163. The detector array 163 can be configured with multiple pixels arranged on the measurement axis of the wavelength detector 162. Each pixel receives a respective wavelength signal and provides output spectral profile data.

[0022] The electrical component 160 is connected to the optical element 120 through an optical path including a fiber optic cable 112. The optical path may be any of several alternative embodiments. For example, the optical path may include a fiber optic cable 112 having first and second optical fiber segments 112A and 112B connected by a connector (CONNECT-D) on the second optical fiber segment 112B, and a coupler (COUPLER-O) connecting the second optical fiber segment 112B to a third optical fiber segment 112C connected to the electrical component 160. A light source 164 under the control of a signal processor 166 is connected to the optical element 120 via a path including an illumination fiber segment (165I), a 2×1 coupler (COUPLER-E, CONNECT-E), and the fiber optic cable 112 to input multi-wavelength light that constitutes an input spectral profile. The optical element 120 includes an input / output optical fiber subassembly 105, a housing 130, and an optical system 150. The input / output optical fiber subassembly 105 comprises an input / output optical fiber 113 extending from the optical fiber cable 112 and an optical fiber connector 108. The input / output optical fiber 113 outputs an output beam through an aperture 195 and also receives the measurement signal light reflected through the aperture 195.

[0023] In operation, light exiting the fiber end through aperture 195 is focused by optical system 150. Optical system 150 includes an axially chromatically dispersive lens such that the focal point on optical axis OA is at a different distance depending on the wavelength of the light, as is known for CRS systems. During measurement operations, the light is focused near workpiece 170, which has first and second workpiece surfaces 190-1 and 190-2 (i.e., the top and bottom surfaces of the workpiece). In various embodiments, workpiece 170 may be transparent or may include a material that allows at least a partial view of bottom surface 190-2 through top surface 190-1 (i.e., some light from optical element 120 can pass through top surface 190-1 to reach bottom surface 190-2). For simplicity, the following discussion will begin with a case in which workpiece 170 is opaque or has properties such that only top surface 190-1 reflects light to CRS system 100.

[0024] In this embodiment, only the top surface 190-1 reflects light, and during measurement, the light is focused by the optical system 150 onto the top surface 190-1 (as opposed to the case where light is also focused on the bottom surface 190-2, as shown in another example below). The light reflected from the top surface 190-1 is refocused onto the aperture 195 by the optical system 150. Due to this on-axis chromatic dispersion, only one wavelength of light has a focal length that matches the measurement distance "Z1." The measurement distance "Z1" is the distance from a reference position RP fixed relative to the optical element 120 to the position of the workpiece surface 190-1. The wavelength of light that is best focused on the workpiece surface 190-1 is also the wavelength of light that is best focused at the aperture 195. Therefore, primarily the best-focused wavelength of light passes through the aperture 195 and travels down the core of the optical fiber 113 of the fiber optic cable 112. Fiber optic cable 112 transmits the signal light to wavelength detector 162 for use in determining the wavelength exhibiting a predominant intensity corresponding to the measured distance Z1 to the location of workpiece surface 190-1.

[0025] In this embodiment, the reflected wavelength-dependent light intensity returns to the electronics unit 160 through a fiber optical path including a coupler (COUPLER-E), and a certain percentage (e.g., approximately 50%) of the light is directed through a signal fiber segment 165S to the wavelength detector 162. The wavelength detector 162 receives the wavelength-dependent light intensity as an output spectral intensity profile (also simply referred to as an output spectral profile) distributed across the pixel array on the measurement axis of the detector array 163, and operates to provide corresponding output spectral profile data based on the pixel data output from the detector array 163.

[0026] A sub-pixel resolution distance indication coordinate (DIC) for the profile data (see, for example, FIG. 3) is calculated by the signal processor 166, and the (sub-pixel resolution) DIC indicates the measurement distance Z1 (in microns) to the position of the workpiece surface 190-1 using a distance calibration lookup table or the like stored in the calibration unit 173 of the memory unit 168, as shown in FIGS. 4A and 4B (described below). According to conventional techniques, the DIC can be determined based on the center of gravity of the intensity profile data contained in the peak region. In various embodiments, the profile data can also be used to determine the sub-pixel resolution DIC, as described below.

[0027] It will be appreciated that, in one embodiment, if workpiece 170 is transparent, the same operational principles described above can be applied to determining the measurement distances to workpiece surfaces 190-1 and 190-2. More specifically, during a measurement operation, light is focused by optical system 150 onto workpiece surfaces 190-1 and 190-2 (i.e., some of the light can pass through workpiece surface 190-1 and reach workpiece surface 190-2). Light reflected from workpiece surfaces 190-1 and 190-2 is refocused by optical system 150 at aperture 195. Due to this on-axis chromatic dispersion, only one wavelength of light has a focal length that matches the measurement distance "Z1." Measurement distance "Z1" is the distance from a reference position RP, which is fixed relative to optical element 120, to the location of workpiece surface 190-1. Similarly, only one wavelength of light has a focal length that matches the measurement distance "Z2" from reference position RP to the location of workpiece surface 190-1. The wavelengths of light that are best focused at each location on workpiece surfaces 190-1 and 190-2 are best focused at aperture 195. Thus, these best focused dominant wavelengths pass through aperture 195 and travel down the core of optical fiber 113 of fiber optic cable 112. Fiber optic cable 112 sends signal light to wavelength detector 162 for use in determining the wavelength exhibiting the dominant intensity corresponding to the measured distance to each location on workpiece surfaces 190-1 and 190-2. As will be described in detail with reference to FIG. 5B, in various embodiments, the indicated measured distance to workpiece surface 190-2 can be an apparent measured distance Z2' related to measured distance Z2, depending on the refractive index of the workpiece material (e.g., as shown in equations (3) to (5)).

[0028] In this embodiment, the reflected wavelength-dependent light intensity returns to the electronics unit 160 through a fiber optical path including a coupler (COUPLER-E), and a certain percentage (e.g., approximately 50%) of the light is directed through a signal fiber segment 165S to the wavelength detector 162. The wavelength detector 162 receives the wavelength-dependent light intensity as an output spectral intensity profile (also simply referred to as an output spectral profile) distributed across the pixel array on the measurement axis of the detector array 163, and operates to provide corresponding output spectral profile data based on the pixel data output from the detector array 163.

[0029] Two corresponding sub-pixel resolution distance-indicating coordinates (DICs) for the profile data (see, for example, FIG. 5A) are calculated by the signal processor 166. The two (sub-pixel resolution) DICs indicate the respective measurement distances (in microns) using a distance calibration lookup table stored in the calibration unit 173 of the memory unit 168, as shown in FIGS. 4A and 4B (described below). According to a conventional technique, the two DICs can be determined based on the centroids of the intensity profile data contained in each of the two corresponding peak regions. In various embodiments, the profile data can be used to determine each of the two DICs with sub-pixel resolution, as described in detail below. In various embodiments, a DIC corresponding to the apparent measurement distance Z3′ relative to the actual measurement distance Z3 can also be determined (e.g., by calculating the centroid of the two merged wavelength peaks), or alternatively, can be determined (e.g., as shown in Equations (6) to (10)). As explained in more detail below, distance Z3 may be between distances Z1 and Z2 and may correspond to a distance to an interior location of workpiece 170, such as the midpoint of workpiece surfaces 190-1 and 190-2.

[0030] As further shown in FIG. 1 , the user interface 171 is connected to the electronics unit 160 and configured to receive user input for operation of the CRS system 100, such as user commands for selecting various operational parameters, via any suitable means, such as a keyboard, touch sensor, mouse, etc. In an exemplary embodiment, the user interface 171 may include one or more operational mode selection elements (e.g., user-selectable buttons) operable by a user to select one of a plurality of operational modes of the CRS system 100 (e.g., thickness measurement mode and / or distance measurement mode, etc.). The user interface 171 is further configured to display information, such as distance and / or thickness, successfully determined / measured by the CRS system 100 on a screen. As described in more detail below, in various embodiments, the memory unit 168 includes a thickness measurement unit 169 and a distance measurement unit 172, which can implement corresponding modes and / or provide other corresponding operations. Such corresponding modes and / or corresponding operations may be provided for workpiece thickness measurement and / or distance measurement (e.g., depending on the distance between the first and second workpiece surfaces). In various embodiments, thickness measurement portion 169 and distance measurement portion 172, and / or their associated operations, may be merged and / or may be indistinguishable.

[0031] 1 shows XYZ Cartesian coordinate axes as a basic frame. The Z direction is parallel to the optical axis (OA) of the optical element 120, which is the distance measurement axis. As shown in FIG. 1, in operation, a workpiece 170 is positioned along the optical axis OA. The workpiece 170 may be mounted on a translation stage 175 advantageously arranged to move along the Z axis direction constrained by guide bearings 175A.

[0032] FIG. 2 provides an overview of known background signal processing and / or calibration operations. FIG. 2 is a diagram 200 of a system noise (bias) profile for a CRS system, showing the voltage offset signal level Voffset(p) for pixels of the detector array 163 when no measurement surface is present across the nominal full measurement range of the CRS system. In this case, there is no intentionally reflected light, and no significant or dominant wavelength peaks in the resulting intensity profile. The voltage offset signal Voffset(p) is plotted as a normalized voltage value for each of the 1,024 pixels along the "wavelength" measurement axis. For "normalized voltage value," a value of 1.0 is assigned to the saturated voltage value of the detector array 163. The voltage offset signal Voffset(p) includes a bias signal level Vbias, which is relatively constant across the detector array, and a background signal component Vback(p), which exhibits variation across the detector array.

[0033] 3, 4(A), 4(B), 5(A), and 5(B) outline a specific signal processing operation for determining a distance indicating coordinate (DIC) with sub-pixel resolution based on effective wavelength peaks occurring in a wavelength dispersion intensity profile from a CRS system, and for determining a measured distance (e.g., in microns) to a workpiece surface based on the determined DIC. An example operation described herein is described in detail in the '456 patent. The purpose of this description is to provide additional technical background information that is helpful in providing a comprehensive understanding of the CRS measurement operation described herein.

[0034] FIG. 3 is a diagram 300 of a wavelength-dispersed intensity profile from a CRS, showing effective wavelength peaks 302 resulting from a subset (wavelength band) of measurement profile signals MS(p) representing wavelengths of light focused on and reflected from one measurement surface of a (e.g., non-transparent) workpiece. In this example, diagram 300 includes a single wavelength peak 302 corresponding to a single reflective surface (e.g., the top surface) of an opaque workpiece. Each measurement profile signal MS(p) has a signal level (in normalized voltage) associated with each pixel p of a detector array (e.g., detector array 163). The wavelength peaks 302 have sufficient height (good signal-to-noise ratio) and are relatively symmetrical, allowing for a good estimation of the peak location or measurement of a distance-indicating coordinate (DIC) 304 along the measurement axis of the detector array. FIG. 3 also shows the bias signal level (MVbias) (in normalized voltage), the peak pixel coordinate (ppc), and a data threshold (MVthreshold) that defines the lower limit of the distance-indicating subset of measurement profile signals MS(p) that form wavelength peaks 302. All values ​​(including, for example, "MV" values) are normalized voltages.

[0035] Briefly, in one embodiment, the operations of determining a distance indicating coordinate DIC (in pixels) and determining a corresponding measurement distance (in microns) based on the determined DIC may include the following steps: The target surface is positioned on the optical axis OA and the resulting wavelength dispersion intensity profile is captured as shown in diagram 300. Determine the peak pixel coordinate (ppc), which is the pixel with the highest level signal. Determine the measured bias signal level (MVbias) at a given sampling rate. Determine the data threshold (MVthreshold) (e.g., as a percentage of the peak height). ·Determine a distance-indicating coordinate (DIC) with sub-pixel resolution based on a distance-indicating subset of the measurement profile signal MS(p) that forms wavelength peaks having values ​​greater than MVthreshold. The measured distance is determined using the correlation between the DIC and the corresponding distance in the stored distance calibration data (e.g., a distance calibration curve such as in FIG. 4(A) or a look-up table such as in FIG. 4(B)).

[0036] In the above operation, DIC can be determined with sub-pixel resolution based on the distance-indicating subset of measurement profile signals MS(p) that exceed a data threshold (MVthreshold). According to conventional techniques, DIC can be determined as the sub-pixel resolution coordinate of the centroid Xc of the distance-indicating subset of signals MS(p). For example, for a detector having 1024 pixels (i.e., each pixel has a corresponding pixel number (p) ranging from 1 to 1024), the centroid Xc can be determined using Equation (1):

[0037]

number

[0038]

number

[0039] As one example, let n = 2 in equation (1). It will be appreciated that equation (2) limits the measured profile signals MS(p) used in the centroid calculation to a range-indicating subset.

[0040] FIG. 4A is a diagram 400A of CRS measurement distance calibration data 410A, showing a correlation between a distance-indicating coordinate DIC with sub-pixel resolution and a known measurement distance ZOUT in microns along the CRS optical axis OA (e.g., stored in calibration unit 173 of FIG. 3). It will be understood that the specific values ​​in FIG. 4A are for illustrative purposes only and may not correspond to the specific values ​​shown in other examples (e.g., to the specific values ​​described with respect to FIGS. 1-3 and / or to the specific table values ​​in FIG. 4B, described in detail below, although the concepts are similar). The example shown in FIG. 4A is data for an optical element (e.g., an optical pen) with a nominal total measurement range MR of 300 μm, corresponding to DICs in a range of approximately 150 to 490 pixels. However, if desired, the CRS system can be calibrated over a wider pixel range in the detector array 163. One typical laboratory calibration method for determining CRS measurement distance calibration data 410A is to use a mirror that moves along optical axis OA (e.g., replacing surface 190-1 in FIG. 1 and moving it in steps of, e.g., approximately 0.1 or 0.2 microns). At each actual mirror position, a corresponding calibration DIC for the CRS system is determined based on the corresponding intensity profile data, as described above with respect to FIG. 3. The calibration DIC and its corresponding actual position (in microns along optical axis OA) are then recorded to provide calibration data 410A. During a workpiece measurement operation (e.g., during an operation to determine a measurement distance to a workpiece surface such as workpiece surface 190-1), the stored calibration data 410A is referenced to determine the measurement distance ZOUT corresponding to the measurement DIC obtained by the CRS system. While the distance calibration data 410A appears to form a smooth curve, it will be appreciated that in some cases the distance calibration data and / or power spectral profile data for CRS systems in general, and less expensive CRS systems in particular, may exhibit variability / irregularities over some short ranges (e.g., as partially described in the '456 patent).

[0041] FIG. 4(B) is a diagram 400B of a CRS distance calibration lookup table 410B for referencing measured distances of distance-indicating coordinates for a chromatic point sensor (e.g., stored in the calibration unit 173 of FIG. 3). As noted above, FIG. 4(B) is merely an example diagram of a lookup table, and any discrepancies between the table values ​​identified in FIG. 4(B) and the displayed values ​​of the calibration data 410A identified in FIG. 4(A) are not significant or relevant for purposes herein. It will be understood that the table values ​​in FIG. 4(B) are for illustrative purposes only and may not correspond to the specific values ​​shown in other examples, such as those in FIG. 4(A), although similar in concept. The calibration DICs input values ​​in the left column of FIG. 4(B) range from 1 to 1024 pixel coordinates in 0.1 pixel increments, and the corresponding measured distance ZOUT (in microns) is input in the right column. During operation, a stored calibration lookup table is referenced to determine the measurement distance (in microns) corresponding to the measured DIC calculated by the CRS system. If the measured DIC falls between adjacent calibrated DIC values, the measurement distance can be determined by interpolation. The specific values ​​in the example of Figure 4(B) represent several narrow ranges around DICs, each containing approximately 104,604,990 pixel locations, with corresponding measurement distances in the ranges of approximately 37 μm, 381 μm, and 486 μm, respectively.

[0042] During operation (e.g., to measure the distance to one work surface 190-1 in FIG. 3 ), the optical pen 120 is connected to the electronics unit 160 and operatively positioned relative to the work surface 190-1 to perform a measurement operation. The measurement operation includes the optical pen 120 receiving an input spectral profile from the illumination source 164, outputting corresponding illumination to the work surface 190-1, receiving reflected light from the work surface 190-1, and outputting the reflected light to provide an output spectral profile to the CRS wavelength detector 162, which then provides output spectral profile data. The output spectral profile includes a distance-dependent profile component and a distance-independent profile component. The distance-dependent profile component includes a wavelength peak (e.g., peak 302 in FIG. 3 ) that indicates the measured distance (e.g., measured distance Z1) from the optical pen 120 to the work surface 190-1. As described above, the measurement distance (e.g., measurement distance Z1) corresponding to the measured DIC determined according to the center of gravity calculation by the CRS system is determined by referring to the stored calibration data (e.g., FIG. 4(A) or 4(B)). If the measured DIC falls between adjacent calibrated DIC values, the measurement distance corresponding to the measured DIC is determined by interpolation (e.g., between the measurement distances corresponding to the adjacent calibrated DIC values).

[0043] In contrast to the example of FIG. 3 relating to determining a measured distance to a single workpiece surface, as described in more detail below, FIG. 5A illustrates an example of determining measured distances to first and second workpiece surfaces (e.g., measured distances Z1 and Z2, respectively, to workpiece top surfaces 190-1 and 190-2 in FIG. 1). Such techniques (referred to herein as "centroid-type techniques") can be used to determine workpiece thickness (e.g., based at least in part on the difference between measured distances Z1 and Z2' as a function of DICs determined from a centroid calculation). However, as described in more detail below (e.g., with particular reference to FIGS. 6-9H), in accordance with the principles disclosed herein, utilizing certain alternative techniques for workpiece thickness determination can provide certain advantages (e.g., including more accurate determination of the thickness of a given workpiece).

[0044] 5A is a wavelength dispersion intensity profile diagram 500A showing effective wavelength peaks 502-1 and 502-2 from a CRS system. Wavelength peak 502-1 is generated by a first subset of measurement profile signals MS(p) that represent first wavelength light that is focused on and reflected from a first measurement surface (e.g., surface 190-1 of workpiece 170). Effective wavelength peak 502-2 is generated by a second subset of measurement profile signals MS(p) that represent second wavelength light that is focused on and reflected from a first measurement surface (e.g., surface 190-2 of workpiece 170). In this example, the workpiece (e.g., workpiece 170) is transparent, allowing light to pass through first workpiece surface 190-1 and be visible to second workpiece surface 190-2 (a portion of light from optical pen 120 passes through first measurement surface 190-1 and reflects off second workpiece surface 190-2).

[0045] Each measurement profile signal MS(p) has a signal level (expressed as a normalized voltage value) corresponding to each pixel p of a detector array (e.g., detector array 163). Each effective wavelength peak 502-1 and 502-2 has sufficient height (good signal-to-noise ratio) and is relatively symmetrical, allowing good positioning of each peak and determination of distance-indicating coordinates (DICs) 504-1 and 504-2 along the measurement axis of the detector array (i.e., using a centroid calculation). FIG. 5(A) shows the bias signal level (MVbias) (at normalized voltages), peak pixel coordinates (ppcl and ppc2), and a data threshold (MVthreshold) that defines the lower limit of the distance-indicating subset of the measurement profile signal MS(p) that forms the corresponding wavelength peak 502-1 and 502-2. All values ​​(e.g., including the "MV" value) are normalized voltages.

[0046] During operation (e.g., for measuring two workpiece surfaces), the optical pen 120 is connected to the CRS electronics 160 and operatively positioned relative to the workpiece 170 (i.e., including workpiece surfaces 190-1 and 190-2) to perform a measurement operation. The measurement operation includes the optical pen 120 receiving an input spectral profile from the illumination source 164, outputting corresponding illumination to the workpiece surfaces 190-1 and 190-2, receiving reflected light from the workpiece surfaces 190-1 and 190-2, and outputting the reflected light to provide an output spectral profile to the CRS wavelength detector 162, which then provides output spectral profile data. The output spectral profile includes a distance-dependent profile component and a distance-independent profile component. The distance-dependent profile component includes wavelength peaks (e.g., peaks 502-1 and 502-2 in FIG. 5A) that indicate the measured distances from the optical pen 120 to the workpiece surfaces 190-1 and 190-2, respectively. As described above, the CRS system determines measurement distances (e.g., measurement distances Z1 and Z2') corresponding to the measured DICs 504-1 and 504-2 calculated (i.e., using center of gravity calculations) by referencing stored calibration data (e.g., FIG. 4(A) or 4(B)). If either or both of the measured DICs fall between adjacent calibration DIC values, the measurement distance corresponding to the measured DIC is determined by interpolation. In various embodiments, according to conventionally known techniques, the thickness of the workpiece 170 is measured according to the difference between the determined measurement distances, as will be described in detail later with reference to FIG. 5(B).

[0047] Figure 5(B) is a diagram 500B illustrating the measurement distances to top surface 190-1 and bottom surface 190-2 of transparent workpiece 170, including actual measurement distance Z2 and apparent measurement distance Z2' to bottom surface 190-2. As shown in Figure 5(B), light rays R1A and R1B (e.g., from optical system 150) are shown to be focused on workpiece surface 190-1 (i.e., according to the wavelength light best focused on workpiece surface 190-1), resulting in wavelength peak 502-1 with corresponding DIC 504-1 that indicates measurement distance Z1 in Figure 5(A). Similarly, as shown in Figure 5(B), light rays R2A and R2B (e.g., from optical system 150) are shown to be focused on work surface 190-2 (i.e., according to the wavelength light best focused on work surface 190-2), resulting in wavelength peak 502-2 with corresponding DIC 504-2 that indicates apparent measurement distance Z2' in Figure 5(A). Apparent measurement distance Z2' differs from actual measurement distance Z2 due to the refractive index of the work material at the corresponding wavelength (i.e., wavelength peak 502-2).

[0048] More specifically, as shown in FIG. 5B, once rays R2A and R2B enter workpiece 170, they transmit through the workpiece material, as shown by rays R2A' and R2B', with an angular change depending on the refractive index, resulting in focusing at workpiece surface 190-2 corresponding to actual measurement distance Z2. However, expected rays R2A-P and R2B-P in the workpiece (i.e., no angular change) exhibit focusing corresponding to apparent measurement distance Z2', coinciding with wavelength peak 502-2 with corresponding DIC 504-2. Several mathematical equations describing the relationships between various distances are discussed in detail below.

[0049] The relationship between the actual measurement distances Z1 and Z2 and the actual workpiece thickness d is expressed by the following equation (3).

[0050]

number

[0051]

number

[0052]

number

[0053] As another example, FIG. 5A shows a dotted image of a wavelength dispersion intensity profile from a CRS system, including wavelength peaks 502-1a and 502-2a and corresponding peak pixel coordinates ppc1a and ppc2a. Wavelength peaks 502-1a and 502-2a are similar to wavelength peaks 502-1 and 502-2, except that the peaks in measurement profile signal 510a and the corresponding first and second subsets are partially merged due to their proximity to one another. This example corresponds to a relatively thin version of workpiece 170 (i.e., a workpiece with top and bottom surfaces that are relatively close to one another and have a small spacing between them). This can pose particular challenges when using conventional techniques (e.g., using centroid calculations to determine DICs) to determine the measurement distance to each surface and, depending on the difference in the determined measurement distances, to determine the workpiece thickness. For example, in some embodiments, the process described above may result in the evaluation of partially merged wavelength peaks 502-1a and 502-2a, which may result in the determination of a single DIC 504a (e.g., according to the above-described center of gravity calculations relating to Equations (1) and (2)), and accordingly, only a single measured distance to the workpiece. In other embodiments, the process may result in the determination of two DICs. However, it will be appreciated that the partial merger of wavelength peaks 502-1a and 502-2a may, under certain conditions, result in an inaccurate determination of the corresponding DICs. And / or, under certain conditions, an error message may be issued (e.g., indicating a problem with the determination of the corresponding DICs). In any of the above situations, the determination of the corresponding workpiece thickness may be challenged (e.g., certain inaccuracies or inability to be determined, etc.).

[0054] According to various embodiments of the present invention, a CRS system is provided with a novel approach for determining workpiece thickness, providing high measurement accuracy even for relatively thin workpieces (e.g., only 7 μm, 5 μm, or 3 μm thick in various embodiments). In particular, in certain embodiments, CRS electronics 160 ( FIG. 1 ) may include a thickness measurement unit 169 that may be utilized by CRS system 100 to perform workpiece thickness measurements in accordance with the principles disclosed herein. It may also include a distance measurement unit 172 that may be utilized by CRS system 100 to perform one or more distance measurements to one or more surfaces of the workpiece (e.g., by conventionally known techniques). In various embodiments, selection of a thickness measurement mode or operation (e.g., using thickness measurement unit 169) or selection of a distance measurement operation or mode (e.g., using distance measurement unit 172) may be performed automatically or in response to user input received via user interface 171. For example, depending on the positioning of the optical pen relative to the (transparent) workpiece, a thickness measurement operation or mode thereof may be automatically or manually selected to determine workpiece thickness.

[0055] As described in more detail below, in various embodiments of thickness measurement operations or modes thereof, the signal processor 166 may execute an algorithm (e.g., including a transform or other process) as stored in the thickness measurement section 169 of the memory section 168. The algorithm (e.g., utilizing a transform process) is executed on the spectral profile data output from the wavelength detector 162 to determine the workpiece thickness. During operation, the optical pen 120 is connected to the CRS electronics 160 and operatively positioned relative to the workpiece 170 to perform a thickness measurement operation. The thickness measurement operation involves the optical pen 120 receiving an input spectral profile from the illumination source 164 and outputting corresponding illumination to two surfaces 190-1 and 190-2 (i.e., "Surface 1" and "Surface 2") of the workpiece 170. The optical pen 120 also receives reflected light from the two surfaces and outputs the reflected light to the CRS wavelength detector 162. The CRS wavelength detector 162 then provides output spectral profile data, and based on processing of the output spectral profile data (e.g., utilizing a transform such as a Fourier transform or a Fourier-related transform), the workpiece thickness is determined.

[0056] 6 is a flow diagram illustrating an exemplary method 600 of operation of the CRS system 100 for determining workpiece thickness. In block 610, a CRS system is provided that includes an optical pen 120 exhibiting on-axis chromatic dispersion and having a confocal optical path including an optical system 150 configured to focus different wavelengths of light at different distances near the workpiece 170, an illumination source 164, and a CRS wavelength detector 162 having a plurality of pixels positioned at respective positions along a measurement axis of the CRS wavelength detector, the plurality of pixels receiving respective wavelengths of light and providing output spectral profile data. At block 620, the CRS system 100 is operated with an optical pen operatively positioned relative to the workpiece for a measurement operation, where the measurement operation includes the optical pen receiving an input spectral profile, outputting responsive illumination to a first work surface and a second work surface of the workpiece, receiving reflected light from the first work surface and the second work surface, and outputting the reflected light to a CRS wavelength detector capable of providing output spectral profile data.

[0057] At decision block 630, it is determined whether one or more measurement distances should be determined. If one or more measurement distances are not to be determined, the procedure proceeds to block 650, described below. If one or more measurement distances are to be determined, the procedure proceeds to block 640, where the output spectral profile data is processed to determine one or more measurement distances to the workpiece. (For example, determining the measurement distances may include determining distance-indicating coordinates that correspond to pixel locations of the CRS wavelength detector that indicate the measurement distances.)

[0058] After block 640, or if one or more measurement distances are not determined in block 630, the procedure proceeds to block 650, where the power spectral profile data is processed to determine a workpiece thickness as a function of the separation between the first and second workpiece surfaces. (For example, the thickness determination process may be independent of the determination of the measurement distances to the workpiece and / or may utilize a transform process, such as a Fourier transform process or a Fourier transform-related process.) In various embodiments, Fourier transform-related operations may include linear transformations of functions associated with Fourier analysis (e.g., the transformation may map a function to a set of coefficients of a basis function, such as a sine function or other functions localized in the frequency spectrum). Some examples may include cosine transforms, sine transforms, Laplace transforms, etc.

[0059] As noted above (e.g., with respect to the dotted measurement profile signal 510a in FIG. 5(A) ), in some embodiments, it is difficult to accurately determine the distance indicating coordinates of the opposite side of a workpiece (e.g., a thin workpiece that produces wavelength peaks that at least partially merge). In some cases, multiple wavelength peaks may merge to some extent, and only one distance-indicating coordinate can be determined. In other conditions, even if two distance-indicating coordinates are determined, inaccuracies and / or processing problems may occur (e.g., the centroid calculation of the first wavelength peak to determine the corresponding first distance-indicating coordinate may be partially shifted or otherwise affected by the merged second wavelength peak, or vice versa). Because of these potential challenges, it may be desirable in some cases to be able to determine distance workpiece thickness without having to determine the distance using a center of gravity calculation as shown in the previously known techniques described above. As described in more detail below, an alternative technique for determining workpiece thickness (e.g., a method using a conversion process) is presented herein.

[0060] In various embodiments, one or more distances determined by conventional techniques may still provide useful information (e.g., in some circumstances, conventional techniques may be utilized in combination with thicknesses determined in accordance with the principles presented herein). For example, as described in more detail below (with respect to Equations (6)-(10)), in some cases, a measured distance to the workpiece determined (e.g., at block 640) may be utilized in combination with a workpiece thickness determined in accordance with the principles presented herein (e.g., at block 650).

[0061] As noted above, in some cases, wavelength peaks corresponding to opposing workpiece surfaces merge to some extent, and only one wavelength-indicating coordinate (e.g., DIC 504a in FIG. 5A) is determined. In some embodiments, such a single wavelength-indicating coordinate (e.g., DIC 504a in FIG. 5A) may represent a measurement distance, such as distance Z3' in FIG. 5B. Distance Z3' is between distances Z1 and Z2' (e.g., may also correspond to the average of distances Z1 and Z2' and may extend to the midpoint of apparent thickness d'). This distance Z3' is related to distance Z3, which is the same as a point in workpiece 17 between first workpiece surface 190-1 and second workpiece surface 190-2. In some embodiments, the point in workpiece 17 corresponding to distance Z3 may be the midpoint between first workpiece surface 190-1 and second workpiece surface 190-2. Therefore, distance Z3 corresponds to the average of first distance Z1 and second distance Z2 to first work surface 190-1 and second work surface 190-2, respectively (e.g., distance Z3 extends to the midpoint of actual thickness d).

[0062] For example, in the dotted measurement profile signal 510a in FIG. 5A corresponding to a thin version of the workpiece 170, if the wavelength peaks 502-1 and 502-2 (e.g., having peak pixel coordinates ppc1a and ppc2a) corresponding to the workpiece surfaces 190-1 and 190-2 have similar characteristics (e.g., similar magnitudes), performing a centroid or similar calculation will result in a single distance-indicating coordinate (e.g., DIC 504a) that is approximately / nominaly centered between the two wavelength peaks. In such a case, the corresponding measurement distance (e.g., distance Z3') will be approximately the average of the first distance Z1 and the second distance Z2'. In some embodiments, the measurement distance so determined (e.g., distance Z3') may be used in combination with the determined value of the workpiece thickness d (e.g., determined in block 650) to determine the first distance Z1 and / or the second distance Z2 to the first workpiece surface 190-1 and the second workpiece surface 190-2. For example, if we assume that distance Z3' is the midpoint between distance Z1 and distance Z2', if we assume that apparent thickness d' corresponds to the difference between distance Z1 and distance Z2' (denoted by d' = Z2' - Z1), and if we assume that thickness d of workpiece 170 corresponds to the spacing between workpiece surfaces 190-1 and 190-2 (denoted by d = Z2 - Z1), then the following equations indicate the relationships between the various distances:

[0063]

number

[0064]

number

[0065] As noted above, in some cases, conventional processing techniques may generate two DICs corresponding to two partially merged wavelength peaks. In some cases, the two determined DICs may be inaccurate (e.g., due to peak merging). In some such examples (e.g., assuming that the two DICs are shifted toward the center by a similar amount due to peak merging), the two measured distances corresponding to the two DICs may be averaged (e.g., similar to distance Z3') to determine an approximate median measured distance. This may then be used in combination with the determined workpiece thickness d (e.g., as determined in block 650) to provide more accurate determinations of distances Z1 and Z2 (e.g., using processing such as Equation (10)).

[0066] In some embodiments, a selection may be made and / or processing may be performed to determine whether the workpiece thickness should be determined for a particular output spectral profile data set according to the principles set forth herein or according to an alternative, conventionally known method (e.g., using a centroid calculation to determine the difference between multiple distance indicating coordinates and their corresponding measurement distances, as described above with respect to DICs 504-1 and 504-2 in FIG. 5(A), equations (4) and (5)). For example, for certain workpieces (e.g., relatively thin workpieces with wavelength peaks spaced slightly apart, creating challenges such as peak merging), thickness determination may be performed using the techniques described herein by user selection and / or (e.g., automatic) processing. Alternatively, for certain workpieces (e.g., relatively thick workpieces with relatively sufficient separation between wavelength peaks to avoid significant peak merging issues), thickness determination may be performed using conventionally known methods, either by user selection and / or by (e.g., automatic) processing. In some embodiments, such a determination may be made based, at least in part, on processing the output spectral profile data using both conventionally known methods and the methods described herein, and comparing / analyzing the different results to determine which thickness determination method to select / use (e.g., most accurate, practical, etc.) for a particular embodiment (which may depend on additional factors, such as the measurement range of the particular optical pen / CRS system being utilized).

[0067] 7A-7D illustrate a process for measuring workpiece thickness using a transform process as part of a deconvolution process. Deconvolution is used to improve the process for determining the wavelength positions of two peaks that are unclear and / or not sufficiently separated from one another. In particular, in one embodiment, deconvolution is used to reduce the peak width, allowing for a better and more accurate determination of the distance between the two peaks (e.g., by corresponding calculations, etc.).

[0068] FIG. 7A shows an intensity profile 701 (or "output spectral profile data") output from wavelength detector 162. Similar to FIGS. 3, 5A, and 10A-10F described above, this intensity profile 701 is comprised of signal level or signal intensity (vertical axis) expressed as a function of pixel coordinates or "pixel position" (horizontal axis). Intensity profile 701 includes a first wavelength peak 702-1 and a second wavelength peak 702-2 corresponding to a first workpiece surface and a second workpiece surface (e.g., workpiece surfaces 190-1 and 190-2 in FIG. 1).

[0069] In various embodiments, the transformation process performed to determine the thickness of the workpiece (e.g., workpiece 170) may include calculating a transform (e.g., a Fourier transform) of the output spectral profile data 701, as shown in Figure 7(B). The Fourier transform of the output spectral profile data may then be divided by the Fourier transform of the reference (e.g., as shown in Figure 7(C)), resulting in the Fourier transform of the target spectral profile data. In various embodiments, the criterion may be a mathematical criterion function (e.g., a Lorentzian criterion, etc.). In other embodiments, the criterion may be, for example, output spectral profile data acquired for a "reference" workpiece (e.g., a top surface positioned the same distance from the optical pen as workpiece surface 190-1). The reference workpiece may be a thick glass workpiece, such that reflections from only the top surface appear in the output spectral profile data (e.g., the bottom surface of the thick glass workpiece may be outside the measurement range of the optical pen, etc., such that reflections from the bottom surface are not included in the output spectral profile data). Finally, the inverse Fourier transform of the Fourier transform of the target spectral profile data is calculated, which recovers the first wavelength peak 702-1x and the second wavelength peak 702-2x as shown in Figure 7(D) and makes their pixel locations easier / more accurately determined, thereby enhancing and improving the determination of the corresponding pixel locations of the two original peaks 702-1 and 702-2 contained in the output spectral profile data of Figure 7(A). In various embodiments, distance Z1 and distance Z2' may be determined from the position of the peak indicated by the pixel position by referring to a distance calibration lookup table (e.g., as described above with respect to Figures 4(A) and (B)), and the work thickness may be determined according to calculations such as those shown in equations (4) and (5).

[0070] Additional techniques, described in more detail below with reference to FIGS. 8 and 9A-H, are presented here. In some embodiments, these techniques offer several advantages over techniques such as those described with reference to FIGS. 7A-7D. For example, while deconvolution-type techniques such as those described with reference to FIGS. 7A-7D improve and strengthen the determination of pixel locations of two peaks in the output spectral profile data, in some embodiments, division by a reference (e.g., FIG. 7C) can increase the signal portion associated with high-frequency noise contained in the Fourier transform of the profile data. While such high-frequency noise can sometimes be mitigated with the use of a low-pass filter, fine-tuning the low-pass filter can present certain challenges (e.g., fine-tuning the filter must balance narrow-band peaks and signal stability). In various embodiments, at least some of these potential challenges can be addressed by alternative techniques, such as those described in more detail below with reference to FIGS. 8 and 9A-9H.

[0071] FIG. 8 illustrates one embodiment of a thickness measurement process using a transform process (e.g., including a Fast Fourier Transform (FFT) process). Some curves shown in FIG. 8 are similar and / or related to some of the curves shown in FIGS. 7(A)-7(D), as explained in more detail below. FIG. 8 shows a Fourier transform curve 801 of the original data. This curve 801 is the Fourier transform of the intensity profile (or "output spectral profile data"), and is expressed as a function of frequency (pixel -1 ) (horizontal axis) and the determination of a Fourier transform such as curve 801 is similar to the determination of the Fourier transform curve of FIG. 7(B). Deconvolution output curve 802 shows the result of processing similar to that described above with respect to Figure 7(C) above, where Fourier transform curve 801 is divided by the reference Fourier transform, and in this example, no low-pass filter is used to reduce high frequency noise, resulting in the noise content in curve 802 increasing (e.g., exponentially or otherwise) with high frequency content. Techniques for avoiding this issue are discussed in more detail below with reference to curve 803.

[0072] Regarding curve 803, if this is a profile including two wavelength peaks of equal intensity according to two criteria (i.e., from two workpiece surfaces), it will be understood that, as shown in FIG. 8 and in accordance with the following equation (11), the measurement distance Z1 corresponding to the top surface and the apparent measurement distance Z2' corresponding to the bottom surface are included at equal distances, i.e., +d' / 2 and -d' / 2, from their corresponding midpoints (e.g., coinciding with the measurement distance Z3' of the apparent midpoint) (see, for example, equations (8) and (9)). It will also be understood that the ideal Fourier transform of intensity profile data including two criteria of equal intensity is the cosine Fourier transform curve 803.

[0073]

number

[0074] In various embodiments, the processes of FIGS. 7(A)-7(D) can be characterized as attempting to recover a cosine curve similar to cosine curve 803 of FIG. 8, although the low-pass filter described above is required to mitigate the increased high frequency noise shown in curve 802, and particular challenges may arise in fine-tuning the low-pass filter. Since the ideal end result is expected to be a cosine curve according to the principles presented here, an alternative approach is to determine the first dip in the absolute value of the Fourier transform (e.g., the absolute value of an FFT that outputs a complex number), which indicates the cosine period and the corresponding workpiece thickness according to the following equation (12):

[0075]

number

[0076] For some specific values ​​exemplified with respect to Equation (12) and the description of Figure 8, if the workpiece material is fused silica with a refractive index n = 1.46, and the workpiece is positioned relative to the optical pen in a range where ∂z / ∂λ = 5.49 μm / pixel, then according to Equation (12), the thickness d is determined as follows: d = (1.46 (5.49 μm / pixel)) / (2 (0.16 / pixel)). Thus, according to various exemplary embodiments of the present invention, a transform process (e.g., a fast Fourier transform (FFT) process) may be used to determine the first dip 804 (e.g., indicating a cosine period) in the Fourier transform curve 801, and based thereon, the workpiece thickness "d" can be determined (e.g., following calculations according to the example of Equation (12)).

[0077] In various embodiments, f dip may be determined according to a dip detection algorithm (e.g., similar to or included in a peak detection algorithm that, in some embodiments, is repurposed as a dip detection algorithm). In various embodiments, such an algorithm may include operations such as taking derivatives and finding zero-crossing points (e.g., the first zero crossing that coincides with the first dip). In various embodiments, false peaks / dip's, such as those caused by noise or similar factors, may also be suppressed or addressed by smoothing and / or thresholding.

[0078] As noted above, in various embodiments, certain techniques described herein (e.g., as shown in Equation (12)) may enable accurate thickness measurements / determinations for workpieces thinner than can be measured using conventionally known techniques. As examples, several calculations for the minimum thickness measurable using conventionally known techniques will be described in more detail below and compared to the minimum thickness measurable using the techniques described herein.

[0079] The following numerical examples relate to conventionally known centroid-type techniques, such as those described above with respect to distance-indicating coordinates 504-1 and 504-2 in FIG. 5(A), that determine DICs according to calculated centroid values. In one specific embodiment, an optical pen for a CRS system can be provided in which the blue wavelength portion of the range has ∂z / ∂λ=8 μm / pixel and the red wavelength portion of the range has ∂z / ∂λ=2 μm / pixel. In one embodiment, using the conventionally known centroid-type techniques described above, it may be determined that wavelength peaks can be measured (i.e., pixel locations can be accurately determined) when the peaks are separated by at least a full width at half maximum (FWHM) value corresponding to FWHM=3 pixels in the blue wavelength portion of the range and FWHM=6 pixels in the red wavelength portion of the range. In such an embodiment, when the refractive index of a glass workpiece is n=1.5, the minimum thickness of the workpiece that can be accurately measured is d min = n(∂z / ∂λ)(FWHM), and the above value in the blue wavelength part of the range is d min = 36 μm, and the above value in the red wavelength part of the range is d min =18μm.

[0080] In comparison, with the specific technique presented here (e.g., as shown in Equation (12) for the "first dip" type specific technique), assuming the use of the same CRS system with the same optical pen, the minimum thickness of the workpiece that can be accurately measured corresponds roughly to the Nyquist frequency, f dip is approximately equal to 0.5, so that equation (12) becomes d min =n(∂z / ∂λ), and the blue wavelength portion of the range is d min = 12 μm (i.e., d in the conventionally known centroid-type technique described above) min = about 1 / 3 of 36 μm or 33.3%), and the red wavelength portion of the range is d min = 3 μm (i.e., d in the conventionally known centroid-type technique described above) min = about 1 / 6 or 16.7% of 18 μm.

[0081] Additionally, according to certain other techniques described herein (e.g., as described above with reference to FIGS. 7(A)-7(D) for certain "deconvolution" type techniques), assuming the use of the same CRS system with the same optical pen, in one embodiment, the minimum thickness of a workpiece that can be accurately measured corresponds to a peak separation of approximately (0.85) FWHM, with the blue wavelength portion of the range being d min = 30.6 μm (i.e., d in the conventionally known centroid-type technique described above) min = 85% of 36 μm, and the red wavelength portion of the range is d min = 15.3 μm (i.e., d in the conventionally known centroid-type technique described above) min It has been experimentally determined that the thickness of the thin film may be 85% of 18 μm. Given the complexity of CRS systems, it will be appreciated that such improvements in the measurable thickness range may be difficult to achieve and may be advantageous in certain implementations, particularly where accurate measurement of thin workpieces and / or thin portions of the workpiece is required.

[0082] Further, additional examples of first dips determined for the "first dip" type technique described in Figure 8 are shown in Figures 9(A)-9(D). The examples of Figures 9(A)-9(D) illustrate that regardless of the position of a workpiece (e.g., a workpiece having a thickness of 25 μm) within the measurement range operable for the optical pen, the processing method of Figure 8 results in the Fourier transform of the signal having a first dip, and the position of this first dip can be used to determine the workpiece thickness (e.g., according to a calculation such as that shown in Equation 12). In the examples of Figures 9(A)-9(D), the workpiece 170 is placed on the stage 175 of the CRS system 100 (see Figure 1) so that its top surface 190-1 is positioned 300 μm, 400 μm, 500 μm, and 800 μm, respectively, from the measurement reference position RP of the optical pen 120 of the CRS system 100. (That is, each wavelength peak is shown to have a different corresponding relative location within the pixel / wavelength range of the detector, as indicated by its corresponding X-axis position.)

[0083] 9(A)-9(D) show corresponding intensity profiles 902A, 905A, 908A, and 911A output from wavelength detector 162, which, similar to FIGS. 3, 5A, 7A, and 10A-10F above, consist of signal level or intensity (vertical axis) expressed as a function of pixel coordinate or "pixel position" (horizontal axis). Note that in the red wavelength portion of the CRS system 100's Z-range far from optical pen 120, as in FIGS. 9(C) and 9(D), the two wavelength peaks of the intensity profiles are at least somewhat recognizable / distinguishable, whereas in the blue wavelength portion of the CRS system 100's Z-range near optical pen 120, as in FIGS. 9(A) and 9(B), the two wavelength peaks of the intensity profiles appear completely merged. 9(A)-9(D) also show intensity profiles 903A, 906A, 909A, and 912A corresponding to a "reference" workpiece, which may be a thick glass workpiece, so that reflections from only the top surface of the reference workpiece appear in the intensity profile, as described above in FIG.

[0084] Figure 9(E) shows the Fourier transforms 902B / 903B of the intensity profiles 902A / 903A of Figure 9(A). Figure 9(F) shows the Fourier transforms 905B / 906B of the intensity profiles 905A / 906A of Figure 9(B). Figure 9(G) shows the Fourier transforms 908B / 909B of the intensity profiles 908A / 909A of Figure 9(C). Figure 9(H) shows the Fourier transforms 911B / 912B of the intensity profiles 911A / 912A of Figure 9(D).

[0085] Figures 9(E)-9(H) show the Fourier transforms 902B, 905B, 908B, and 911B of the intensity profile of the measured workpiece, including first dips 904, 907, 910, and 913, regardless of the position of the workpiece relative to the optical pen 120 of the CRS system 100, and these first dips can be used to determine the workpiece thickness as described above. On the other hand, it should be noted that the Fourier transforms 903B, 906B, 909B, and 912B of the intensity profiles of the reference workpieces do not contain any significant dips. In various embodiments, it has been experimentally determined that the "first dip" type techniques described herein (e.g., including calculations such as those shown in Equation (12)) can be used to determine workpiece thickness (e.g., 25 μm thickness) with little error (e.g., less than 10% or less than 5%) over most of the wavelength range (e.g., 90% or 75%) corresponding to the operating range of a CRS system including an optical pen. This determination includes determinations using data such as those shown in FIGS. 9(A)-9(D), particularly data where wavelength peaks merge (e.g., as in FIGS. 9(A) and 9(B) where the two peaks visually appear as a single peak).

[0086] The principles underlying the operation of the first-dip techniques described herein (e.g., involving the use of transforms such as Fourier transforms, which are sums of sine waves) may make such techniques particularly effective in determining workpiece thicknesses corresponding to the spacing between first and second workpiece surfaces, as discussed above. However, in some embodiments, such techniques may be less effective at simultaneously determining thicknesses between multiple layers in workpieces having multiple layers (i.e., workpieces including top and bottom workpiece surfaces and additional, intermediate workpiece surfaces as separate layers). Therefore, in certain embodiments, corresponding thickness measurement operations and / or modes thereof performed in accordance with the principles described herein may only be capable of determining thicknesses corresponding to the spacing between first and second workpiece surfaces, but may not be capable of determining spacings between three or more workpiece surfaces (e.g., multi-layer structures).

[0087] In some of the above-described embodiments, the workpiece thickness determined according to the techniques described herein can be used in combination with a distance to the workpiece determined according to conventional methods. For example, considering the illustration of two merged wavelength peaks in Figures 9(A)-9(D) (e.g., particularly Figures 9(A) and 9(B) where the two peaks visually appear as a single peak), it can be appreciated that in such cases, determining DICs using conventional centroid-type techniques may not accurately determine the two distances to the two surfaces, but can be used to accurately determine a distance to an approximate midpoint (e.g., distance Z3') (e.g., distance Z3' is determined by determining the centroid of the merged peaks and lies between distances Z1 and Z2'). Distances calculated in this manner (e.g., distance Z3') can be referenced or used in combination with thickness measurements performed according to the principles described herein (e.g., as described above for determining distance Z1 with respect to Equation (10)).

[0088] In some exemplary embodiments, the CRS system 100 may be included in and / or utilized with a machine system (e.g., a vision inspection system or a coordinate measuring machine, etc.). In some cases, the machine system may automatically perform corresponding functions, such as workpiece thickness measurement along the optical axis OA of the CRS system 100. Some examples of such machine systems include suitable machine vision inspection systems described in U.S. Patent No. 8,085,295 and U.S. Patent No. 7,454,053.

[0089] While preferred embodiments of the present invention have been described, many variations on the configurations and operational procedures shown herein will be apparent to those skilled in the art based on this disclosure. For example, while the particular techniques shown herein involve particular types of Fourier transform processing (e.g., FIGS. 7A-9H), it will be understood that other types of transform processing can similarly be utilized in accordance with the principles shown herein to achieve similar results (e.g., utilizing Fourier transform-related processes such as cosine transforms, sine transforms, Laplace transforms, etc.). As another example, while a CRS system including a chromatic point sensor (optical pen) is shown herein, a CRS system including a chromatic line sensor can also be configured to operate in accordance with the systems and methods shown herein. It will be understood that these and various other alternatives can be used to implement the principles shown herein. In addition, the various embodiments described above can be combined to provide further embodiments.

Claims

1. 1. A chromatic range sensor (CRS) system for determining workpiece thickness, comprising: The CRS system comprises: an optical pen having a confocal optical path including an optical system that exhibits axial chromatic dispersion and is configured to focus different wavelengths of light at different distances near the workpiece; an illumination source that generates multi-wavelength incident light having an input spectral profile that is input to the optical pen; a CRS wavelength detector having a plurality of pixels disposed at respective positions along a measurement axis of the CRS wavelength detector; The CRS system comprises: When the optical pen is at a position where it can perform a measurement operation on the workpiece, The optical pen The input spectral profile is input, outputting corresponding irradiation to a first work surface and a second work surface of the work; receiving reflected light from the first work surface and the second work surface; configured to output the reflected light to the CRS wavelength detector, which provides output spectral profile data; The CRS system further includes a signal processing unit that processes the output spectrum profile data to determine a workpiece thickness according to the distance between the first workpiece surface and the second workpiece surface, and the workpiece thickness determination process uses a conversion process; the transform process includes at least one of a Fourier transform process and a Fourier transform-related process; The conversion process is performed by: determining a transform of the output spectral profile data; and determining the workpiece thickness based at least in part on a first characteristic included in the transform of the output spectral profile data; The CRS system, wherein the first characteristic corresponds to a first dip in the transform of the output spectral profile data.

2. 2. The CRS system of claim 1, the output spectral profile data comprises distance-dependent profile components having a first wavelength peak and a second wavelength peak corresponding to the first work surface and the second work surface, respectively, and the determination of the work thickness is independent of determined values ​​of a first measured distance to the first work surface and a second measured distance to the second work surface.

3. 3. The CRS system according to claim 2, The CRS system, wherein the first wavelength peak and the second wavelength peak visually appear as a single peak in an image of the output spectral profile data provided by the CRS wavelength detector.

4. 2. The CRS system of claim 1, A CRS system characterized in that, for a workpiece in which the distance between the first workpiece surface and the second workpiece surface is 5 μm, the use of the conversion process enables the signal processing unit to determine the workpiece thickness with an accuracy of less than 10% error.

5. 2. The CRS system of claim 1, The CRS system, wherein the signal processing unit is further configured to process the output spectral profile data to determine a measurement distance to the workpiece.

6. 6. The CRS system according to claim 5, A CRS system characterized in that the process of determining the measurement distance to the workpiece includes processing the output spectrum profile data to determine a center of gravity that indicates the measurement distance to the workpiece, but the process of determining the workpiece thickness does not depend on the determined value of the center of gravity.

7. 6. The CRS system according to claim 5, The CRS system, wherein the determined measurement distance is a distance to a point inside the workpiece between the first workpiece surface and the second workpiece surface.

8. 8. The CRS system according to claim 7, The CRS system, characterized in that the measurement distance is determined according to a distance indicating coordinate that is an average of a first wavelength peak and a second wavelength peak corresponding to the first work surface and the second work surface, respectively.

9. 6. The CRS system according to claim 5, The CRS system is characterized in that the signal processing unit is configured to determine at least one of a first measurement distance to the first work surface and a second measurement distance to the second work surface using the determined measurement distance and the determined work thickness.

10. 2. The CRS system of claim 1, The conversion process is performed by: determining a Fourier transform of the output spectral profile data; dividing the Fourier transform of the output spectral profile data by a Fourier transform of a reference to obtain a Fourier transform of target spectral profile data; calculating an inverse Fourier transform of the Fourier transform of the target spectral profile data to determine first and second wavelength peaks corresponding to a first measured distance to the first workpiece surface and a second measured distance to the second workpiece surface, respectively; and determining a thickness of the workpiece based at least in part on a difference between the measured distances to the first workpiece surface and the second workpiece surface.

11. 2. The CRS system of claim 1, The CRS system is operable in multiple modes, including a thickness measurement mode, and for this purpose, the signal processing unit performs processing to determine the thickness as part of the thickness measurement mode.

12. 12. The CRS system of claim 11, The CRS system is characterized in that the multiple modes further include a distance measurement mode for measuring the distance to the workpiece, and the CRS system is configured such that during the distance measurement mode, when the optical pen is in a position where it can perform a measurement operation on the workpiece, the optical pen inputs the input spectral profile, outputs irradiation corresponding to the input spectral profile to the workpiece, receives reflected light from the workpiece, and outputs the reflected light to the CRS wavelength detector, which is capable of providing output spectral profile data, and the output spectral profile comprises distance-dependent profile components having wavelengths of light that indicate the measured distance to the workpiece.

13. 12. The CRS system of claim 11, The CRS system is characterized in that the thickness measurement mode can only determine the thickness depending on the distance between the first work surface and the second work surface, and cannot determine multiple distances between three or more work surfaces.

14. 2. The CRS system of claim 1, The CRS system is at least one of a chromatic point sensor system and a chromatic line sensor system.

15. 1. A method of operating a chromatic range sensor (CRS) system for determining workpiece thickness, comprising: The CRS system comprises: an optical pen having a confocal optical path including an optical system that exhibits axial chromatic dispersion and is configured to focus different wavelengths of light at different distances near the workpiece; an illumination source that generates multi-wavelength incident light having an input spectral profile that is input to the optical pen; the CRS wavelength detector having a plurality of pixels disposed at respective positions along a measurement axis of the CRS wavelength detector, the plurality of pixels being capable of receiving light of each wavelength and providing output spectral profile data; The method of operation comprises: operating the CRS system with the optical pen operatively positioned relative to the workpiece to perform a measurement operation, the measurement operation including the optical pen receiving the input spectral profile and outputting illumination responsive thereto onto a first work surface and a second work surface of the workpiece, receiving reflected light from the first work surface and the second work surface, and outputting the reflected light to the CRS wavelength detector, which provides the output spectral profile data; and processing the output spectrum profile data to determine a workpiece thickness according to a distance between the first workpiece surface and the second workpiece surface, wherein the workpiece thickness determination process utilizes a conversion process; the transform process includes at least one of a Fourier transform process and a Fourier transform-related process; The conversion process is performed by: determining a transform of the output spectral profile data; and determining the workpiece thickness based at least in part on a first characteristic included in the transform of the output spectral profile data; The method of operating, wherein the first characteristic corresponds to a first dip in the transform of the output spectral profile data.

16. 16. The method of claim 15, further comprising: the output spectral profile data comprises distance-dependent profile components having first and second wavelength peaks corresponding to the first work surface and the second work surface, respectively, and the determination of the work thickness is independent of determined values ​​of a first measured distance to the first work surface and a second measured distance to the second work surface.

17. 16. The method of claim 15, further comprising: The conversion process is performed by: determining a Fourier transform of the output spectral profile data; dividing the Fourier transform of the output spectral profile data by a Fourier transform of a reference to obtain a Fourier transform of target spectral profile data; calculating an inverse Fourier transform of the Fourier transform of the target spectral profile data to determine first and second wavelength peaks corresponding to a first measured distance to the first workpiece surface and a second measured distance to the second workpiece surface, respectively; and determining a thickness of the workpiece based at least in part on a difference between the measured distances to the first workpiece surface and the second workpiece surface.

18. 16. The method of claim 15, further comprising: The CRS system is operable in multiple modes, including a thickness measurement mode and a distance measurement mode for measuring the distance to the workpiece; performing a process to determine the thickness as part of the thickness measurement mode; The method of operation further comprises: operating the CRS system during the distance measurement mode with the optical pen operatively positioned relative to the workpiece to perform a measurement operation, wherein the measurement operation includes the optical pen receiving the input spectral profile, outputting illumination to the workpiece in response thereto, receiving reflected light from the workpiece, and outputting the reflected light to the CRS wavelength detector capable of providing the output spectral profile data comprising distance-dependent profile components having wavelengths of light indicative of a measured distance to the workpiece; and A method of operation comprising: processing the output spectral profile data to determine a distance to the workpiece.

19. 20. The method of claim 18, further comprising: The thickness measurement mode is characterized in that it is only possible to determine the thickness according to the distance between the first work surface and the second work surface, and is not possible to determine multiple distances between three or more work surfaces.

20. 1. A method of operating a chromatic range sensor (CRS) system, comprising: The CRS system comprises: an optical pen having a confocal optical path including an optical system that exhibits axial chromatic dispersion and is configured to focus different wavelengths of light at different distances near the workpiece; an illumination source that generates multi-wavelength incident light having an input spectral profile that is input to the optical pen; the CRS wavelength detector having a plurality of pixels disposed at respective positions along a measurement axis of the CRS wavelength detector, the plurality of pixels being capable of receiving light of each wavelength and providing output spectral profile data; The method of operation comprises: operating the CRS system with the optical pen operatively positioned relative to the workpiece to perform a measurement operation, the measurement operation including the optical pen receiving the input spectral profile and outputting illumination responsive to the input spectral profile onto a first work surface and a second work surface of the workpiece, receiving reflected light from the first work surface and the second work surface, and outputting the reflected light to the CRS wavelength detector providing the output spectral profile data; processing the output spectral profile data to determine a measured distance to the workpiece, wherein determining the measured distance includes determining distance indicating coordinates according to pixel locations of the CRS wavelength detector that represent the measured distance; and processing the power spectrum profile data to determine a workpiece thickness as a function of a distance between the first workpiece surface and the second workpiece surface; A method of operation characterized in that the process of determining the workpiece thickness does not depend on the determined value of the measured distance to the workpiece.

Images