Chromatic confocal measuring device and method for measuring the distance to an object

Abstract

A chromatic confocal measuring device for measuring the distance to an object (40; 50) contains a spectrograph (46) which converts charge carriers released in light-sensitive cells (49; 49') and having two different amplifications into electric voltages (V1, V2) such that two voltages are output (V1, V2) each time a light-sensitive cell (49; 49') is exposed. An evaluation unit (30) is configured to calculate a distance value for each exposure cycle taking into consideration the two voltages (V1, V2) generated by the light-sensitive cells (49; 49').

Description

[0001] Chromatic confocal measuring device and method for measuring the distance to an object

[0002] BACKGROUND OF THE INVENTION

[0003] 1. Field of the invention

[0004] The invention relates to a chromatic confocal measuring device and a method for measuring the distance to a workpiece or other object. The invention particularly relates to measurements in which the intensity of the measuring light reflected from the object varies considerably due to the nature and / or inclination of the surface.

[0005] 2. Description of the state of the art

[0006] Chromatic confocal distance measurement devices have been used in industrial metrology for many years when distances to workpieces or other objects need to be measured non-contact and with high accuracy. If the distances are measured at multiple measuring points, a surface profile of the object can be determined from the distance values. For transparent objects, such devices can also be used to measure wall thicknesses or other thicknesses, since a distance can usually be determined for each optical interface, and the thickness is calculated as the difference between the measured distances. A typical measurement task for such a device is to check the dimensional accuracy and surface quality of gears installed in transmissions.The permissible tolerances are small, as even minor deviations from the nominal dimensions and roughness lead to unwanted noise and increased wear. Conventional devices for chromatic confocal distance measurement contain a light source that generates broadband measuring light and directs it onto a pinhole with a very small aperture, often referred to as a diaphragm. The aperture is projected onto the surface to be measured by imaging optics. At least part of the imaging optics exhibits significant longitudinal chromatic aberration. Different-colored images of the aperture are therefore arranged one behind the other on the optical axis of the imaging optics.Only the spectral portion of the measuring light, where the image of the aperture lies exactly on the surface to be measured, is reflected back by the surface to be measured, allowing it to enter the same aperture or an aperture arranged optically conjugate to it. The wavelength of this spectral portion is recorded using a spectrograph. Each wavelength is assigned a specific distance from the imaging optics.

[0007] Although the other spectral components are also partially reflected by the object surface, they cannot be focused by the imaging optics into the same or an optically conjugated aperture and therefore do not contribute significantly to the part of the measuring light that is spectrally analyzed by the spectrograph.

[0008] The spectrograph typically contains a prism, a diffraction grating, or other dispersive element, as well as a linear array of light-sensitive cells (pixel array), each of which measures the intensity of the measuring light deflected in a specific direction by the dispersive element. The measuring light typically strikes multiple cells, not just a single one. The electrical signals generated by the cells are used to determine which cell exhibits the maximum intensity. Therefore, the cells quantitatively record the intensity of the incident measuring light with the highest possible accuracy, since errors in this intensity measurement directly translate into errors in the distance measurement. To obtain accurate measurements, the intensity of the incident measuring light must be high enough to allow the electrical signal generated by the cells to stand out from the signal noise.However, if the intensity is too high, some of the cells may become saturated, so that the intensity of the incident measuring light can no longer be determined quantitatively.

[0009] The proportion of specularly or diffusely reflected measuring light from the object's surface depends on the surface's properties, but also on its inclination relative to the measuring beam. If the surface is uniform throughout and the object is stationary during the measurement of a surface profile, the device can be positioned so that the measuring beam strikes the surface to be measured approximately perpendicularly or at least at approximately the same angle. The device's light source can then be adjusted so that the intensity of the reflected light is optimally matched to the spectrograph's sensitivity, preventing saturation of the light-sensitive cells.

[0010] However, such a feed is usually not possible with fast-moving objects. An example of such fast-moving objects are rotating gears that are to be measured by a stationary device. If the measuring beam is directed radially onto the rotating gear, the angle of incidence of the measuring beam on the surface of the teeth varies very rapidly. While the measuring beam hits the tip of a tooth perpendicularly at one measuring point, resulting in a very strong reflection, the next measuring point may be on the steeply sloping tooth flank, which reflects very little measuring light towards the imaging optics. The very different intensity of reflections can lead to the intensity of the received signal being too low at some measuring points and too high at others. In general, it is not possible to select the intensity so that none of the aforementioned effects occur during a measurement.DE 10 2019 107 188 A1 proposes a solution to this problem: controlling the light source so that the intensity of the measuring light generated is variably adapted to the measurement locations. Using the example of measuring a gear, this means, for example, that the intensity is reduced when the measuring beam hits the tip surface and maximized when the measuring beam hits the tooth flank. However, such a control requires that the geometry of the workpiece at the measurement location is at least roughly known at every measurement time. For a rotating gear with its highly jagged surface, even the smallest deviations from the target rotational frequency of the gear can make such a prediction impossible.

[0011] If the shape of the workpiece is unknown or cannot be predicted accurately enough, a control system could be used that automatically adjusts the optimal intensity of the measuring light. However, experiments have shown that such a control system cannot respond quickly enough to rapidly changing measurement conditions.

[0012] Regardless of whether the intensity of the measuring light generated by the light source is changed by means of a control or a regulation, fluctuating temperatures of the light source occur, which can have adverse effects on the spectral composition of the measuring light.

[0013] Instead of changing the intensity of the measuring light, the exposure time can also be changed during the measurement of a surface profile. For example, a measuring point on the tip surface of a gear tooth can be exposed for a shorter time, which releases fewer charge carriers in the light-sensitive cells and prevents saturation. However, this means that a consistently high measuring rate can no longer be achieved, which is disadvantageous for many measuring tasks. In addition, varying exposure times when the workpiece moves lead to smaller and larger measuring points, so that these are no longer equidistant. Similar difficulties arise when a device is used that uses two different measuring beams, as is known, for example, from DE 10 2019 122 866 A1.The diameters of the two apertures, which can also be designed as fiber ends, differ, causing the two measuring beams to produce measuring spots of different sizes on the surface. Alternatively or additionally, different exposure times can be selected. Here, too, the different parameters make it difficult to combine the two measurement signals into a grid of equidistant measuring points, as the two measuring beams impinge on the workpiece at different locations.

[0014] US 2004 / 0041077 A1 discloses a light-sensitive cell with a cascaded charge storage system for improving the dynamic range. The charges accumulated in a first capacitor flow into a second capacitor when the charge storage capacity of the first capacitor is exceeded.

[0015] SUMMARY OF THE INVENTION

[0016] The object of the invention is to provide a chromatic-confocal measuring device for measuring the distance to a workpiece, with which reliable measurement results can be achieved even when the intensity of the reflected measuring beam varies considerably, e.g., because the nature and / or inclination of the surface changes significantly from measuring point to measuring point. Furthermore, the object of the invention is to provide a method based on the chromatic-confocal measuring principle that is suitable for such measurements.

[0017] With regard to the device, this object is achieved by a chromatic confocal measuring device for measuring the distance to an object, wherein the measuring device has a light source for generating broadband measuring light, a first aperture for the passage of the measuring light, and imaging optics that images the first aperture onto the object. The imaging optics are chromatically uncorrected, so that images of the first aperture are created at different distances from the imaging optics that depend on the wavelength. A spectrograph spectrally analyzes measuring light that has been reflected from the object and passed through a second aperture. The second aperture coincides with the first aperture or is optically conjugated to the first aperture. The spectrograph has several light-sensitive cells in which charge carriers are released when measuring light strikes them. These charge carriers are converted into electrical voltages.An evaluation unit is configured to calculate the distance to the object from the electrical voltages generated by the spectrograph. According to the invention, the spectrograph is further configured to convert charge carriers released in the light-sensitive cells into electrical voltages with two different amplifications, so that two voltages are output each time a light-sensitive cell is exposed. The evaluation unit is configured to calculate a distance value for each exposure cycle, taking into account the two voltages generated by the light-sensitive cells.

[0018] Since, according to the invention, each light-sensitive cell generates two different electrical voltages obtained by converting radiation-induced charge carriers with different amplifications, it is possible to evaluate only one of the two voltages or both voltages for the distance calculation, depending on the intensity of the incident measuring light. For example, one of the electrical voltages can be evaluated for measuring points that reflect a lot of measuring light, and the other for measuring points that reflect little measuring light. In contrast to conventional HDR schemes (HDR stands for High Dynamic Range), the two electrical voltages originate from the same exposure cycle and thus from measuring light that was generated at the same measuring point during the same exposure cycle.In this way, a grid of equidistant measuring points can be advantageously obtained without having to change the intensity of the measuring light generated by the light source or the exposure time.

[0019] When we talk about a calculation taking voltages into account, this means that the voltages are fed to the evaluation unit and, in a further step, this unit decides on the basis of the supplied voltages whether to include one of the two voltages or both voltages together in the calculation of the distance value.

[0020] The evaluation unit determines the wavelength at which the intensity of the measuring light applied to the spectrograph is maximum from the intensity values ​​derived from the voltage(s) for each light-sensitive cell. Based on this wavelength, the evaluation unit calculates the distance value, as is well known in the art.

[0021] The evaluation unit can be configured to calculate at least some distance values ​​using an average of the two voltages, where the average can be arithmetic or weighted, for example. Such averaging is particularly useful in cases where so little measuring light reaches the light-sensitive cell that saturation, which would render one of the two voltages unusable, does not occur. The weighting can be selected such that the contribution of one voltage is smaller, and thus the contribution of the other voltage is larger, the closer one voltage is to a predetermined threshold.

[0022] A light-sensitive cell that provides two voltages can be realized, for example, by dividing the cell into two separate individual cells along a direction perpendicular to the pixel line. These cells convert the generated charge carriers into electrical voltages with different gains. However, this means that only a portion (e.g., half) of the measurement light is available to each individual cell as an input signal.

[0023] It is therefore preferred if each light-sensitive cell is assigned an analog circuit which is designed to (a) transfer the charge carriers to a first capacitance and to a second capacitance, (b) derive a first voltage with a first gain from the charge carriers in the first capacitance and (c) derive a second voltage with a second gain which is smaller than the first gain from the charge carriers in the second capacitance.

[0024] So instead of having two individual cells with their own photodiodes or other light-sensitive components, each cell has only one light-sensitive component but two capacitors that convert charge carriers into voltages with different gains.

[0025] The different gain values ​​can be achieved very easily by choosing different capacitances. This takes advantage of the fact that for a capacitance C, the voltage U per charge Q is inversely proportional to the capacitance C. Thus, an additional charge carrier generates a higher voltage increase in a small capacitance C1 than in a large capacitance C2 with C2 > C1. Alternatively, the different gains can also be achieved with the same capacitances C1 = C2 and different downstream amplifier circuits.

[0026] The capacitances will generally be capacitors that are part of an integrated circuit, since analog circuits with soldered SMD components require too much installation space. In one embodiment, the analog circuit in a cell constructed in this way is designed to transfer charge carriers that can no longer be absorbed by the first capacitance entirely to the second capacitance. If only a small amount of measuring light reaches such a cell, the first, smaller capacitance does not saturate, so the first voltage can be used to calculate the distance. If, on the other hand, a large amount of measuring light hits the cell, the first, smaller capacitance can saturate, while the larger, second capacitance still supplies a usable voltage.

[0027] Preferably, after the first voltage is read and before the second voltage is read, the charge carriers stored in the first capacitor are transferred to the second capacitor. This ensures that all charge carriers generated during exposure contribute to the measurement. In principle, however, it is also possible to allow the charge carriers stored in the first capacitor to flow out of the cell after the first voltage is read, so that they do not contribute to the second voltage.

[0028] Typically, the transfer of charge carriers between the first capacitance and the second capacitance is controlled by at least one transistor or similar switching element configured to direct charge carriers to the second capacitance instead of the first capacitance when the first voltage exceeds a predetermined threshold voltage. The threshold voltage is preferably set such that it is only exceeded when the first capacitance can no longer accommodate charge carriers due to saturation or when this condition is imminent.

[0029] To allow the evaluation unit to decide which voltages to use when calculating the distance, two thresholds can be set, enabling the following decision tree: ■ If the first voltage generated by the first capacitance with the higher gain is lower than a first threshold, only the first voltage is used in the calculation. This corresponds to the case where only a small amount of measurement light reaches the light-sensitive cell, so a high gain is desirable.

[0030] ■ If the first voltage lies between the first threshold and a second threshold that is greater than the first threshold, a (possibly weighted) average of the first voltage and the second voltage is used to calculate the distance. As mentioned above, the weighting can be set so that the contribution of the first voltage is smaller the closer the first voltage is to the second threshold, which corresponds to saturation. The contribution of the second voltage, generated by the second capacitance with the lower gain, then becomes correspondingly larger. This corresponds to the case where a lot of measuring light reaches the light-sensitive cell, but the first capacitance has not yet saturated.

[0031] ■ If the first voltage is above the second threshold, only the second voltage is used to calculate the distance. This corresponds to the case where the intensity of the incident measurement light is so high that the first capacitance saturates, and many charge carriers must be absorbed by the second capacitance, so the lower gain of the second capacitance is sufficient to obtain a sufficiently strong measurement signal.

[0032] Instead of the serial filling of the capacitances described above, the analog circuit can be configured to transfer a first portion of the charge carriers to the first capacitance and, in parallel, transfer a second portion of the charge carriers to the second capacitance. If the first voltage is below a threshold value, saturation has not occurred, so that the first voltage, the second voltage, or a (possibly weighted) average of the two voltages can be used to calculate the distance. Here, too, weighting can be performed such that the contribution of the first voltage is smaller the closer the first voltage is to the threshold value, which corresponds to saturation. If, on the other hand, the first voltage is above a threshold value, this indicates saturation of the first capacitance, so that only the second voltage is evaluated.

[0033] Chromatic confocal measuring devices can also be used to measure the thickness of transparent objects. When measuring the wall thickness of bottles, for example, the measuring light is reflected simultaneously from the front of the glass wall facing the measuring device and from the back of the glass wall. After passing through the second confocal aperture, only those spectral components of the measuring light contribute to the detected measurement signal for which the measuring light is focused either on the front or the back of the opaque object.

[0034] As a result, two spectrally separated high-intensity regions appear in the detected measurement signal. The corresponding measurement distance can be determined for each of these regions as described above. The thickness of the measured opaque object can be determined from the difference between the two measurement distances.

[0035] If the reflectivity of the measuring light for the boundary layers of the opaque object differs significantly, the problem arises that no amplification factor can be found that provides optimal signal quality for both measurement distances to be determined. At low amplification, the signal from the boundary layer with low reflectivity results in a poor signal-to-noise ratio. At high amplification, on the other hand, the signal from the boundary layer with high reflectivity results in oversaturation of the detector. In this case, the measuring device described here can advantageously be used to determine a distance value separately for each of the two spectral ranges in the manner described above by taking into account the two voltage values ​​generated by the individual light-sensitive cells.

[0036] The measuring device thus advantageously enables the highly precise measurement of opaque objects in which the reflectivity of the individual boundary layers for the measuring light differs significantly. An example of such an object is glass bottles made of glass types that absorb a large portion of visible light, so that only a very small portion of the measuring light reaches the back wall of the glass and can contribute to the measurement signal.

[0037] With regard to the method, the above-mentioned task is solved by a method for chromatic-confocal measurement of the distance to an object with the following steps:

[0038] ■ a light source generates broadband measuring light;

[0039] ■ the measuring light passes through a first aperture;

[0040] ■ an imaging optics images the first aperture onto the object, whereby the imaging optics are chromatically uncorrected, so that images of the first aperture are created at different distances from the imaging optics, which depend on the wavelength;

[0041] ■ a spectrograph spectrally analyses measuring light which has been reflected from the object and has passed through a second aperture, the second aperture coinciding with the first aperture or being optically conjugated to the first aperture, and the spectrograph having a plurality of light-sensitive cells in which charge carriers are released when measuring light strikes them and are converted into electrical voltages;

[0042] ■ Charge carriers released in the light-sensitive cells of the spectrograph are converted into electrical voltages with two different amplifications, so that two voltages are output each time a light-sensitive cell is exposed; and

[0043] ■ An evaluation unit calculates a distance value for each exposure cycle, taking into account the two voltages generated by the light-sensitive cells.

[0044] The advantageous embodiments explained above for the device apply accordingly to the method.

[0045] The device and method can generally be extended to more than two amplification stages. For example, if a measurement object has areas with such low reflectivity that even with high amplification the signal-to-noise ratio is too poor, a third amplification stage with even higher amplification could be added and used to further increase the dynamic range of the measurement device.

[0046] In general, the evaluation of two voltages according to the invention can also be applied to measuring devices according to other measuring principles, for example to laser triangulators.

[0047] BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Exemplary embodiments of the invention are explained in more detail below with reference to the drawings. These show: Figure 1 shows a device for distance measurement according to the invention according to a first exemplary embodiment in a schematic meridional section;

[0049] Figure 2 is a plan view of a pinhole diaphragm with a two-dimensional arrangement of apertures for the device shown in Figure 1;

[0050] Figure 3 is a plan view of a pinhole diaphragm with a one-dimensional arrangement of apertures;

[0051] Figure 4 shows a part of a gear wheel with a measuring beam at two different measuring beam positions in a side view;

[0052] Figure 5 is a circuit diagram of an analog circuit arrangement of a light-sensitive cell which is part of the spectrograph contained in the device, according to a first embodiment in which charge carriers are distributed serially between two capacitors;

[0053] Figure 6 is a flow chart for explaining an embodiment of a method according to the invention carried out by the device according to Figure 1;

[0054] Figure 7 is a circuit diagram of an analog circuit arrangement according to a second embodiment in which the charge carriers are distributed in parallel between two capacitors; and

[0055] Figure 8 shows a distance measuring device according to the invention according to a third embodiment, in which only one measuring point is provided and the measuring light is guided via optical fibers. DESCRIPTION OF PREFERRED EMBODIMENTS

[0056] 1. Structure of the measuring device

[0057] Figure 1 shows a device for distance measurement according to the invention and designated overall by 10 according to a first embodiment in a schematic meridional section.

[0058] The device 10 contains a light source 12 that generates broadband measurement light ML. The measurement light ML preferably has a continuous spectrum. However, the use of comb spectra or other discrete spectra is also generally possible, provided the wavelength peaks are sufficiently close together. In the illustrated embodiment, the light source 12 comprises a broadband light-emitting diode (LED) 14 and a converging lens 16 arranged behind it in the direction of light propagation.

[0059] The measuring light ML generated by the light source 12 is first collimated by the converging lens 16. A further converging lens 16' focuses the measuring light ML such that it just completely illuminates a pinhole 18. In the plan view of the pinhole 18 according to Figure 2, it can be seen that the pinhole contains a two-dimensional pattern of apertures 22. If a pinhole with a one-dimensional, i.e. linear, pattern of apertures 22 is used, as shown in plan view in Figure 3, the converging lens 16' can be designed as a cylindrical lens or as another anamophotic optical element that has a different refractive power along orthogonal directions. In this way, light losses can be minimized. Alternatively, a light source with an elongated exit surface can be used, which is imaged onto the pinhole and illuminates it accordingly.

[0060] When illuminated with the measuring light ML, the apertures 22 of the pinhole 18 represent the first apertures from which the measuring light ML emerges divergently. A collimator lens 32 is arranged in the light path behind the pinhole 18, the front focal plane of which coincides with the plane E1. The beams emerging from the apertures 22 are therefore collimated upon passing through the collimator lens 32.

[0061] The collimated beams strike a non-polarizing beam splitter cube 33, which has a beam splitter surface 34 inclined by 45° to the optical axis. A predetermined portion of the incident light, e.g., 50%, is reflected by the beam splitter surface and is lost for measurement. The remaining measurement light ML passes through the beam splitter cube 33 without deflection due to the collimated beam path and strikes a chromatically uncorrected objective lens 36, which has an object plane lying at infinity. Together with the imaging optics 24 and the collimator lens 32, the objective lens 36, due to its chromatic longitudinal aberration, images the diaphragm openings 22 arranged in the plane E1 into image planes B1 to B3, the axial positions of which are wavelength-dependent. In Figure 1, this is indicated by different dashed lines. Each wavelength is assigned exactly one image plane B1 to B3.If - as in the present embodiment - the spectrum generated by the light source 12 is continuous, a continuous sequence of image planes is created.

[0062] If the image of an aperture 22 lies in an image plane at the level of which an at least partially reflective surface 38 of a workpiece 40 is located, the light beam incident there is partially reflected back into itself and returns to the beam splitter surface 34 of the beam splitter cube 33 via the lens 36. For the configuration shown here, for example, this requirement is met for the aperture 22 located on the optical axis OA of the device 10 at a wavelength for which the image of the aperture 22 lies in the image plane B2. For the light beam 43 that has passed through a different aperture 22, this requirement is met for the object 40 assumed here at a different wavelength for which the image of the aperture 22 in question lies in the image plane B1. Of course, not only light of a single wavelength is reflected at the surface 38.However, there is only one wavelength at which the measuring light beam is reflected in such a way that a larger proportion of the reflected light can contribute to an image of the measuring spot on the plane E1 or a plane conjugated to it E2.

[0063] At the beam splitter surface 34, a predetermined portion of the reflected measurement light ML is reflected by 90° and focused by a converging lens 42 onto a spatial filter 44 arranged in a second plane E2. The spatial filter 44 is transparent to the measurement light ML at certain locations 45 (or reflective in the case of a reflective spatial filter), while blocking the measurement light ML at the remaining locations. The transparent locations 45 form second apertures through which the measurement light ML must pass in order to be detected.

[0064] In the simplest case, the spatial filter 44 is the same pinhole diaphragm shown in plan view in Figure 2, possibly reduced or enlarged by the image scale of the intervening optics. The entire optics in the light path between the pinhole diaphragm 18 and the spatial filter 44, i.e., the collimator lens 32, the objective lens 36, and the focusing lens 42, cause the pinhole diaphragm 18 to be imaged onto the spatial filter 44. The planes E1 and E2 are therefore optically conjugated. Measuring light ML, which has passed through a specific first aperture (one of the apertures 22 of the pinhole diaphragm 18), therefore passes through a second aperture (transparent location 45 in the spatial filter 44), which is assigned to the first aperture by optical conjugation. Intermediate images are created in the image planes B1 to B3 on the surface 38 of the workpiece 40.

[0065] The spatial filter 44 thus ensures that only measuring light beams that have been reflected at a specific wavelength from the surface 38 can be further evaluated. Measuring light ML that does not meet this requirement is blocked by the static spatial filter 44 according to the chromatic-confocal measuring principle.

[0066] A spectrograph 46 with multiple input channels is arranged in the light path behind the static spatial filter 44. In each input channel, reflected measurement light ML, which has passed through a translucent location of the static spatial filter 44 associated with the respective input channel, is spectrally analyzed. The spectrograph 46 typically contains a dispersive optical element, e.g., an optical grating or a prism 47, as well as a line-like arrangement 48 of light-sensitive cells 49 for each input channel, as schematically indicated in Figure 1 for some input channels.

[0067] During a measurement, a control and evaluation device 30 calculates distances from measurement points on the surface 38 to the measuring device 10 from wavelengths measured by the spectrograph 46, as is known per se for chromatic confocal measuring devices. The measurement points are images of the apertures 22. Each aperture 22 is assigned its own input channel, so that the distance measurement can be performed for all apertures simultaneously, provided the apertures are spaced far enough apart that no optical channel crosstalk occurs. If a higher lateral resolution is desired, switchable spatial filters can be used, as described in detail in the German patent application filed on July 13, 2022, with the file number 10 202 211 7536.7.

[0068] In an alternative embodiment, the device 10 is designed such that the light passes through completely separate optical systems on its path from the apertures 22 to the workpiece 40 and from the workpiece 40 to the light-transmitting locations 45 of the spatial filter 44, as disclosed, for example, in DE 10 2019 118 600 A1. 2. Possible measurement task

[0069] Figure 4 illustrates a typical measuring task for the device 10. It shows a side view of a section of a gear 50, whose dimensional accuracy and surface quality are to be measured using the device 10. With a linear arrangement of apertures 22, as shown in Figure 3, the alignment of the pinhole 18 is parallel to the axis of symmetry of the gear 50, i.e., perpendicular to the plane of the paper.

[0070] During the measurement, the gear 50 rotates around its axis of symmetry, as indicated by an arrow. As a result, the measuring beams 52 gradually sweep over the radial surface 38 of the gear 50.

[0071] If a measuring beam 52 strikes the surface 38 of the gear 50 perpendicularly, e.g., a root surface 53 or, as in the illustrated rotational position, the tip surface 54 of a tooth 55, a large amount of measuring light is reflected back into the objective lens 36 and is available for evaluation by the spectrograph 46. In other rotational positions, however, the measuring beams 52 strike the flank 56 of the teeth 55. In Figure 4, this configuration is indicated by a dashed measuring beam 52'; in reality, it is not the measuring beam 52' that moves, but the gear 50.

[0072] In the area of ​​the flank 56, the surface normal of the surface 38 forms such a large angle with the direction of the measuring beam 52' that only a small amount of the measuring light ML is reflected back into the objective 36 and can be evaluated in the spectrograph 46. Even if the gear 50 has the same surface quality throughout, including in particular roughness and material, the intensity of the measuring light ML entering the spectrograph 46 varies considerably during rotation of the gear 50 due to the different angles of inclination of the surface 38. This can lead to the light-sensitive cells 49 in the spectrograph 46 becoming saturated at high intensities or receiving so little measuring light ML that the measuring signal does not stand out clearly enough from the noise. In both cases, the light-sensitive cells 49 do not provide precise values ​​for the intensity of the incident measuring light ML.

[0073] 3. Structure of the light-sensitive cells - first embodiment

[0074] In order for the light-sensitive cells 49 to provide reliable values ​​despite the above-described strongly varying intensities of the incident measuring light ML, the cells 49 can be constructed according to a first embodiment in a manner as shown in Figure 5 using a circuit diagram for an analog circuit 58.

[0075] The cell 49 shown therein has a radiation-sensitive element, which may be, for example, a photodiode 60. The photodiode 60 is connected to a first capacitor 62 having a capacitance C1. When measuring light ML strikes the photodiode 60, charge carriers are generated there, which flow into the first capacitor 62 and are stored there. This causes the voltage V1 across the first capacitor 62 to rise.

[0076] If the voltage V1 exceeds a threshold voltage VS, which is determined by a gate voltage applied to a transistor 64, the transistor 64 opens, allowing the charge carriers generated in the photodiode 60 to flow into a second capacitor 66. The second capacitor 66 has a capacitance C2 that is significantly higher than the capacitance C1 of the first capacitor, e.g., C2 = 10 x C1.

[0077] If the intensity (or more precisely, the radiant energy integrated over the exposure time) of the measuring light ML is low, only a few charge carriers are released in the photodiode 60 and transferred to the first capacitor 62. The voltage V1 generated by the first capacitor 62 then remains below the threshold voltage VS, so that the transistor 64 is blocked and no charge carriers reach the second capacitor 66. The first voltage V1 is then read out via a drain circuit indicated at 68.

[0078] By closing a readout switch 70, the charge carriers flow from the first capacitor 62 into the second capacitor 66, generating a voltage V2 there that is lower than the voltage V1 due to the higher capacitance C2 of the second capacitor 66. This second voltage V2 is also read out via the drain circuit 68. The cell 49 is reset via reset switches 71, 72.

[0079] Further details on the structure of cell 49 can be found in US 2004 / 0041077 A1 mentioned above.

[0080] Two voltages V1, V2, obtained from the same amount of charge but with different amplifications, are thus transmitted to the evaluation unit 30. Preferably, an average value is calculated from the two voltages, which forms the output signal for cell 49 and is used to calculate the distance to surface 38. However, it is also possible to use only one of the two voltages, V1 or V2.

[0081] If, on the other hand, the intensity of the measuring light is high, so many charge carriers are released in the photodiode 60 that the first capacitor 62 quickly goes into saturation. This is indicated by the first voltage V1 exceeding the threshold voltage VS. This causes the transistor 64 to open, so that all further charge carriers flow into the second capacitor 66 with the larger capacitance C2. After the exposure time has elapsed, the first voltage V1 is read out via the drain circuit 68. By closing the readout switch 70, the charge carriers then flow from the first capacitor 62 into the second capacitor 66, where they additionally contribute to the second voltage V2, which is also read out via the drain circuit 68. Since the second capacitor 66 has a larger capacitance C2, less amplification takes place there, i.e. each additional charge carrier increases the voltage V2 less than is the case with the first capacitor 62.

[0082] Both voltages V1 and V2 are supplied to the evaluation unit 30. When the first voltage V1 is approximately equal to VS, the evaluation unit 30 detects that the first capacitor 62 has saturated, which is why it discards the first voltage V1 and uses only the second voltage V2 to determine the spectral intensity maximum, from which the distance is subsequently derived.

[0083] Alternatively, the charge transfer from the first capacitor 62 to the second capacitor 66 can be omitted. To do this, after the first voltage V1 has been read, the readout switch 70 remains open, while the reset switch 71 is closed, thus short-circuiting the first capacitor 62. Then, the two switches 70 and 71 are flipped, allowing the second voltage V2 to be read. This voltage is somewhat lower in this alternative, since the charge carriers from the first capacitor 62 do not contribute to it.

[0084] In order to obtain even more meaningful measurement results, the evaluation unit 30 can also take the voltages V1, V2 into account in the following way:

[0085] ■ If the first voltage V1 generated by the first capacitor 62 is less than a first threshold value S1, only the first voltage V1 is used in the calculation. This corresponds to the case already explained above in which only a small amount of measurement light ML reaches the light-sensitive cell 49, so that a high gain is desirable.

[0086] ■ If the first voltage V1 lies between the first threshold value S1 and a second threshold value that is greater than the first threshold value, an average value M = M(V1, V2) from the first voltage and the second voltage is used to calculate the distance. The average value can be arithmetic, ie M = (V1 + V2) / 2, but can also be weighted. The weighting can be set such that the contribution of the first voltage V1 is smaller the closer the first voltage V1 is to the second threshold value S1, which corresponds to saturation. The contribution of the second voltage V2, which was generated by the second capacitor 66 with the lower gain, then becomes correspondingly larger. This corresponds to the case where a lot of measuring light ML reaches the light-sensitive cell 49, but the first capacitor is not yet saturated.

[0087] ■ If the first voltage V1 is above the second threshold S2, only the second voltage V2 is used to calculate the distance. This corresponds to the case where the intensity of the incident measurement light ML is so high that the first capacitor 62 saturates and many charge carriers must be absorbed by the second capacitor 66, so that the lower gain of the second capacitor is sufficient to obtain a sufficiently strong measurement signal.

[0088] According to this decision tree, if only very few charge carriers were generated, only the first voltage V1 obtained through high amplification is used to calculate the spectral intensity maximum. If more charge carriers are available without saturation occurring, both voltages V1 and V2 are included in the calculation. If the first capacitor 62 saturates, only the second voltage V2 is used in the calculation, as already mentioned above.

[0089] 4. Procedure

[0090] Figure 6 illustrates the process sequence described above using a flow chart.

[0091] In a first step S1, broadband measurement light ML is generated. In a second step S2, the measurement light ML passes through a first aperture, e.g., a diaphragm opening 22 in the pinhole 18.

[0092] In a third step S3, the first aperture is imaged onto the workpiece, e.g. the gear 50.

[0093] In a fourth step S4, the measuring light ML reflected from the workpiece passes through a second aperture, e.g. an opening 45 in the spatial filter 44.

[0094] In a fifth step S5, the reflected measuring light ML is spectrally analyzed after it has passed through the second aperture.

[0095] As part of the spectral analysis, in a first sub-step S5a, charge carriers generated by the photodiode 60 are transferred to a first capacitance C1 (capacitor 62) and to a second capacitance C2 (capacitor 66) with C2 > C1.

[0096] In a second sub-step S5b, a first voltage V1 with a high gain is derived from charge carriers in the first capacitance (capacitor 62).

[0097] In a third sub-step S5c, a second voltage V2 with a low gain is derived from charge carriers in the second capacitance (capacitor 66).

[0098] In a fourth sub-step S5d, the first voltage V1 is compared with two threshold values ​​S1, S2 where S1 < S2. If the first voltage V1 is less than the first threshold value S1, the first voltage V1 is used in the fifth sub-step S5e to calculate the distance value in step S6. If the first voltage V1 lies between the two threshold values ​​S1, S2, an average value M = M(V1, V2) is used in the fifth sub-step S5e to calculate the distance value in step S6, whereby the average value M can be arithmetic or weighted. If the first voltage V1 is greater than the second threshold value S2, the second voltage V2 is used in the fifth sub-step S5e to calculate the spectral intensity maximum and thus the distance value in step S6.

[0099] 5. Second embodiment

[0100] In the embodiment shown in Figure 7 for a light-sensitive cell 49', a first portion of the charge carriers in the analog circuit 58' is transferred to the first capacitor 62 and, in parallel, a second portion of the charge carriers is transferred to the second capacitor 66. The transistor 64, which in the first embodiment shown in Figure 5 transfers the charge carriers to the second capacitor 66 when saturation is reached in the first capacitor 62, is thus omitted in the analog circuit 58'.

[0101] If the evaluation unit 30 determines that the first voltage V1 is below a threshold value VS, it assumes that saturation has not occurred, so that the first voltage V1, the second voltage V2, or a (possibly weighted) average of the two voltages V1, V2 can be used to calculate the distance. If, however, the first voltage V1 is above the threshold value VS, this indicates saturation of the first capacitor 62, which is why only the second voltage V2 is used in the calculation.

[0102] By distributing the charge carriers in parallel between both capacitors 62, 66, fewer charge carriers accumulate in the first capacitor 62 than in the first embodiment. This is particularly disadvantageous when only a very small amount of measurement light ML hits the photodiode 60, because then correspondingly fewer charge carriers can contribute to the higher amplified voltage V1.

[0103] 6. Third embodiment

[0104] Figure 8 shows a device according to the invention, designated 10', in a third exemplary embodiment in a schematic meridional section. In this exemplary embodiment, measurements are not taken simultaneously at multiple measuring points, but rather only at one measuring point located on the optical axis OA of the objective 36. Because of the resulting very small object field, a simpler-designed objective 36 and a simpler-designed spectrograph 46 with only one channel can be used.

[0105] In addition, the beam is guided between the light source 12, the objective 36, and the spectrograph 46 via optical fibers 80, 82, as is known per se in the prior art, cf., for example, DE 10 2019 122 866 A1. The end of the optical fiber 80 forms a confocal aperture 22', which functionally corresponds to one of the diaphragm openings 22 of the first exemplary embodiment (cf. Figure 1). In this way, it is possible to design the objective 36 as a compact measuring head 84, which is connected to the other parts of the device 10' only via the optical fiber 80. These other parts are housed together in a housing 86, on which various operating elements and interfaces can be formed. The measuring head 84 is small and movable, so that it can be arranged (possibly movable) even in spatially confined conditions.

Claims

PATENT CLAIMS 1. Chromatic-confocal measuring device for measuring the distance to an object (40; 50), the measuring device (10) comprising: a light source (12) for generating broadband measuring light (ML), a first aperture (22) for the passage of the measuring light (ML), an imaging optics (32, 36) which projects the first aperture (22) onto the object (40; 50), wherein the imaging optics (32, 36) are chromatically uncorrected, so that images of the first aperture (22) are created at different distances from the imaging optics (32, 36) which depend on the wavelength, a spectrograph (46) which spectrally analyses measuring light (ML) which has been reflected by the object (40; 50) and has passed through a second aperture (45), wherein the second aperture (45) coincides with the first aperture (22) or is optically conjugated to the first aperture (22), and wherein the spectrograph (46) has a plurality of light-sensitive cells (49; 49') in which charge carriers are released upon impingement of measuring light (ML), which charge carriers are converted into electrical voltages, and an evaluation unit (30) which is designed to calculate the Distance to the object (40;50), characterized in that the spectrograph (46) is further designed to convert charge carriers released in the light-sensitive cells (49; 49') into electrical voltages (V1, V2) with two different amplifications, so that with each exposure one; light-sensitive cell (49; 49') two voltages (V1, V2) are output, and that the evaluation unit (30) is designed to calculate a distance value for each exposure cycle, taking into account the two voltages (V1, V2) respectively generated by the light-sensitive cells (49; 49').

2. Device according to claim 1, characterized in that the evaluation unit (30) is designed to calculate at least some distance values using an average value from the two voltages (V1, V2).

3. Device according to claim 1 or 2, characterized in that each light-sensitive cell (49; 49') is assigned an analog circuit which is designed to transfer the charge carriers to a first capacitor (62) and to a second capacitor (66), to derive a first voltage (V1) with a first gain from the charge carriers in the first capacitor (62) and to derive a second voltage (V2) with a second gain which is smaller than the first gain from the charge carriers in the second capacitor (66).

4. Device according to claim 3, characterized in that the first capacitance (62) is smaller than the second capacitance (66).

5. Device according to claim 4, characterized in that the analog circuit (58) is designed to transfer charge carriers which can no longer be absorbed by the first capacitor (62) completely to the second capacitor (66).

6. Device according to claim 5, characterized in that the analog circuit (58) is designed to transfer the charge carriers stored in the first capacitor (62) to the second capacitor (62) after the first voltage (V1) has been read out and before the second voltage (V2) has been read out.

7. Device according to claim 5 or 6, characterized in that the analog circuit has a switching element (64) which is designed to no longer conduct charge carriers to the first capacitance (62) but to the second capacitance (66) when the first voltage (V1) exceeds a predetermined threshold voltage (VS).

8. Device according to one of claims 5 to 7 with reference back to claim 2, characterized in that the evaluation device (30) is set up to use only the first voltage (V1) when calculating the distance a) if the first voltage (V1) is less than a first threshold value (S 1 ), b) to use an average value (M) from the first voltage (V1) and the second voltage (V2) if the first voltage (V1) is between the first threshold value (S1) and a second threshold value (S2) which is greater than the first threshold value (S 1 ), and c) to use only the second voltage (V2) if the first voltage (V1) is above the second threshold value (V2).

9. Device according to claim 4, characterized in that the analog circuit (58') is designed to transfer a first part of the charge carriers to the first capacitor (62) and, in parallel thereto, to transfer a second part of the charge carriers to the second capacitor (66).

10. Method for chromatic-confocal measurement of the distance to an object (40; 50) with the following steps: a light source (12) generates broadband measuring light (ML); the measuring light (ML) passes through a first aperture (22); an imaging optics (32, 36) images the first aperture (22) onto the object (40; 50), wherein the imaging optics (32, 36) are chromatically uncorrected, so that images of the first aperture (22) are created at different distances from the imaging optics (32, 36) which depend on the wavelength; a spectrograph (46) spectrally analyses measuring light (ML) which has been reflected by the object (40; 50) and has passed through a second aperture (45), the second aperture (45) coinciding with the first aperture (22) or being optically conjugated to the first aperture (22), and the spectrograph (46) having a plurality of light-sensitive cells (49; 49') in which charge carriers are released upon incidence of measuring light (ML), which charge carriers are converted into electrical voltages;an evaluation unit (30) calculates the distance to the object (40; 50) from the electrical voltages generated by the spectrograph (46); characterized in that charge carriers released in the light-sensitive cells (49; 49') of the spectrograph (46) are converted into electrical voltages (V1, V2) with two different amplifications, so that two voltages (V1, V2) are output for each exposure of a light-sensitive cell (49; 49'), and in that the evaluation unit calculates a distance value for each exposure cycle, taking into account the two voltages (V1, V2) respectively generated by the light-sensitive cells (49; 49').

11. The method according to claim 10, characterized in that the evaluation unit calculates at least some distance values on the basis of an averaging of the two voltages (VI, V2).

12. Method according to claim 10 or 11, characterized in that each light-sensitive cell (49; 49') is assigned an analog circuit (58; 58') which transfers the charge carriers to a first capacitor (62) and to a second capacitor (66), derives a first voltage (V1) with a first gain from the charge carriers in the first capacitor (62) and derives a second voltage (V2) with a second gain, which is smaller than the first gain, from the charge carriers in the second capacitor (66).

13. The method according to claim 12, characterized in that the first capacitance (62) is smaller than the second capacitance (66).

14. The method according to claim 13, characterized in that the analog circuit (58) transfers charge carriers which can no longer be absorbed by the first capacitor (62) completely to the second capacitor (66).

15. The method according to claim 14, characterized in that the analog circuit (58) transfers the charge carriers stored in the first capacitor (62) to the second capacitor (62) after reading out the first voltage (VI ) and before reading out the second voltage (V2).

Images