Determining radial withdrawal and cutter edge geometry of a cutter on a rotating tool

Abstract

A method of determining radial wear and a method of determining sharpness of one or more cutters (21-n) arranged on and along a tool (1) having a longitudinal axis (z) and wherein the tool is rotating around said axis A plurality of digital silhouette images (I1-n) of silhouettes of at least one cutter edge along at least a portion of said longitudinal axis are captured, wherein adjacent images are separated by a separation angle (α), and a three-dimensional model of the cutter edge based on said plurality of digital images is generated. Cutter edge geometry is determined based on a plurality digital images.

Description

[0001] Determining radial withdrawal and cutter edge geometry of a cutter on a rotating tool

[0002] Technical field of the invention

[0003] The invention concerns image-based object recognition and measurement. More specifically, the invention concerns methods for determining radial withdrawal and shape of an object by use of optics, as specified in the preambles of claims 1 and 2.

[0004] Background of the invention

[0005] The high thermal and mechanical loads during milling operations result in wear of the tool's cutters (in some publications referred to as "teeth"), - more specifically in the cutter edge - producing a radial withdrawal of the cutter edge and a change in the cutter edge geometry. The change in the cutter edge geometry may comprise an increase of the cutter edge radius of curvature and / or a general variation of the cutter edge microtopography and leads to a reduction of tool efficiency and may compromise the surface quality of the workpiece. In addition, the changed cutter edge geometry results in extended contact on the tool's clearance face, which may induce larger cutting forces - leading to larger tool deflection and consequently increased geometrical errors on the workpiece. All these contributions serve to deteriorate the quality of the machined surface during the course of machining operation.

[0006] The prior art includes US 2021 / 0294297 Al, which describes a method for determining the state of wear of a tool. At least one optical image of a surface of the tool is recorded and image data are processed in order to detect a wear zone. A surface extent and / or spatial extent of the wear zone is determined. The state of wear of the tool is classified on the basis of the extent determined. A corresponding apparatus for determining the state of wear of a tool comprises a camera for recording optical images of the tool surface, an image processing module configured for processing image data in order to detect a wear zone, a computation module configured for determining a surface extent and / or spatial extent of the wear zone, and a classifier configured for classifying the state of wear of the tool on the basis of the extent determined.

[0007] So-called "on-machine tool presetting systems" are widely used on milling machines to characterize tool diameter and length for radius and length compensation. Laser beam interruption systems have been in use for several years and base the dimensional measurement on recording machine tool coordinates when the beam is interrupted, and radial wear is determined based on the intensity in one or more pixels illuminated by a laser. By performing periodic measurements, the variation of the effective tool geometry (e.g. radius) at the specific axial elevation can be monitored. However, such systems cannot attribute wear to the individual cutters' edges of the rotating tool and, as the measurement exploits the displacement along the machine tool axis, measurement accuracy is limited by the machine tool positioning accuracy. However, on-machine camera-based presetters have recently been introduced as an alternative to laser beam interruption systems.

[0008] The prior art also includes WO 2009 / 038465 Al, which describes an apparatus and a method for finding positional relationships and geometrical changes of a rotating part (e.g. a milling tool) attached to a machine when the part is in a rotating mode. The milling tool cutter edge is backlit (e.g. by a strobe light), and shadow images of the cutter edge are captured in different angular orientations.

[0009] The prior art documents CN 116252398 A, JP 2001-293642 A, CN 102581700 B, and TW M500649 U describe related technology.

[0010] Objectives of the invention are to determine (i) radial wear of a cutter and (ii) change of cutter edge geometry (i.e. cutter sharpness) along the cutting tool longitudinal axis - before and after use, and while the cutting tool is rotating, for example at a rotational speed used during a milling operation.

[0011] Summary of the invention

[0012] The invention is set forth and characterized in the main claim, while the dependent claims describe other characteristics of the invention.

[0013] It is thus provided a method of determining radial wear of one or more cutters arranged on and along a tool having a longitudinal axis and wherein the tool is rotating around a rotation axis or a camera is rotating around the tool, characterized by:

[0014] - capturing a plurality of digital silhouette images of silhouettes of at least one cutter edge along at least a portion of said longitudinal axis in an xy plane, wherein adjacent images are separated by a known separation angle; and

[0015] - by using the separation angle and digital silhouette images obtained in the xy plane, converting the two-dimensional digital silhouette images to a three-dimensional digital model of the cutter edge in the xy plane and along the tool z axis. It is also provided a method of determining sharpness of one or more cutters arranged on and along a tool having a longitudinal axis and wherein the tool is rotating around a rotation axis or a camera is rotating around the tool, characterized by:

[0016] - capturing a plurality of digital silhouette images of silhouettes of at least one cutter edge along at least a portion of said longitudinal axis in an xy plane, wherein adjacent images are separated by a known separation angle;

[0017] - based on the separation angle and digital silhouette images of silhouettes, converting the two-dimensional digital silhouette images to a three-dimensional digital model of the cutter edge in the xy plane and along the tool z axis;

[0018] - determining cutter edge geometry based on a plurality of digital silhouette images.

[0019] In one embodiment, the digital silhouette images are captured by a single camera. For example, the digital silhouette images are captured by an on-machine camera-based system and the silhouettes are produced by backlighting, for example by a light source arranged on the opposite side of the tool from the camera.

[0020] In one embodiment of both methods, the camera is stationary and the tool is rotating at its normal operating speed, and the separation angle is calculated or determined based on cutting tool rotational speed, and / or light source frequency, and / or data provided by the camera.

[0021] In one embodiment, the method of determining the sharpness is based on using at least three adjacent digital silhouette images.

[0022] In one embodiment of both methods, one or more cutters are arranged in a helical fashion along the tool longitudinal axis, and the tool is a cutting tool or a milling tool.

[0023] The invention thus involves two separate aspects: Determining (i) radial withdrawal and (ii) cutter edge geometry of a cutter on a rotating tool, based on captured digital silhouette images of the cutter edge. In one embodiment, the cutter may be a helical cutter, and the tool may be a cutting tool. The invention is based on captured digital silhouette images, where the silhouette is generated by backlight, as opposed to the method described in the above- mentioned US 2021 / 0294297 Al, which is based on optical images of an illuminated (i.e. front lit) tool surface. Brief description of the drawings

[0024] These and other characteristics of the invention will become clear from the following description of embodiments of the invention, given as non-restrictive examples, with reference to the attached schematic drawings, wherein:

[0025] Figure la is a perspective view of a cutting tool (in some publications referred to as a milling tool) having four helical cutters, and indicates a tool longitudinal axis (the tool z axis);

[0026] Figure lb is a cross-sectional drawing in an arbitrarily chosen xy plane perpendicular to the tool z axis, in the helical portion of the tool illustrated in figure la, also illustrating a camera and two lines representing two digital silhouette images for respective cutting tool positions or camera positions;

[0027] Figure 2a is a sketch of a virtual camera path around a cutting tool or a rotating cutting tool relative to a stationary camera, and illustrates a sequence of digital silhouette images; and figure 2b is an enlarged view of the circle "A" in figure 2a;

[0028] Figure 3 illustrates general principles of radial wear measurement using an on-machine camera-based system (CBS) on a cutting tool;

[0029] Figure 4 displays plots of measurements of a cutter edge along the cutting tool's longitudinal axis (tool z axis), for a new cutter and after three randomly selected passes (here: passes 40, 80, 120);

[0030] Figure 5 corresponds to figure lb and illustrates captured digital silhouette images of the respective cutter's edge in an xy plane;

[0031] Figure 6 is a processed representation of captured digital silhouette images of the cutter edge of each of the four cutters on an unused cutting tool;

[0032] Figure 7 is an enlarged view of the upper left-hand corner of figure 6 and illustrates the edge of one of the cutters in an xy plane, each line representing a digital silhouette image of the cutter edge as the tool is rotating;

[0033] Figure 8 corresponds to figure 6, and illustrates captured digital silhouette images of the cutter edge of each of the four cutters on a worn cutting tool; Figure 9 is a three-dimensional representation of four individual helical cutter edges and three projection planes along the tool z axis;

[0034] Figure 10 illustrates a cutter edge along the tool z axis for a new cutter and after three randomly selected passes (here: passes 40, 80, 120, cf. figure 4), based on the procedure described above with reference to figures 1 and 4, and cutter edge sharpness (i.e. the cutter edge geometry); and

[0035] Figure 11 illustrates a cutter edge sharpness along the tool z axis for a new cutter and after three randomly selected passes (here: passes 40, 80, 120).

[0036] Detailed description of embodiments of the invention

[0037] The following description may use terms such as "horizontal", "vertical", "lateral", "back and forth", "up and down", "upper", "lower", "inner", "outer", "forward", "rear", etc. These terms generally refer to the views and orientations as shown in the drawings and that are associated with a normal use of the invention. The terms are used for the reader's convenience only and shall not be limiting.

[0038] As mentioned above, on-machine camera-based presetters have recently been introduced as an alternative to laser beam interruption systems. They operate by recording multiple images of the rotating tool, detecting the tool profile and providing as standard output the tool length and radial position of each cutter's edge at a preset value of the axial elevation. The measurement cycle is typically very fast (< 4 seconds) and similarly to the laser systems, periodic assessment of tool dimensions can be exploited to monitor the wear condition of the tool. However, the assessment is limited to the preset evaluation position. Existing solutions for tool wear measurement either require interruption of the machining operation to dismantle the cutting tool and perform the inspection or provide limited information by evaluating wear in a single position or with insufficient resolution. In order to overcome the limitations of current wear monitoring systems, this invention enables the utilization of an on- machine camera-based system (CBS) for fast and automated high-resolution characterization of wear of rotating tools over the entire tool length and suitable for industrially viable on- machine tool wear monitoring in fine finishing machining of mould surfaces. One embodiment of an on-machine CBS is described in the above-mentioned prior art document WO 2009 / 038465 Al. Figure la illustrates a cutting tool 1 having four cutters 21-4 arranged in a helical fashion along the cutting tool's longitudinal axis (tool z axis). Although not illustrated, it should be understood that cutters can be extending in a non-helical fashion along the tool z axis, and the invention is not limited to the illustrated helical cutter arrangement. It should furthermore be understood that the cutting tool may have fewer or more cutters 2i-nthan the four illustrated. In one embodiment of the invention, the cutting tool 1 can be arranged in a milling machine (not shown) and be rotating (for example at a normal operating speed) when the digital silhouette images described below are generated, for example in a set-up as described in the above-mentioned WO 2009 / 038465 Al.

[0039] Figure lb illustrates a cross-section in an arbitrarily chosen xy plane perpendicular to the tool z axis in a portion of the tool of figure la in which cutters 21-4 are arranged. Figure lb also illustrates a camera 10 and two lines representing individual digital silhouette images k, h captured at two different relative rotational positions between the camera and the tool. In the context of this disclosure, a digital silhouette image shall be understood as a digital image of the silhouette of a cutter, the silhouette generated by a light source on the opposite side of the cutter from the camera.

[0040] Figures 2a and 2b illustrate a plurality of digital silhouette images k-n- In a conventional set-up, the camera 10 is stationary while the tool 1 is rotating about its rotation axis, but the invention is also applicable for a set-up in which the camera 10 is rotating around a tool, i.e. any relative rotation between tool and camera.

[0041] In one embodiment, the cutting tool 1 is installed in a milling machine (not illustrated) and rotating at a predetermined speed or at a speed controlled by a milling machine operator or control system. The invention does not require that tool rotation speed is controlled. As a nonlimiting example, the rotation speed may be between 800 and 60000 rpm (revolutions per minute). When the tool is rotating, the tool's rotation axis may not always be coaxial with the tool's longitudinal axis (z axis). A deviation between these axes can be due to tool imperfections or anomalies, and / or misalignment between the tool and the milling machine spindle. The rotating tool's effective milling radius may therefore differ from the tool's true radius, i.e. its radius in a static, non-rotating, state. The effective milling radius can therefore be defined as the true radius plus machine-induced deviations and / or milling tool mounting deviations. Although a backlight is not illustrated in figures lb and 2a, it should be understood that a light source (for example, but not limited to, a strobe light) is arranged on the opposite side of the tool 1 from the camera 10, and the camera 10 captures digital silhouette images (i.e. shadow images) of the cutter edge in different angular orientations, for example as described in the above-mentioned WO 2009 / 038465 Al, for example using an on-machine CBS. Adjacent digital silhouette images k, h are separated by an angle a, the magnitude of which is calculated or determined in a manner known in the art, for example based on cutting tool rotational speed, light source frequency, and / or data provided by the camera. Data provided by the camera may comprise captured images and camera clock speed. As a non-limiting example, the separation angle a is around 1 to 2 degrees.

[0042] Using a single camera, the system can provide a digital silhouette image of the tool profile which can be assessed at any position along the tool z axis within the field of view of the system and independently for each individual cutter 2i-n. Tool wear can be determined based on a single measurement as described below, or be characterized in terms of the withdrawal of the cutter's edge profile with respect to a reference edge profile - by comparing the digital tool profile measured at any time during the machining operation with a reference measurement recorded on an unused tool. At each desired level along the cutting tool z axis, and for each cutter 2nedge, the sensor (i.e. the camera 10) detects the point with the largest distance from the rotation axis, which - as mentioned above - may be deviating from the cutting tool's longitudinal axis (z axis). This point is effectively in contact with the machined surface and is determining the tool's effective milling radius. However, as wear changes the morphology of the cutter's edge, the largest radial wear is not necessarily at the detected position as shown in figure 3. While this approach does not allow full characterization of all wear features according to established standards (e.g. ISO 8688), estimation of local radial wear and edge profile waviness in the axial direction directly correlate with the surface generation phenomena described above and can therefore be used as a quantitative representation of the edge wear.

[0043] Aspect 1 - determining cutter radial wear (cutter edge radial withdrawal):

[0044] As illustrated in figure 4, the method according to a first aspect (Aspect 1) of the invention can be used to determine tool wear in terms of cutter edge radial withdrawal. Figure 4 displays plots of selected measurements of a cutter edge, along the tool's longitudinal axis (tool z axis). The line "New Tool" denotes a measurement of a new and unused tool before a milling operation starts, and individual lines represent respective consecutive measurements made after a milling operation. For example, the line "Pass 40" denotes a measurement after the fortieth milling operation. The plots in figure 4 show that the dimension in the xy plane (i.e. the cutter's radial dimension) decreases with increasing number of passes - due to wear. Tool wear can therefore be determined based on a single measurement, either before a milling operation commences or at any one of the subsequent passes. Tool wear can also be determined by comparing two or more series of digital silhouette images.

[0045] Cutter radial wear is thus determined by (a) arranging a cutting tool (1) having one or more cutters 2i-n in a machine such that the cutting tool is rotating around a rotation axis that may or may not be coaxial with the tool longitudinal axis (tool z axis), and (b) capturing a plurality of digital silhouette images of at least one cutter edge along at least a portion of the tool longitudinal axis, and generate a three-dimensional digital model of the cutter edge. By using the separation angle a (which is calculated or determined as described above) and the captured digital silhouette images, the two-dimensional images are converted to a three- dimensional digital model of the cutter edge, in the xy plane and along the tool z axis. Radial withdrawal of said cutter may thus be determined based on the three-dimensional model.

[0046] As an alternative, multiple radial withdrawal measurements may be compared to determine progressive wear. in cutter

[0047] While the invention according to Aspect 1 as described above determines the radial wear (i.e. radial withdrawal) of a cutter 2i-nalong its length on a rotating cutting tool 1, it does not provide any information as to the cutter's morphology; which is a decisive factor for its sharpness. As a second aspect (Aspect 2), it is therefore providing a method for determining the change in cutter edge's form, shape or structure (i.e. its geometry) along the cutting tool's longitudinal axis (tool z axis) while the tool is rotating, as described in the following.

[0048] Figure 5 corresponds to figure lb and illustrates digital silhouette images k - lnof the respective cutters' edge in an xy plane, captured by the camera 10 while the cutting tool is arranged and rotating as described above. The separation angle a between adjacent images is calculated or determined as described above. Figure 6 is a processed representation of the captured digital silhouette images of the edges of each of the four cutters. In this embodiment, the images are obtained in the same manner as described above, i.e. using a single camera 10 to capture digital silhouette images of the cutter edge in different angular orientations, and along the cutting tool's longitudinal axis. By using the separation angle a and the digital silhouette images, the two-dimensional images are converted to a three-dimensional digital model of the cutter edge, in the xy plane and along the tool z axis.

[0049] Figure 7 is an enlarged view of the upper left-hand corner of figure 6 and illustrates the edge of one of the cutters in an xy plane. Each line represents a digital silhouette image k - lnof the cutter edge as the tool is rotating, as described above, and the angle between adjacent lines in the figure corresponds to the separation angle a between adjacent digital silhouette images as described above. In the illustrated, non-limiting example, a circle fitting procedure has been used to determine the cutter edge geometry, by using three or more images. It should be understood that methods other than a circle-fitting procedure can be used to determine the geometry. Figure 8 corresponds to figure 6, and illustrates captured digital silhouette images of the cutter edge of each of the four cutters on a cutting tool after several more passes following the condition illustrated in figures 6 and 7. The reduced sharpness is thus apparent in that the radius of curvature for a used tool (n>, figure 8) is greater than the radius of curvature for a tool that has been subjected to fewer passes (ra, figure 6).

[0050] Figure 9 corresponds to figures 6 and 8 and is a three-dimensional representation illustrating three projection (xy) planes along the tool z axis. Varying radii of curvature ra, rb illustrate change in edge morphology along each cutter 2i - 24.

[0051] Although the above example involves determining a single radius, any set of parameterized geometries may be fitted to the lines representing digital silhouette images. The parameters and the deviation of the fit result from the geometrical shape can thus be determined based on the lines.

[0052] A similar procedure may be performed at subsequent stages of the milling procedure, whereby a change in cutter edge geometry over subsequent passes is an indication of how the cutter's sharpness changes (e.g. deteriorates), as shown in figure 8.

[0053] Cutter sharpness (i.e. the cutter edge geometry) is thus determined by (a) arranging a cutting tool 1 having one or more cutters 2i-nin a machine such that the cutting tool is rotating around a rotation axis that may or may not be coaxial with the tool longitudinal axis (tool z axis), and (b) capturing a plurality of digital silhouette images of at least one cutter edge along at least a portion of the tool longitudinal axis, and generate a three-dimensional digital model of the cutter edge.

[0054] Each digital silhouette image is represented by a corresponding line k .n, and adjacent digital silhouettes are separated by an angle a. The separation angle a is calculated or determined as described above. The curve fitting method to determine the cutter edge geometry is preferably based on at least three digital silhouette images k - h.

[0055] Generating a three-dimensional model:

[0056] The three-dimensional digital model comprises cutter edge points extracted from the digital silhouette images k-n using digital edge detection and known techniques for three-dimensional reconstruction from multiple two-dimensional images, for example triangulation. The cutter edge points are positioned in the three-dimensional model at locations and orientations which correspond to the surface of the milling tool being measured. Data processing detects effects caused by rotation speed inaccuracies and tool runout introduced during image recording, and the three-dimensional model is updated to remove these influences. The result is an accurate three-dimensional geometrical representation of each individual cutter edge as free from disturbances caused by environmental factors. The cutter edge sharpness is determined by digitally analysing the three-dimensional representation of each cutter and providing individual results for each cutter and aggregate results.

[0057] Correlation between cutter wear and cutter sharpness:

[0058] Figure 10 illustrates the wear of a cutter edge for a new tool and after three randomly selected passes, based on the procedure described above with reference to figures 1 and 3. In addition, the bar on the right-hand side in figure 10 indicates the cutter edge sharpness (i.e. the cutter edge geometry), determined as described in Aspect 2. The bar is presented in grayscale ranging between black (100 % sharpness, representing a new, unused, cutter edge) and light grey (0 % sharpness). A corresponding grayscale is seen in the cutter wear plots, i.e. the wear (i.e. radial withdrawal) of a new cutter is displayed in a black colour and the wear of a cutter which has been used 120 times is displayed in light grey. Figure 10 thus illustrates a strong correlation between cutter edge wear (i.e. radial withdrawal) and cutter edge sharpness (i.e. the cutter edge geometry). Figure 11 illustrates the cutter edge sharpness after a given pass and the sharpness of an unused cutter edge. The figure has been generated by inverting and scaling the data displayed in figure 10 and is a further illustration of the correlation between cutter edge wear and cutter edge sharpness. The correlation demonstrates that the method of Aspect 2 distinguishes random noise from actual cutter edge geometry.

[0059] General:

[0060] In the embodiments described above, various features and details are shown in combination. The fact that several features are described with respect to a particular example should not be construed as implying that those features by necessity have to be included together in all embodiments of the invention. Conversely, features that are described with reference to different embodiments should not be construed as mutually exclusive. As a person skilled in the art readily will understand, embodiments that incorporate any subset of features described herein and that are not expressly interdependent have been contemplated by the inventor and are part of the intended disclosure. However, explicit description of all such embodiments would not contribute to the understanding of the principles of the invention, and consequently some permutations of features have been omitted for the sake of simplicity or brevity.

[0061] Although the claimed invention has been described with reference to an on-machine camerabased system and a tool rotating at a rotational speed used during a milling operation, it should be understood that the invention comprises also other camera systems and operational environments.

Claims

Claims1. A method of determining radial wear of one or more cutters (2i.n) arranged on and along a tool (1) having a longitudinal axis (z) and wherein the tool is rotating around a rotation axis or a camera is rotating around the tool, characterized by:- capturing a plurality of digital silhouette images (li.n) of silhouettes of at least one cutter edge along at least a portion of said longitudinal axis in an xy plane, wherein adjacent images are separated by a known separation angle (a); and- by using the separation angle (a) and digital silhouette images (li.n) obtained in the xy plane, converting the two-dimensional digital silhouette images to a three-dimensional digital model of the cutter edge in the xy plane and along the tool z axis.

2. A method of determining sharpness of one or more cutters (2i.n) arranged on and along a tool (1) having a longitudinal axis (z) and wherein the tool is rotating around a rotation axis or a camera is rotating around the tool, characterized by:- capturing a plurality of digital silhouette images (li.n) of silhouettes of at least one cutter edge along at least a portion of said longitudinal axis in an xy plane, wherein adjacent images are separated by a known separation angle (a);- based on the separation angle and digital silhouette images (li.n) of silhouettes, converting the two-dimensional digital silhouette images to a three-dimensional digital model of the cutter edge in the xy plane and along the tool z axis;- determining cutter edge geometry based on a plurality of digital silhouette images.

3. The method of claim 1 or claim 2, wherein said digital silhouette images are captured by a single camera.

4. The method of claim 3, wherein said digital silhouette images are captured by an on- machine camera-based system (CBS).

5. The method of any one of claims 3-4, wherein the silhouettes are produced by backlighting.

6. The method of claim 5, wherein the silhouettes are produced by a light source arranged on the opposite side of the tool from the camera.

7. The method of any one of claims 1-6, wherein the camera is stationary and the tool is rotating at its normal operating speed.

8. The method of any one of claims 1-7, wherein the separation angle (a) is calculated or determined based on cutting tool rotational speed, and / or light source frequency, and / or data provided by the camera.

9. The method of any one of claims 1-8, wherein the method is based on using at least three adjacent digital silhouette images (I1-3).

10. The method of any one of claims 1-9, wherein one or more cutters (2i.n) are arranged in a helical fashion along the tool (1) longitudinal axis (z), and the tool is a cutting tool or a milling tool.

Images