METHOD OF PRODUCE OR PROCESSING A DENTURE

DE502020009782D1Active Publication Date: 2024-12-05GLEASON PFAUTER MASCHFAB
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
DE502020009782
Authority / Receiving Office
DE · DE
Patent Type
Patents
Current Assignee / Owner
Priority Date
2019-04-15
Filing Date
2020-03-27
Publication Date
2024-12-05
Estimated Expiration
2040-03-27

AI Technical Summary

Technical Problem

Existing methods for machining workpieces with rotating tools face challenges in maintaining tool life and reliability, leading to potential errors in workpiece production due to tool wear, with current monitoring systems often detecting issues too late to prevent batch-wide defects.

Method used

A monitoring system that uses Fourier transformation of time signals from sensors to detect local tool wear, generating warning signals when thresholds are exceeded, allowing for timely tool changes and preventing further processing with worn tools, thereby reducing production errors and extending tool life.

Benefits of technology

The system effectively reduces the probability of workpiece errors by detecting tool wear early, allowing for immediate action to prevent machining errors and optimizing tool life, while minimizing follow-up efforts by ensuring tools are replaced before tolerance limits are breached.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader
Need to check novelty before this filing date? Find Prior Art

Description

[0001] The invention relates to a method for producing or machining a gear tooth on a workpiece, in which the workpiece, driven to rotate about its axis of rotation, is brought into rolling machining engagement with a tool tooth rotating about an axis of rotation, in particular at a non-zero cross angle to the workpiece axis of rotation, and to a gear cutting machine designed for this purpose. Such methods are well known to those skilled in the art, for example as the gear skiving described in EP 2 537 615 A1, which document is included herein by reference with regard to the terminology and machining engagement of gear skiving, wherein gear skiving represents a preferred gear machining operation of the invention.

[0002] When machining particularly large batches of workpieces, the tools with their tooth profiles are replaced at regular intervals because progressive wear causes the machining quality to fall below the desired tolerances. Since tool changes reduce the average machining time per workpiece, efforts are made to maximize tool life, for example, by using wear-increasing coatings. Furthermore, attempts are made to postpone tool replacement as long as possible, i.e., to avoid changing tools too early. This is achieved, for example, by regularly taking samples of the produced workpiece teeth, checking their tooth flanks for profile deviations, and changing the tool only when the deviations of the workpiece flanks from their intended ideal form exceed a suitably defined tolerance threshold.

[0003] Occasionally, a tool failure, such as a broken tool tooth, can occur. Such a singular event is detected by the monitoring mechanisms standard in modern CNC machines, and an emergency stop is initiated to prevent further damage to the machine.

[0004] WO 2018 / 073047 A1 teaches the improvement of the service life of peeling wheels by rake surface phases on the cutting teeth with varying phase angle and discloses a method according to the preamble of claim 1.

[0005] EP 1 195 668 A1 teaches process monitoring systems for process-accompanying wear detection on hob cutters using several normalized and weighted monitoring parameters.

[0006] Despite all such safety measures, it occasionally happens that a workpiece defect is only discovered later during the use of the geared workpieces, which is attributable to workpiece gear data that is not within the desired tolerances. Even if such a deviation is discovered before later use, for example during a final inspection, uncertainties arise as to whether this is due to an isolated workpiece or whether preceding or subsequent workpieces in the batch might also be affected.

[0007] In view of these difficulties, the invention is based on the objective of improving a method of the type mentioned at the outset, in particular with regard to a good combination of tool life on the one hand and reliability on the other.

[0008] This problem is solved by the invention, from a process engineering perspective, through a further development of the method with the features of claim 1. Due to the monitoring according to the invention, the probability of a workpiece defect or defective workpiece production is reduced, since information about the locally greatest tool wear is available during the machining operation. Accordingly, suitable measures can be taken early on, such as interrupting the machining after the workpiece has just been machined, during the machining process, or even immediately stopping the process. Tool life is increased by maximizing the utilization of the currently used tool, thereby reducing subsequent costs should the tolerance limit for the tools be approached too closely or exceeded.

[0009] In a preferred embodiment, the monitoring incorporates periodicities on time scales larger than the product of the number of teeth and the rotational speed of the workpiece. This allows for the particularly suitable determination of the relative influence of the most worn tool tooth.

[0010] A particularly preferred method is to subject a time signal generated by sensory acquisition to a transformation into its complementary space, especially a Fourier transform. This simplifies the evaluation and allows for the introduction of suitable and desired termination criteria.

[0011] In this context, it is preferable that the monitoring includes a monitoring window superimposed on the transformed signal. This allows for targeted testing of the transformed signal's behavior in a region where the influence of the most worn tool tooth is quantitatively most pronounced.

[0012] As explained above, after potentially variable termination criteria, the workpiece batch should no longer be processed unchanged. Accordingly, the monitoring system is designed to generate a warning signal if the irregularity exceeds a predefined threshold. This warning signal can be selected from a class of signals representing different warning levels, to which the operator can respond, but also includes signals to which the machine reacts automatically. For example, at a pre-warning level, a machine operator can be notified to review the current quantitative measurement of the irregularity. If only a few workpieces are missing to complete a batch, the operator can decide to complete the batch and, for example, initiate a tool change before starting the next batch.If necessary, such a decision could also be made by the machine itself using suitable algorithms. In any case, it is preferred that a warning signal of such a significant level be included that no new workpiece is machined with the currently used tool.

[0013] Preferably, a criterion is used for issuing the warning signal and / or selecting the warning signal from the available warning levels; more precisely, the criterion for generating the warning signal is the exceedance of the irregularity beyond a predetermined level due to the transformed signal exceeding a predefined threshold related to the monitoring window. This threshold could, for example, be a threshold for the transformed signal integrated within the monitoring window. However, a particularly simple yet reliable approach uses an amplitude threshold. Naturally, this threshold is first exceeded by the fastest-growing peak in the monitoring window, thus also taking advantage of the influence of the most worn tool tooth.

[0014] Regarding the detection of the machining operation, an acoustic signal is preferably recorded using sensors. This could, for example, involve a recording made via a microphone. However, it is preferable to utilize structure-borne sound generated by the machining operation and employ sensors such as vibration sensors / accelerometers for signal acquisition, which are, for example, mounted on the tool head supporting the tool. Usually, one sensor is sufficient, but for verification or redundancy, several sensors can be used. An accelerometer used, for example, operates on the basic principle that a seismic mass is spring-suspended within the accelerometer housing. When the sensor is accelerated, the mass's position relative to the housing changes due to inertia, and this change is measured via the spring deflection.Sensors with the highest possible natural frequency are preferred, for which piezoelectric sensors are considered suitable. However, the physical unit of the amplitude measured by the sensor, for example, an amplitude specified in mg (g = 9.81 m / s²), is not the only possible implementation; the velocity of the vibration, in units such as mm / s, could also be used. Suitable sensors are readily available to those skilled in the art. Furthermore, a power value of the process could also be used as an output signal, such as the currents from the rotary drive of the tool spindle or workpiece spindle.

[0015] The evaluation of the transformed signal preferably uses time ranges of the sensor-acquired time signal that are recorded in at least one operation where the contact area considered over the gear development is larger than in a majority of the other operations. This allows for the most effective detection of the most advanced wear.

[0016] It is also provided that the relevant sensor data is acquired in a single pass, at least within the range between the point at which the tool teeth fully engage due to the feed rate and just before they begin to disengage. This increases the stability and reliability of the transformed signal. Conversely, it is also preferred that the changes in the machining engagement associated with the axial feed rate as the tool teeth engage are awaited before sensor measurements are incorporated into the transformed signal. For example, it is provided that the incorporated time values ​​are only recorded when 20%, 35%, or 50% of the workpiece's tooth width has been reached.

[0017] The time interval used to determine the transformed signal is preferably a recording length of at least 0.1 seconds, but preferably at least 0.3 seconds or at least 0.5 seconds. Due to clipping, the transformed signal is inherently inaccurate at very low frequencies and, in any case, cannot handle full load with respect to the amplitudes at those frequencies. However, this does not pose a major obstacle for the invention, since harmonics are also captured in the transformed signal, and an amplitude threshold can be referenced to the harmonic with the largest amplitude.

[0018] In a preferred embodiment, the number of teeth on the workpiece and the tool are commensurate or at least quasi-commensurate. Commensurable means that a greater common divisor of the number of teeth on the tool and the workpiece exists that is not equal to one, in particular an odd greater common divisor. Quasi-commensurable means that this condition holds at least for hypothetical numbers of teeth that differ in total by a maximum of n from the actual number of teeth, where n ≤ 1 / 3 of this greater common divisor. For example, the specific embodiment below with numbers of teeth 85 and 51 is commensurate with a greater common divisor of 17, and numbers of teeth 53 and 83 could be reduced to this by changing them by 2 respectively (with 2 + 2 = 4 < 17 / 3), so that quasi-commensurability exists for this pairing. Preferably, the greater common divisor is 5 or greater, in particular 7 or greater.

[0019] In this context, a difference in the number of teeth of the workpiece and tool of at least 3, in particular at least 5, and in particular at least 7, is also preferred. With such conditions, lower beat effects with other machine frequencies and thus a clearer signal in the complementary space are achieved, which facilitates the determination of more precise threshold values.

[0020] Furthermore, a portion of the transformed signal that depends on the tool's rotational speed within the workpiece is preferably included in the monitoring. This dependency can, for example, consist of the monitoring window encompassing a value calculated from the workpiece rotational speed multiplied by the rotational speed factor, and / or at least one of the nearest (e.g., up to the fifth or sixth) harmonics of this value.

[0021] In terms of device technology, a gear cutting machine for producing or machining a gear on a workpiece, with a rotatably driven bearing for the workpiece and a rotatably driven bearing for a geared tool and a control device for controlling a machining operation taking place in rolling coupling, is also protected, which has the further features of claim 12.

[0022] The advantages of the gear cutting machine according to the invention result from the advantages of the method according to the invention stated above. The method is also protected in the form of a monitoring program that can be stored on a data carrier, as per claim 15.

[0023] Further features, details and advantages of the invention will become apparent from the following description with reference to the accompanying figures, of which Fig. 1 A cross-sectional view of a peeling wheel and internal gearing shows that Fig. 2a bis 2c The wear conditions of a tool tooth show, Fig. 3 a representation of a transformed detection signal suitable for Fig. 2a shows, Fig. 4 a representation of a transformed detection signal suitable for Fig. 2b shows, Fig. 5 a representation of a transformed detection signal suitable for Fig. 2c shows, as well as Fig. 6 shows a roller peeling machine.

[0024] The invention is illustrated below by means of a specific example of gear skiving of a workpiece 3 with internal teeth 2 using a skiving wheel 4 with tool teeth 5. As shown from Fig. 1 As can be seen, the workpiece rotation axis C and the tool rotation axis C2 are at an axis intersection angle Σ. In the chosen representation, machining takes place at the axis intersection point and the axis distance axis runs orthogonally to the paper plane; however, off-center machining operations are also conceivable, as is the machining of externally toothed workpieces.

[0025] In this specific embodiment, the workpiece rotates at 353 revolutions per minute and the tool at 589 revolutions per minute. The workpiece has 85 teeth (z), while the tool has 51 teeth (z2). The teeth thus have a greatest common divisor of 17, and the tool passes over the workpiece five times. This means that after five rotations of the tool, the same tool tooth engages with the same workpiece tooth gap again. However, other commensurate or quasi-commensurate tooth ratios could also be used.

[0026] With the selected tool, all tooth gaps of the workpiece are machined per workpiece revolution without a step factor, from which an engagement frequency can be calculated from the product of the number of teeth of the workpiece and the rotational speed of the workpiece, which is 500 Hz for the specific example used.

[0027] In this embodiment, a sensor 64 mounted on the tool head of the machine records a time signal of the vibrations occurring during machining. However, the sensor signal could also be acquired by other sensors, as explained earlier. Recording can be continuous, but in any case, it can be performed at intervals that are used for further evaluation of the signal. In this embodiment of gear skiving with, for example, eight passes of increasing feed and deeper cutting depth, the fifth cutting pass is selected, in which the contact area between the tool teeth 5 and the workpiece teeth 2 is comparatively large. Starting with the machining process at the midpoint of the workpiece tooth width in this embodiment, a signal recorded for one second is transformed using a Fourier transform. As in this embodiment, repeat measurements can also be performed, e.g.,two to four further measurements and transformations, but preferably only values ​​are used as long as the axial feed has not progressed so far that there is no longer a full machining engagement.

[0028] In this embodiment, an averaged representation of the sensor signal in the complementary space is formed from the three transformed recordings by averaging the individual transformed signals.

[0029] Such a signal is in Fig. 3 shown that the one with the new cutting wheel ( Fig. 2a The processing performed is reflected in the data. A peak can be seen at the aforementioned intervention frequency of 500 Hz.

[0030] During the further processing of a batch of workpieces, the tool teeth wear down, but usually not uniformly; rather, one tool tooth 51 will typically be the first to show signs of wear 52 and where the wear progresses most rapidly. This progression is in Fig. 2 depicted for such a tooth, in Fig. 2b with a wear in the range of 0.2 mm, in Fig. 2c with a wear of already 2 mm, which leads to machining errors in the engagement that are outside the tolerances for the manufactured workpieces.

[0031] By using the Fig. 3 By continuing the described measurements and evaluations of the measurement signal, scrap of workpieces caused by tool wear can be avoided, since the machining process of a batch of workpieces is interrupted when a predefined threshold in the measurement signal below the intervention frequency is exceeded. This threshold is defined in Fig. 5 This threshold S is shown and labeled S. This threshold S is not yet reached in comparison with a tool tooth whose wear, although relatively the most advanced, is still within the usable range. In contrast, peaks form, which in this embodiment primarily occur in the range of 120 (and also almost 150) Hz, and increase significantly with greater wear of this tool tooth, possibly even exceeding the peak at the engagement frequency in amplitude. This can be understood by the fact that the worn tool tooth, during the given rolling motion, alters the tooth gap shape (in the sequence 1, 52, 18, 69, 35, and 1 relative to it) of the corresponding tooth gaps in the workpiece, whose engagement during workpiece rotation also occurs with other tool teeth, but in the sequence 1, 18, 35, 52, and 69, and thus at a fundamental frequency of approximately 120 Hz.29.4 Hz, which can be identified as the interval between the peaks lying within the monitoring window, so that in this specific example, the relevant monitored peaks are those of the 4th and 5th harmonics. Therefore, in this specific embodiment, a monitoring window could be set in the range between 80 and 160 Hz, whereby different individual values ​​may result depending on the recording time and other tooth ratios. To avoid excessive individualization, the monitoring window could be chosen to be larger, for example, in the range of 30 to 350 Hz, or a first percentage of the intervention frequency (e.g., 5%, 10%, or 15%) to a second percentage (e.g., 80%, 65%, or 50%) of the intervention frequency.

[0032] It goes without saying that the threshold height can be set to a meaningful value by pre-determination. This only requires test runs with tools where a tooth is worn down around the wear limit and the creation of the... Fig. 3 bis Fig. 5 The Fourier-transformed sensor signals shown and explained above. This allows for the establishment of a very precise criterion, particularly for larger batches of workpieces, according to which a tool change can be triggered in a timely manner, but not prematurely, and which also responds when average tool wear would still be considered sufficient for further machining.

[0033] A corresponding monitoring control could be integrated into a (overall) control system 99 of the in Fig. 6 The gear skiving machine shown is integrated, which, as is otherwise quite common, has a machine bed 80 with a workpiece table 70 around which a workpiece can be rotated about the axis of rotation C using a clamping device (not shown), and on which CNC-controlled machine axes are provided on the tool side to perform the necessary movements for and during machining. These include a radial infeed axis X, an axial feed axis Z, a swivel axis A for adjusting the axis cross angle between the tool axis of rotation C2 and the workpiece axis of rotation C, and, in this embodiment, a tangential axis Y downstream of the swivel axis A for, e.g., off-center machining settings. Accordingly, in Fig. 6 The figure shows a radial slide 61, which carries a vertical slide 62 on which a slide 63, movable linearly in the plane orthogonal to the radial direction X, is arranged with the tool head for the tool. The sensor, used for example to detect the state of the machining operation, is in Fig. 6 The sensor is labelled with reference number 64. The time signal captured by the sensor could also be used for other process monitoring purposes.

[0034] The invention is not limited to the specific embodiment shown.

Claims

1. A method of producing or machining a toothing (2) on a workpiece (3), wherein the workpiece, which is rotationally driven around its axis of rotation (C), is brought into rolling machining engagement with a tool toothing (5) rotating around an axis of rotation (C2) which in particular is at a non-null crossed-axes angle (Σ) to the axis of rotation of the workpiece, and wherein the machining engagement is automatically monitored through a capture thereof using sensors which is already taking place during the machining engagement, characterised in that the monitoring includes monitoring of periodicities on time scales which are greater than the product of the number of teeth (Z) and the rotational speed (n) of the workpiece (3) for a recurring irregularity originating from tool wear (52) of at least one tool tooth (51), which is greater than that of other tool teeth.

2. The method according to claim 1, wherein a time signal generated during the sensor-based capture is subjected to a transformation in the complementary space, in particular to a Fourier transformation.

3. The method according to claim 2, wherein the monitoring includes a monitoring window superimposed on the transformed signal.

4. The method according to one of the preceding claims, wherein the monitoring generates a warning signal if the irregularity exceeds a predetermined level.

5. The method according to claims 3 and 4, wherein this exceedance is determined based on the detection of an exceedance of a predetermined threshold (S) relating to the monitoring window by the transformed signal.

6. The method according to one of the preceding claims, wherein an acoustic signal is recorded during the sensor-based capture.

7. The method according to one of the preceding claims, wherein the sensor-based capture takes place in at least one machining pass, in which the contact region, viewed across the course of the toothing, is larger than in a plurality of the other passes.

8. The method according to one of the preceding claims, wherein the sensor-based capture takes place in one pass in at least one region which lies between a full entry of the tool toothing, caused by the feed rate, and the point at which it starts to exit.

9. The method according to one of the preceding claims, wherein multiple recordings of in particular at least 0.1 seconds in length are made during the sensor-based capture, and the transforms of the plurality of recordings are averaged.

10. The method according to one of the preceding claims, wherein the numbers of teeth (Z, Z2) of the workpiece and tool are commensurable or at least quasi-commensurable.

11. The method according to one of claims 2 to 10, wherein a region of the signal which is dependent on the tool rolling over in the workpiece is included in the monitoring.

12. A gear cutting machine (100) for producing or machining a toothing on a workpiece, comprising a rotatably driven bearing arrangement for the workpiece and a rotatably driven bearing arrangement for a toothed tool and a control device (99) for controlling a machining engagement that takes place in a rolling coupling, said control device comprising an automatic monitoring device which comprises at least one sensor (64) and monitors the machining engagement through a capture thereof, using the at least one sensor, which is already taking place during the machining engagement, characterised in that the monitoring device is configured so as to include monitoring of periodicities on time scales that are greater than the product of the number of teeth (Z) and the rotational speed (n) of the workpiece (3), for a recurring irregularity originating from tool wear of at least one tool tooth which is greater than that of other tool teeth.

13. The gear cutting machine according to claim 12, wherein the monitoring device is further configured to carry out a method according to one of claims 2 to 11.

14. The gear cutting machine according to claim 12 or 13, wherein at least one structure-borne sensor is provided on the tool head forming the tool bearing arrangement.

15. A monitoring program which, when executed on a control device (99) of a gear cutting machine, executes a method according to one of claims 1 to 11 that is carried out by the machine, and which can be stored on a data medium.