Method for machining a workpiece by a continuous generating mill and corresponding gear generator

Abstract

A method for machining a workpiece with preformed teeth by means of continuous generating grinding with a worm grinding wheel (100) is proposed. The worm grinding wheel is dressed and the workpiece is machined with the dressed worm grinding wheel. For each grinding pass, there is a contact zone (104, 104') on the worm helical tooth surface in which the worm grinding wheel comes into machining engagement with the workpiece. Each contact zone has a quality-determining region. The quality-determining region of at least one part of the contact zones is arranged without overlap with the quality-determining regions of all the other contact zones, and a plurality of grinding passes are carried out with the contact zones respectively before the worm grinding wheel is dressed again. A tooth cutting machine of corresponding design is likewise disclosed.

Description

Technical Field

[0001] The present invention relates to a method for processing pre-formed teeth on a workpiece by continuous generating milling, and a tooth-making machine configured to perform the method. Background Technology

[0002] In gear manufacturing, the final hard finishing is one of the most important working steps concerning the quality of the resulting teeth. It shapes the geometry that will subsequently engage. A commonly used method for hard finishing is continuous generating grinding. Basic information on continuously generating grinding of teeth can be found, for example, in H. Schriefer et al., “Continuous Generating Grinding of Teeth,” Reishauer AG, Wallisellen, June 2008, ISBN 978-3-033-01447-3.

[0003] In continuous generating milling, a pre-formed toothed workpiece is machined in generating engagement by means of a rotating grinding wheel in the form of a worm gear (worm wheel). Machining takes place in one or more grinding strokes. During each grinding stroke, the worm wheel moves relative to the workpiece teeth along the workpiece axis (so-called axial feed motion), while the workpiece rotates about its workpiece axis in generating engagement with the worm wheel.

[0004] Typically, a workpiece is first machined in one or more roughing strokes, followed by one or more finishing strokes. Here, the roughing and finishing strokes are distinguished from each other by grinding intensity: in the finishing stroke, a smaller depth of cut and / or axial feed rate is usually selected compared to the roughing stroke, thus selecting a lower grinding intensity to ensure more precise compliance with quality specifications.

[0005] During each grinding stroke, a specific area of ​​the worm wheel engages with the corresponding workpiece. Due to this engagement, the worm wheel experiences wear in that area. To re-engage new, unused areas of the worm wheel with the workpiece, it is known to move the worm wheel discontinuously or continuously relative to the workpiece along its worm axis (the so-called "cutting" of the worm wheel). Different cutting strategies are known for this purpose.

[0006] In discontinuous tool traversal, when machining a workpiece, all roughing strokes are performed at the same traversal position on the worm wheel. Subsequently, a so-called tool jump to a traversal position for finishing is typically performed, and all finishing strokes are executed at this position. After finishing, a so-called tool return is performed to a new traversal position for roughing the next workpiece. The traversal position does not change during either the roughing or finishing strokes. Here, the operator typically selects the magnitude of the traversal jump and return based on preset formulas or empirical values.

[0007] In the "continuous feed" grinding system primarily used today, feed motion is additionally superimposed continuously during each roughing and finishing stroke. The amount of feed motion for each grinding stroke, or the resulting feed rate, is selected based on a preset formula or the operator's experience.

[0008] Dressable worm wheels are frequently used in generating mills. If a worm wheel has spun out over its entire usable area, it must be dressed individually. This is done using dressing tools typically coated with diamond. During dressing, the outer diameter of the wheel decreases. After a certain number of dressing cycles, the worm wheel is worn and must be replaced. The dressing tool also wears out with each dressing cycle. Therefore, the number of dressing cycles affects not only the lifespan of the worm wheel but also the lifespan of the dressing tool. Consequently, the cost of the worm wheel and dressing tool constitutes a significant portion of the operating cost. Furthermore, workpieces cannot be machined during dressing, extending the cycle time for each dressing cycle. Therefore, overall, dressing is not only a significant cost factor but also a significant time factor. Therefore, it is desirable, in principle, to design a spun-out strategy that allows as many workpieces as possible to be machined with the worm wheel while maintaining the required workpiece quality before it must be dressed again.

[0009] However, in practice, worm grinding wheels are often not optimally utilized. Therefore, it is important for many users to avoid manufacturing deviations at the finished workpiece at any cost. To this end, users often act overly cautiously, dressing the worm grinding wheel after machining a relatively small number of workpieces, even though the worm grinding wheel can actually produce other defect-free workpieces. Optimal utilization of the worm grinding wheel is also difficult for the following reasons: In traditional tool-shifting strategies, the wear condition of the worm grinding wheel often affects the obtained machining quality in unpredictable ways. Therefore, reliably predicting when the worm grinding wheel should be dressed is usually not feasible. Furthermore, in traditional tool-shifting strategies, the worm grinding wheel is usually worn unevenly across its width. Thus, it often must be dressed as soon as even a single area of ​​the worm grinding wheel is worn, even though other areas may not yet be worn to the same degree. Summary of the Invention

[0010] The purpose of this invention is to provide a method for machining pre-formed teeth on a workpiece by generating grinding, which achieves better utilization of the worm wheel.

[0011] The objective is achieved by the method of claim 1. Other embodiments are given in the dependent claims.

[0012] Therefore, a method is proposed for machining pre-formed teeth on a workpiece by continuous generating grinding using a worm wheel, the worm wheel rotating about a worm axis, wherein the method has the following characteristics:

[0013] Dressing the worm wheel; and

[0014] The workpiece is machined after the worm gear grinding wheel has been dressed.

[0015] Each workpiece is processed using at least one grinding stroke, and

[0016] During each grinding stroke, the worm wheel moves relative to the workpiece along the workpiece axis, while the workpiece rotates around the workpiece axis during machining engagement with the worm wheel (axial feed motion).

[0017] Each workpiece has workpiece teeth with (left and right) workpiece tooth surfaces. An effective area exists on each of these workpiece tooth surfaces. The area on the workpiece tooth surface that contacts the tooth surface of the mating tooth in subsequent use, thus representing the contact pattern of the tooth, is called the "effective area." The effective area can be preset during tooth design or determined by identifying the contact pattern under preset load conditions. The method may include presetting the effective area, for example, by using an input function to define the effective area in a suitable manner, such as graphically defining the effective area in a diagram of the corresponding workpiece tooth surface. If necessary, the entire tool tooth surface can also be defined as the effective area.

[0018] During each grinding stroke, the workpiece tooth surface is machined using at least one worm helical tooth surface of the worm grinding wheel. Typically, machining is performed using more than one worm helical tooth surface, in the case of multi-start worm grinding wheels and in common double-tooth meshing, for example, using the left and right tooth surfaces of each worm grinding wheel head. However, in the following discussion, only one worm helical tooth surface is considered.

[0019] For each grinding stroke, the following area on the worm gear helical tooth surface is defined as the "contact zone," in which the worm gear helical tooth surface contacts the workpiece during the relevant grinding stroke. Each contact zone has an area that contacts the effective area of ​​the workpiece tooth surface. This area is referred to below as the "quality-determining region" because it affects the workpiece quality in the important area of ​​the workpiece tooth surface, i.e., the effective area. If the entire workpiece tooth surface is considered as the effective area, then the quality-determining region extends over the entire contact zone; that is, the contact zone and the quality-determining region are thus identical.

[0020] According to the present invention, multiple contact areas exist on the helical tooth surface of the worm gear, the quality-determining regions of which substantially do not overlap with all other contact areas, and multiple grinding strokes are performed separately by means of these contact areas before the worm grinding wheel is re-dressed. The contact areas whose quality-determining regions substantially do not overlap with all other contact areas are hereinafter referred to as "decoupled contact areas".

[0021] In other words, there exist decoupled contact areas on the worm gear helical tooth surface, which are used not only for a single grinding stroke but also for multiple grinding strokes. By reusing the decoupled contact areas, the worm grinding wheel can be utilized particularly well. The grinding strokes implemented using specific decoupled contact areas are especially suitable for machining different workpieces; that is, there can be decoupled contact areas for machining two or more workpieces. It can be proposed that each workpiece undergoes only one completely specific, selected grinding stroke, such as the first, second, and third grinding strokes for the corresponding workpiece, using each decoupled contact area. In particular, it can be proposed that when machining a single workpiece, each grinding stroke is implemented using a different contact area (i.e., for example, the first grinding stroke using the first contact area, the second grinding stroke using the second contact area, and so on).

[0022] If the quality-determining area of ​​the contact zone does not overlap with any other contact zone, or the overlap is so small that the remaining overlap has no measurable effect on the quality of the effective area of ​​the tooth made by means of the related contact zone, then the quality-determining area of ​​the contact zone "substantially" does not overlap with another contact zone. For example, if no more than 25%, preferably no more than 15%, and particularly preferably no more than 5% of the area of ​​the quality-determining area of ​​the contact zone overlaps with one or more other contact zones, then the contact zone can still be considered decoupled.

[0023] To move from one decoupled contact zone to another, the worm wheel is moved relative to the workpiece's machining position between two grinding strokes along a traverse direction parallel to the worm axis, similar to the traverse jump and traverse return in the conventional traverse strategy described above. However, in the conventional traverse strategy, traverse jump and traverse return are implemented such that the quality-determining region of each contact zone at least partially overlaps with one or more other contact zones used previously or subsequently. In contrast, using decoupled contact zones allows for better control of the machining process because the wear state of the worm wheel in the quality-determining region of the decoupled contact zone is not affected by the wear state of the previously used, overlapping contact zones.

[0024] The proposed method does not preclude the existence of additional overlapping contact areas on the worm helical tooth surface, similar to those in conventional tool-shifting strategies, besides the decoupled contact areas. Furthermore, only a portion of the grinding stroke for each workpiece is performed using the decoupled contact areas, while one or more additional grinding strokes are performed using non-decoupled contact areas. For example, it is conceivable that only the roughing stroke is performed using the decoupled contact areas, while the finishing stroke's contact areas overlap, or vice versa. However, preferably, all contact areas used on the worm helical tooth surface are decoupled. Particularly preferably, before the worm wheel is re-dressed, all grinding strokes are performed using only the decoupled contact areas when machining all workpieces.

[0025] The proposed method also does not preclude the existence of one or more decoupled contact areas on the worm gear helical tooth surface, wherein each decoupled contact area is used only for performing a single grinding stroke at a single workpiece, for example, performing a first roughing stroke or a final finishing stroke at exactly one workpiece. This can then be meaningful when machining conditions are desired to be particularly controlled for the associated grinding strokes. However, preferably, each decoupled contact area is used for multiple grinding strokes and, in particular, for machining multiple workpieces before the worm wheel is re-dressed.

[0026] In a preferred embodiment, not only does the quality-determining region of the decoupled contact area substantially not overlap with all other contact area arrangements, but the decoupled contact area substantially does not overlap with all other contact area arrangements at all. That is, the decoupled contact area either does not overlap with other contact areas at all, or the decoupled contact area overlaps with other contact areas only to such an extent that this has no measurable impact on the quality of the tooth surface machined by means of the contact area, both within and outside the effective region.

[0027] The number of grinding strokes performed in each decoupled contact zone before the worm wheel is re-dressed can be pre-defined, or this pre-defined number can be automatically determined based on measurements during workpiece machining. Surprisingly, the relevant number can be very high, and can be at least 10, at least 20, or even at least 50, without any quality loss at the finished workpiece. Even 100 or more grinding strokes per contact zone before the worm wheel is re-dressed is feasible. Overall, this allows for machining several times more workpieces compared to conventional tool-shifting strategies before the worm wheel is re-dressed. Consequently, the lifespan of the worm wheel and dressing tool can be significantly increased compared to conventional tool-shifting strategies, thereby significantly reducing the cost of generating grinding methods.

[0028] To automatically determine the number of grinding strokes for each decoupled contact zone, it can be proposed to determine wear characteristic parameters characterizing the wear state of the worm wheel in the corresponding contact zone during the grinding stroke, and automatically decide based on these parameters whether to re-engage the same contact zone for a subsequent grinding stroke or a grinding stroke at a later time, or whether the contact zone will not be used for further grinding strokes until the worm wheel is re-dressed. Therefore, in this manner, each decoupled contact zone is repeatedly used in the grinding stroke until the wear characteristic parameters indicate that the contact zone is worn. This optimizes the number of workpieces processed between two dressing processes.

[0029] Wear characteristic parameters can be, in particular, intensity measures of grinding intensity during a relevant grinding stroke, especially the maximum, integral, average, or low-pass filtered spectral components of a power index determined during the relevant grinding stroke. The power index can be, in particular, the power consumption and / or current consumption of the tool spindle, which drives the worm wheel to rotate. This is based on the consideration that, given a certain depth of cut, less and less worm wheel material is available for the cutting process as wear increases, and furthermore, the surface properties of the worm wheel teeth change. Overall, this manifests as a decrease in grinding intensity. However, wear characteristic parameters do not necessarily need to be based on grinding intensity. For example, wear characteristic parameters can also be determined based on spectral analysis of structure-acoustic signals recorded during the corresponding grinding stroke.

[0030] The determination of whether a contact area is worn can be made, in particular, by comparing wear characteristic parameters with a threshold. However, more complex determination criteria can also be envisioned, such as a determination that additionally considers the corresponding wear characteristic parameters during another grinding stroke at the same workpiece. For example, it can be determined that if, on the one hand, the intensity metric mentioned above during the roughing stroke is below a first threshold, and on the other hand, the intensity metric determined for the subsequent finishing stroke exceeds a second threshold, then the contact area for the roughing stroke of the corresponding workpiece is considered worn and no longer used for further grinding strokes. Instead of simply comparing only the corresponding current value of the wear characteristic parameter with a threshold, considerations that better distinguish between true trends and random fluctuations can be envisioned, for example, by monitoring how many of the last N grinding strokes performed with the relevant contact area exceed or fall below the threshold and making a decision based on this, or by performing a (linear or nonlinear) regression analysis of the wear characteristic parameter and comparing at least one value of the regression function thus determined with a threshold.

[0031] Preferably, for each workpiece, during the grinding stroke of machining the relevant workpiece via a decoupled contact area, a tool traverse motion is performed simultaneously with the axial feed motion; that is, the worm wheel moves continuously relative to the workpiece not only along the axial feed direction but also along the tool traverse direction (so-called diagonal motion). Thus, during the machining engagement in the relevant grinding stroke, the worm wheel generally travels a specific tool traverse distance relative to the workpiece along the tool traverse direction and a specific axial travel distance along the axial feed direction. The ratio of the tool traverse distance to the axial travel distance is called the "diagonal ratio." In the case of a cylindrical worm wheel and a cylindrical workpiece, the contact area on the worm helical tooth surface corresponds to a single contact path without tool traverse motion. Here, if neither axial feed movement nor tool traverse motion occurs, the area on the worm helical tooth surface is called the contact path, in which the worm helical tooth surface achieves cutting engagement with the workpiece during the grinding stroke. Each contact path has a conical spiral extension on the worm wheel tooth surface. Because the contact between the worm wheel and the workpiece is not a precise point, but occurs with a certain cutting width, the contact path has a finite width. If a tool shift occurs, each contact area no longer corresponds to a single contact path, but consists of multiple partially overlapping contact paths that move relative to each other along the tool width direction.

[0032] If a tool traverse is performed during the grinding stroke, only a portion of the contact path in the normal contact area contacts the effective area of ​​the workpiece tooth surface, and this contact path typically does not contact the effective area along its entire length. In this respect, only a portion of the normal contact path extending along the corresponding contact path forms the quality-determining region of the contact area. Conversely, if no tool traverse is performed during the grinding stroke, in the case of a cylindrical workpiece and a cylindrical tool, each contact area corresponds to exactly one contact path. In this case, the central segment of the normal contact path forms the quality-determining region of the contact area. Conversely, if the entire workpiece tooth surface is considered the effective area, all contact paths along their entire length form the quality-determining region of the contact area, meaning the entire contact area becomes the quality-determining region.

[0033] Preferably, the worm wheel in the decoupled contact zone either has no modification of the worm helical tooth surface at all, or the worm wheel in the decoupled contact zone has a modification of the worm helical tooth surface such that, when viewed along the contact path, the modification remains unchanged during the tool-width movement within the contact zone from the contact path to the contact path. The modification of the worm wheel in the decoupled contact zone can particularly be a pure tooth profile modification, i.e., a modification of the worm wheel tooth profile such that the modification does not change in the width direction of the worm wheel across these decoupled contact zones. This can particularly be a tooth profile angle modification, tooth profile convexity, tooth profile waviness, tooth tip trimming, and / or tooth root trimming. Thus, despite the diagonal movement, either no modification is produced on the workpiece tooth surface, or a pure tooth profile modification that does not change along the width direction of the workpiece is produced on the workpiece tooth surface. Therefore, preferably in this method, diagonal motion is not used to transfer the variable width profile of the worm gear tooth surface to the workpiece tooth surface in order to produce a profiled workpiece tooth surface in the workpiece width direction. Correspondingly, the selection of the diagonal ratio is purely from a technical perspective, such as the worm wheel's lifespan, workpiece quality, and / or the roughness structure of the workpiece tooth surface (as in so-called "low noise shifting," where diagonal grinding methods are used to specifically affect the roughness structure, e.g., see US6379217B1), but not from the perspective that a variable width geometric profile of the workpiece tooth surface should be produced.

[0034] To achieve the desired decoupling of the contact area, the following can be done: Assume the worm wheel occupies a first homing position relative to the workpiece along the homing direction at the end of the machining engagement in the first grinding stroke, and the worm wheel occupies a second homing position at the beginning of the machining engagement in the subsequent (but not necessarily immediately following) second grinding stroke. Preferably, there exists a gap between the first and second homing positions such that the gap corresponds at least to the width of the contact path formed on the tooth surface of the worm wheel during the first grinding stroke, measured along the worm axis, and the worm wheel occupies only homing positions outside the region between the first and second homing positions during all grinding strokes. In this way, the contact areas used in the aforementioned grinding strokes are completely non-overlapping.

[0035] The width of the contact path is determined by the variation of the cutting width along the contact path. The cutting width can be variable along the contact path, as the meshing relationship changes along the contact path, and it can also be variable during the grinding stroke, for example, because multiple teeth of the workpiece are machined via the same contact path and the machining allowance of the workpiece may fluctuate between teeth due to pitch errors in the workpiece teeth. Therefore, it is meaningful to determine the variation of the cutting width along the contact path when removing the nominal machining allowance corresponding to the maximum expected machining allowance, thereby determining the nominal width of the contact path during the first grinding stroke, and selecting the spacing between the first and second traverse positions such that the spacing, measured along the worm axis, at least corresponds to the width. In addition to the nominal machining allowance, the macro-geometry of the workpiece (especially module, number of teeth, tooth profile), the macro-geometry of the worm wheel (especially module, number of heads, engagement angle), and machining parameters (especially the depth of cut and the axial feed per revolution of the workpiece) are particularly incorporated into the calculation of the cutting width.

[0036] Expressed numerically, preferably, the distance between the first and second tool shift positions satisfies the following relationship:

[0037] ,

[0038] Where a p,0 This represents the (nominal or actual) width of the contact path extending perpendicularly to the contact path, where α represents the meshing angle of the worm grinding wheel in its axial section. Here, the width of the contact path extending perpendicularly to the contact path corresponds, in a good approximation, to the width of the contact path measured along the tooth profile height direction of the worm grinding wheel in its axial section, and in this regard, this width can also alternatively be used for α. p,0 In worm grinding wheels with different meshing angles on the right and left worm helical tooth surfaces, The values ​​of these parameters may differ for the right and left worm gear helical surfaces. In this case, it is preferable to select the larger of these values.

[0039] In some implementations, for each workpiece, a power index characterizing instantaneous grinding power is continuously measured at least during machining with the aid of the decoupled contact zone, and at least one grinding parameter is controlled in a closed loop based on the power index. By decoupling the contact zones where machining engagement occurs during the selected grinding stroke, this closed-loop control becomes significantly more reliable than when using contact zones whose quality-determining regions overlap with other contact zones, because the wear state of the worm wheel in the other contact zone does not affect the closed-loop control in the corresponding quality-determining region.

[0040] As grinding parameters, the axial feed rate (i.e., the speed of axial feed motion) can be controlled in a closed loop. Here, at least one grinding parameter is preferably controlled in a closed loop such that during the relevant grinding stroke, the power index follows the target curve or does not exceed a preset maximum value.

[0041] To reduce the correlation between power performance metrics and process parameters such as tool diameter, workpiece diameter, workpiece module, or depth of cut, thereby facilitating the preset of target curves or maximum values ​​and improving closed-loop control, a normalization operation can be performed when determining the power performance metrics to normalize them. This normalization operation is related to at least one process parameter, which is at least one geometric parameter of the tool, at least one geometric parameter of the workpiece, and / or at least one setpoint parameter of the machine tool. The normalization operation is performed such that, compared to when no normalization operation is performed, the normalized power performance metrics are less correlated with at least one process parameter.

[0042] If at least one of the mentioned process parameters changes, the normalization operation is preferably recalculated. Recalculating the normalization operation may, in particular, involve applying a model describing the expected correlation between power indices and process parameters, especially a model of process force or process power.

[0043] For further consideration of the normalization operation, please refer to the published document WO2021048027A1, the content of which is fully incorporated into this publication through reference.

[0044] Furthermore, a gear-making machine is proposed for machining pre-formed teeth on workpieces by continuous generating grinding, the gear-making machine being specifically configured to implement the method disclosed above. The gear-making machine has:

[0045] A tool spindle configured to clamp a worm wheel thereon so as to drive the worm wheel to rotate about the tool axis;

[0046] A workpiece spindle, which is configured to clamp a pre-made toothed workpiece thereon so as to drive the pre-made toothed workpiece to rotate about the workpiece axis.

[0047] An axial slider is configured to generate an axial feed motion of the tool spindle relative to the workpiece spindle along an axial feed direction having a component parallel to the workpiece axis.

[0048] A cutting tool slider, configured to generate cutting tool movement along a cutting tool direction extending parallel to the worm axis; and

[0049] A machine tool control device, which is configured to implement one of the methods disclosed above. Attached Figure Description

[0050] Preferred embodiments of the present invention are described below with reference to the accompanying drawings, which are for illustrative purposes only and should not be interpreted restrictively. The drawings show:

[0051] Figure 1 A schematic diagram illustrating a worm gear grinding wheel meshing with a gear-shaped workpiece;

[0052] Figure 2 An exemplary map illustrates the time variation curve of power consumption of the tool spindle driving the worm wheel during the grinding stroke;

[0053] Figure 3A An exemplary map illustrates the grinding intensity during roughing when machining multiple workpieces in a conventional tool-following strategy.

[0054] Figure 3B An exemplary map illustrates the grinding intensity during finishing when machining multiple workpieces using a conventional tool-following strategy.

[0055] Figure 4 The diagram illustrates the helical tooth surface of a worm gear, which has a contact path shown thereon.

[0056] Figure 5 The diagram illustrates a portion of the contact path that corresponds to the instantaneous contact area.

[0057] Figure 6 A diagram illustrating the worm gear helical tooth surface with two partially overlapping contact areas;

[0058] Figure 7 A diagram illustrating the tooth surface of a worm gear with two non-overlapping contact areas;

[0059] Figure 8 A diagram illustrating a virtual planar tooth with a contact path;

[0060] Figure 9The diagram illustrates the contact area on the virtual planar teeth in the first pass strategy;

[0061] Figure 10 The diagram illustrates the contact area on the virtual planar teeth in the second cutting strategy;

[0062] Figure 11A An exemplary map illustrates the grinding intensity during roughing when the same roughing contact area is repeatedly used to machine multiple workpieces;

[0063] Figure 11B An exemplary map illustrates the grinding intensity during finishing when the same finishing contact area is repeatedly used to process multiple workpieces;

[0064] Figure 12 The following chart shows the grinding intensity for roughing (vertical axis) and finishing (horizontal axis) of a large number of workpieces;

[0065] Figure 13A An exemplary map illustrates the time-varying power consumption curve of the tool spindle when the axial feed rate is closed-loop controlled to ensure that the power consumption does not exceed the maximum value.

[0066] Figure 13B An exemplary map illustrates the time-varying curves corresponding to the axial feed rate;

[0067] Figure 14A An exemplary map illustrates the time-varying curve of the tool spindle's power consumption when the axial feed rate is closed-loop controlled so that the power consumption follows a set curve.

[0068] Figure 14B An exemplary map illustrates the time-varying curves corresponding to the axial feed rate;

[0069] Figure 15A An exemplary map illustrates the tooth surface of a workpiece with an effective area;

[0070] Figure 15B Exemplary map illustrations show the worm gear tooth surface with a quality-determining region; and

[0071] Figure 16 Illustrated explanation of a gear grinding machine. Detailed Implementation

[0072] Kinematics of Continuous Spreading Mill

[0073] Figure 1The diagram illustrates the kinematics of continuous generating grinding. A worm wheel (i.e., a worm-shaped forming wheel) 100 rotates about the worm axis B. The worm wheel meshes with a gear-shaped workpiece 200 having workpiece teeth 201 via an alternating-axis helical gear transmission. The workpiece 200 rotates about the workpiece axis C. During the grinding stroke, the worm wheel 100 moves relative to the workpiece 200 along an axial feed direction Z extending parallel or inclined to the workpiece axis, preferably across the entire width of the workpiece teeth 201 (the so-called axial feed motion). The position of the worm wheel 100 relative to the workpiece 200 along the radial cutting direction X determines the amount of material removed during the grinding stroke. To engage new, unused areas of the worm wheel 100 with the workpiece 200, the worm wheel is moved continuously or discontinuously relative to the workpiece 200 along a cutting direction Y extending parallel to the worm axis B (the so-called tangential cutting motion).

[0074] All the above linear motions (cutting, axial feed motion, and tool movement) can be understood as relative motions, which can be selectively generated by the corresponding motion of the worm wheel in space and / or by the motion of the workpiece in space.

[0075] Machining of workpiece 200 is typically performed over multiple grinding strokes, but can also be done in a single grinding stroke. If multiple grinding strokes are performed, they are usually one or more roughing strokes, followed by one or more finishing strokes. Here, the roughing stroke is typically performed with a larger X-direction infeed compared to the finishing stroke. In the following text, the subscript 'r' is used for roughing strokes and the subscript 'f' is used for finishing strokes.

[0076] For the considerations below, it is assumed that both the workpiece and the worm wheel have a basic cylindrical shape. However, these considerations can also be applied to cases where the workpiece and / or the worm wheel have different basic shapes, such as when machining conical workpieces or using conical, barrel-shaped, or spherical worm wheels. The workpiece teeth can be straight or helical.

[0077] Measurement of grinding power

[0078] Figure 2 An exemplary diagram illustrates the time-varying power consumption P(t) of the tool spindle driving the worm wheel 100 to rotate during the grinding stroke. Rapid fluctuations in power consumption, such as those caused by continuously changing contact conditions between the worm wheel and the workpiece, are not considered in this diagram. In practice, such rapid fluctuations can be filtered out using suitable filtering, for example, by means of a low-pass filter.

[0079] Power consumption initially increases continuously during the worm's infeed into the teeth, then levels off, and decreases continuously again during the outfeed. The maximum power consumption is referred to below as the grinding intensity I. s .

[0080] The power consumption of the tool spindle is an example of a power index characterizing instantaneous grinding intensity. Instead of power consumption, the current consumed by the tool spindle can also be used as a power index, which ultimately corresponds to torque measurement and thus approximately corresponds to cutting force measurement. Under constant voltage, the consumed current is proportional to the power, or under constant rotational speed, the torque is proportional to the power. In this respect, current can at least approximately characterize grinding intensity. Grinding intensity I s This is a scalar calculated from the time-varying curve of the power index and characterizing the grinding intensity overall during the grinding stroke. Instead of the maximum value of the power index, another scalar, such as the integral of the power index or (if appropriate, a weighted) time average, can also be used as the grinding intensity.

[0081] Grinding strength depends on a series of parameters, especially the radial infeed between the worm wheel and the workpiece, the speed of the axial feed motion (axial feed rate), the rotational speeds of the tool spindle and the workpiece spindle, the geometry of the worm wheel, the geometry of the workpiece, and the tribological conditions in contact (lubricant, surface properties of the worm and the material, and the material itself).

[0082] However, grinding strength is also affected by the wear condition of the worm wheel. Therefore, a partially worn worm wheel cannot cut the same amount of material during the grinding stroke as a newly dressed worm wheel. Consequently, increased wear of the worm wheel leads to a decrease in cutting power.

[0083] Therefore, given the grinding parameters, the grinding strength I s It can be used as an indicator of the wear condition of the worm grinding wheel in the current use area of ​​the worm grinding wheel.

[0084] Grinding intensity when using traditional tool shifting strategies

[0085] Traditional tool-shifting strategies typically result in the first batch of workpieces being machined after dressing using a worm wheel area that experiences increased wear between workpieces until equilibrium is reached, at which point the wear state no longer changes between workpieces. This should be addressed in relation to traditional tool-shifting strategies. Figure 3A and Figure 3B To elaborate further.

[0086] Therefore, it is first assumed that each workpiece is machined using a unique roughing stroke and a unique finishing stroke, wherein the tool traverses continuously during the roughing and finishing strokes, respectively. Furthermore, it is assumed that between the roughing and finishing strokes, the tool traverses jump to a finishing region of the worm wheel that has not yet been used for roughing, and that after finishing, the tool traverses precisely to the point where the previous roughing stroke ended. This corresponds to the conventional tool traverse strategy, as it is known from the prior art and frequently used in continuous generating mills.

[0087] exist Figure 3A The diagram illustrates the grinding intensity I during roughing when machining multiple workpieces n, specifically addressing this tool-shifting strategy. r (n), in Figure 3B The grinding intensity I during finishing is shown in the figure. f (n), where the worm wheel is re-dressed before machining the first workpiece. Due to increased wear in the roughing zone, the grinding intensity gradually decreases during the roughing of the first batch of workpieces until it reaches a substantially constant value. The opposite is true during finishing: the grinding intensity initially increases during the machining of the first batch of workpieces because less material is removed during the roughing stroke, and now correspondingly more material is removed during the finishing stroke, and then it flattens out.

[0088] The reasons for this performance should be examined in more detail below.

[0089] Contact path on worm grinding wheel

[0090] During the helical generating motion between the worm grinding wheel and the workpiece, at least one tooth surface of the worm grinding wheel helical teeth (worm helical tooth surface) and at least one tooth surface of the workpiece teeth (workpiece tooth surface) are always approximately in point contact. The contact point between the worm helical tooth surface and the workpiece tooth surface moves continuously on the relevant worm helical tooth surface during the machining of a single workpiece tooth surface due to the rotation of the worm grinding wheel 100 and the workpiece 200. The sequence of contact points on the worm grinding wheel is referred to hereinafter as the contact path. Figure 4 The diagram illustrates this contact path 102 on a single worm gear helical tooth surface 101. The contact path 102 has a tapered helical extension. For better distinction, the sequence of contact points generated on the workpiece tooth surface is referred to hereinafter as the contact trajectory.

[0091] However, when removing a limited machining allowance, the contact between the worm wheel 100 and the workpiece 200 is not precisely point-like. Instead, there is a finitely extended contact area at any given time, in which the worm wheel penetrates the workpiece tooth surface with the machining allowance in a cutting manner. Figure 5The schematic diagram illustrates a portion of the contact path 102 with contact area 103 on the worm gear helical tooth surface. The contact area 103 typically has an elongated shape along the contact path 102 and is asymmetrical in shape transverse to the contact path.

[0092] The extension of the contact area 103 transverse to the direction of the contact path is referred to below as the cutting width a. p Cutting width a p This depends particularly on the macroscopic geometry of the workpiece, the macroscopic geometry of the worm wheel, the removed machining allowance on the workpiece tooth surface, and the axial feed per workpiece revolution. The cutting width can be variable along the contact path. For example, in involute teeth, the meshing relationship varies along the involute, thus affecting the cutting width. Consequently, the contact path can have a width that changes along its extension.

[0093] Contact area during continuous cutter movement

[0094] Without a traverse motion, the worm wheel performs cutting work along a single contact path 102. With the introduction of continuous traverse motion, the contact path rotates continuously along the worm helical tooth surface during the grinding stroke, thus axially relative to the worm wheel, causing the contact path to move across the worm helical tooth surface 101.

[0095] In this way, during the grinding stroke, the area of ​​the worm gear helical teeth, called the "contact zone," engages with the workpiece as a whole. The contact zone engaging with the workpiece during a single grinding stroke has a generally conical, helical extension similar to a single contact path, but with a greater width compared to a single contact path due to the superimposed tool movement. Without tool movement, the contact zone corresponds to a single contact path.

[0096] exist Figure 6 Two contact areas 104 and 104' are shown schematically and non-proportional on the worm gear helical tooth surface 101. If the conventional tool-shifting strategy described above is followed, these contact areas engage during the selected grinding stroke (specifically, during the roughing stroke) when machining workpieces that follow each other. In this conventional tool-shifting strategy, the roughing stroke of the corresponding following workpiece begins precisely at the point where the roughing stroke of the preceding workpiece ends in the Y direction. In this case, the two contact areas 104 and 104' overlap due to the finite width of the contact path. The width of the overlapping area approximately corresponds to the width of a single contact path during roughing, i.e., the cutting width a. p .

[0097] In practice, this means that the wear condition of the worm wheel in the contact area 104 affects the machining conditions in the subsequent grinding strokes performed by means of the overlapping contact area 104'.

[0098] Variable grinding intensity in traditional tool shifting strategies

[0099] Based on the above considerations, the reason for this can be understood: in traditional traverse strategies, even if all grinding parameters remain constant, the grinding intensity is initially not constant from workpiece to workpiece. This is because the contact area used when machining subsequent workpieces partially overlaps with the contact area used when machining previous workpieces. In traditional traverse strategies, the amount of traverse by the worm wheel during each roughing or finishing stroke is typically significantly smaller than the cutting width measured along the worm axis. Figuratively speaking, the worm wheel thus gradually wears down at the beginning after dressing: the first workpiece is machined with a contact area that has not yet experienced any wear. The second workpiece is machined with a partially overlapping contact area, where a larger portion of the contact area has been used to machine the first workpiece and a smaller portion has not yet experienced any wear. In the third workpiece, part of the contact area has been used to machine the first two workpieces, another portion is used only to machine the single previous workpiece, and other portions have not yet experienced any wear, and so on. That is, each subsequent workpiece is machined with a contact area that gradually experiences wear until an equilibrium state is reached and the wear state of the worm in each contact area during the grinding stroke is the same as in the previous contact area.

[0100] The increase in wear in the contact area during rough machining manifests as rough machining intensity I. r The decrease is due to the decrease in cutting power as wear increases. As a result, this leads to an increase in finishing intensity, because more cutting work must be performed accordingly during finishing. This corresponds precisely to... Figure 3A and Figure 3B The observed performance is reproduced in the middle.

[0101] The above considerations can also be easily applied to situations where work is carried out using more than one roughing stroke and / or finishing stroke, or using only one grinding stroke.

[0102] Grinding using decoupled contact areas

[0103] To prevent the wear condition of a specific contact area from being affected by the wear condition of another partially overlapping contact area, it is meaningful to decouple the contact areas from each other so that they no longer overlap. In the aforementioned tool tracing strategy, this can be achieved, for example, by reducing tool tracing back.

[0104] This is Figure 7The exemplary diagram illustrates two contact areas 104, 104', which engage during the roughing stroke when machining workpieces that follow each other, provided the tool backlash is appropriately reduced. Now, contact areas 104, 104' extend side-by-side on the worm gear helical tooth surface 101 without overlapping, meaning they are decoupled from each other.

[0105] The amount of adjustment for the blade rebound.

[0106] The rebound of the cutting tool must be reduced in order to... Figure 7 In the example, the values ​​that decouple the contact areas 104 and 104' can be intuitively derived through geometric considerations.

[0107] Therefore, the reference tooth profile in the axial section of the worm grinding wheel is first considered, so that the reference tooth profile can mesh with the teeth of the worm grinding wheel and the teeth of the workpiece made by the worm grinding wheel, so that the teeth are generated on the reference tooth profile without deviation. Figure 8 This reference tooth profile 110 is schematically shown. The left tooth face of the reference tooth profile 110 has a profile angle α. The meshing segment of the left tooth face can be drawn above the reference tooth profile. The meshing segment extends perpendicular to the left tooth face of the reference tooth profile. In this respect, the profile angle α corresponds to the meshing angle in the axial section of the worm grinding wheel. However, instead of the meshing segment, in Figure 8 Draw a width of 'a' in the middle p The band. The band symbolically represents the contact path. At the location where the meshing section intersects with the corresponding left tooth surface, measured perpendicular to the direction of the contact path's extension, the contact path has a width 'a'. p Because the contact path extends in a conical spiral shape on the worm wheel tooth surface, primarily in the circumferential direction of the worm wheel, the width of the contact path measured perpendicular to the contact path direction corresponds well to its width measured in the height direction of the corresponding left tooth surface. Conversely, measured along the worm axis, the width of the band is approximately... This value is referred to below as "the width of the contact path measured along the axis of the worm gear." A similar consideration can be made for the right tooth face.

[0108] Therefore, if two consecutive contact zones should not overlap, then there must be at least a gap along the worm axis between the worm wheel's traverse position at the end of the machining engagement via the first contact zone and the traverse position at the beginning of the machining engagement via the second contact zone. .

[0109] The above considerations are simplified because they do not take into account, for example, the lead of the worm helical tooth and the varying cutting width 'a' between the tooth tip and root regions of the worm helical tooth. pAnd possible modifications to the worm gear tooth surface, but providing practically applicable approximate solutions that can be easily refined accordingly if necessary.

[0110] Calculation of cutting width

[0111] Cutting width a p It can be approximately calculated analytically or determined numerically through penetration calculation. One feasibility of analytical calculation is disclosed in WO2021048027A1. According to that document, a can be approximately calculated from technical parameters and tooth parameters as follows: p :

[0112]

[0113] Here, a p,max This represents the maximum expected width of cut for a large axial feed, given a given length and a given nominal machining allowance. k This represents the contact length of the cutting zone, and s z This indicates the axial feed per revolution of the workpiece.

[0114] The maximum cutting width can be calculated as follows:

[0115]

[0116] Here, L y Indicates the growth rate, and q t This indicates the nominal machining allowance.

[0117] For contact length l k The following parsing relationship can be derived:

[0118]

[0119] Here, r pSS,eq This represents the equivalent worm wheel radius. The equivalent worm wheel radius r pSS,eq From the actual worm wheel radius r pSS The following conclusions can be drawn:

[0120]

[0121] Here, α represents the meshing angle of the worm gear teeth in the axial section.

[0122] Therefore, the cutting width 'a' along the contact path can be approximately calculated for each machining condition. p Alternatively, a penetration calculation can be performed to determine the cutting width 'a' along the contact path, instead of an analytical calculation. p .

[0123] As mentioned above, the cutting width along the contact path can be variable. To achieve reliable decoupling of the contact area even with varying cutting widths, it is meaningful to proceed as follows: Using the maximum expected machining allowance on all teeth of the tooth section as a nominal machining allowance, which is removed from the tooth surface upon entry. Then, the cutting width 'a' along the contact path is determined for this nominal machining allowance. p The cutting width a along the contact path p The maximum value is assumed to be the width 'a' of the contact path measured along the tooth profile height direction of the worm gear helical tooth surface. p,0 The width of the contact path is then measured along the worm axis and divided by... Conclusion:

[0124] .

[0125] Tool shifting strategy using decoupled contact area

[0126] If the machining of each workpiece is not performed using a single stroke, but rather using at least one roughing stroke and at least one finishing stroke, then different tool shifting strategies can be envisioned, which take into account the above considerations and can achieve decoupling of the contact area. Two feasible tool shifting strategies will be illustrated below.

[0127] exist Figure 9 The diagram illustrates the first cutting strategy. This diagram is related to... Figure 8 Similarly, the contact area on the reference tooth profile is shown. However, for clarity, the reference tooth profile is not drawn here, and the shape and extension of the contact area are shown below. Figure 8 Compared to being shown in a significantly compressed manner, the width of the worm grinding wheel is expressed using... express. Figure 9 This can also be understood as a diagram of the (unfolded) worm gear helical tooth surface 101, where the horizontal axis represents the width position along the axis of the worm grinding wheel. The vertical axis represents the height position along the worm gear helical tooth surface. Or it can be developed into a path.

[0128] In this traverse tool strategy, the worm wheel is divided into a roughing zone 111 and a finishing zone 112, as in a conventional traverse tool strategy. A roughing stroke is performed first when machining the first workpiece. During the roughing stroke, traverse tool continuously along the Y direction. Thus, during the roughing engagement, the contact path moves along the worm wheel axis. The roughing stroke is thus achieved using the roughing contact area, the width of which, measured along the worm wheel axis or along the Y direction, corresponds to the amount of traverse feed during the roughing engagement plus the width of a single contact path. The amount of traverse feed during the roughing engagement is shown in the accompanying drawings. The width of a single contact path along the axis of the worm gear grinding wheel is expressed as Therefore, in general, during roughing, the following roughing contact area is utilized, which has a width measured along the worm axis. .

[0129] After the roughing engagement of the first workpiece in the roughing zone 111 is completed, with a (positive) value The tool jumps to the finishing zone 112, reaching the position where the finishing engagement of the first workpiece begins. Tool traversal continues continuously during the finishing stroke, where the traversal speed can be the same as or different from the roughing stroke. The total traversal feed during the finishing engagement is... During finishing, the width of a single contact path along the worm axis is Typically, due to the smaller depth of cut, the width of the contact path becomes smaller during finishing compared to roughing. Therefore, generally, during finishing, a finishing contact area is utilized, the finishing contact area having a width measured along the worm axis. .

[0130] After the finishing engagement ends, the tool traverse and return is performed before the roughing engagement of the second workpiece begins. However, unlike in the traditional traverse strategy, the tool now returns precisely to the same point where the roughing engagement of the first workpiece has also begun. In other words, the same roughing contact area used for the roughing stroke of the first workpiece is now used a second time for the roughing stroke of the second workpiece. The finishing stroke of the second workpiece is also performed using the same finishing contact area that was also used for the finishing stroke of the first workpiece.

[0131] These contact areas are now used to process additional workpieces until one of the contact areas has worn out. After the finishing stroke of the last workpiece processed using these contact areas, the machining is now performed with (negative) values. To reduce tool bounce, select a value that is suitable for: In this way, the roughing contact area of ​​the next workpiece is decoupled from the roughing contact area of ​​the previous workpiece.

[0132] exist Figure 10 The diagram illustrates the second tool shifting strategy. In this strategy, the first set of workpieces is machined via the worm wheel region 121, the second set of workpieces is machined via the adjacent worm wheel region 122, and so on. For each workpiece, a roughing stroke is performed via the roughing contact area, followed by a finishing stroke via the finishing contact area.

[0133] To prevent the roughing and finishing contact areas of the workpiece from overlapping, an additional tool traverse (tool jump) is performed between the end of roughing engagement and the beginning of finishing engagement for each workpiece. The magnitude of this additional tool traverse is... The size is chosen large enough that the first contact path at the start of the finishing engagement no longer overlaps with the last contact path at the end of the roughing engagement. This is achieved when the worm gear moves at least a certain magnitude between the end of the roughing engagement and the start of the subsequent finishing engagement. This is the case during tool shifting. As a result, the contact areas of the roughing stroke and the finishing stroke are decoupled from each other.

[0134] After the finishing engagement of the last workpiece in the first group is completed, and before the roughing engagement of the first workpiece in the subsequent second group begins, in terms of quantity... Perform additional cutting motion, the magnitude of which is at least [value missing]. Thus, the roughing contact area for the workpieces in the second group is decoupled from the finishing contact area for the workpieces in the first group. Subsequently, the machining of the workpieces in the second group is performed in the same manner as the machining of the workpieces in the first group. This decouples all roughing and finishing contact areas.

[0135] If more than one roughing stroke and / or more than one finishing stroke are performed, the corresponding contact areas are preferably also decoupled from each other. The same approach can be taken if only one machining stroke is performed per workpiece.

[0136] Of course, many other tool-shifting strategies with decoupled contact areas are feasible, in which multiple grinding strokes are performed separately in the decoupled contact areas. Here, it is not necessary for all contact areas to be used for the same number of grinding strokes. For example, the finishing contact area can be used for a significantly different number of grinding strokes than the roughing contact area; that is, the switching from one roughing contact area to another can be independent of the switching between finishing contact areas. The switching to another contact area can also be independent of whether the relevant contact area has worn. For example, if there are three roughing contact areas, then, for instance, a first workpiece can be machined using a first roughing contact area, a second workpiece using a second roughing contact area, and a third workpiece using a third roughing contact area, then a fourth workpiece is machined again using the first roughing contact area, a fifth workpiece using the second roughing contact area, and a sixth workpiece using the sixth roughing contact area, and so on. In this way, all roughing contact areas always have a similar wear state.

[0137] A variant is also conceivable in which a pool of decoupled contact areas of equal width exists. Each of these contact areas is initially used as a finishing contact area during one or more grinding strokes. Once the relevant contact area no longer provides sufficient quality during finishing (which can be determined, for example, based on the grinding intensity), the contact area is subsequently used as a roughing contact area for one or more roughing strokes.

[0138] Considerations for the selection of grinding parameters

[0139] When machining with decoupled contact zones, it is advantageous to select grinding parameters (especially radial infeed and axial feed rates) that are not too aggressive, so that the wear of the worm wheel in the corresponding contact zone is kept small during each grinding stroke. It is also advantageous to select a relatively large ratio (i.e., diagonal ratio) between the traverse distance and the axial travel distance for each contact zone, so that the wear of each grinding stroke is distributed over a relatively large contact zone.

[0140] This makes it possible to grind a surprisingly high number of workpieces using the same contact area without unacceptable quality loss. For example, it has been shown that dozens, sometimes even more than 100 grinding strokes can be performed per contact area without quality loss. Consequently, the lifespan of the worm wheel can be increased several times compared to traditional tool-shifting strategies. Thus, the slight reduction in productivity that might result from more careful selection of grinding parameters is largely offset.

[0141] Grinding intensity when grinding repeatedly using the same contact area

[0142] exist Figure 11A and Figure 11BThe diagram illustrates the variation curves of grinding intensity when using a single roughing contact area and a finishing contact area in one of the tool-shifting strategies described above.

[0143] Because the roughing contact area is used to machine multiple workpieces, it gradually wears down. This increased wear manifests as follows: Figure 11A In the roughing grinding process, the grinding intensity I r The value decreases continuously between workpieces without reaching a substantially constant value, as in this case. Figure 3A This is similar to the situation in traditional tool-shifting strategies. Increased wear in the roughing contact area causes the need to gradually provide more cutting work during finishing. Therefore, in... Figure 11B In the middle, the grinding strength I in finishing is f The amount increases continuously between workpieces, but does not reach a basically constant value.

[0144] Unlike traditional tool-shifting strategies, this behavior allows for automatic determination of when to switch to a new, unused roughing or finishing contact area. In one implementation, for example, a lower threshold roughing intensity I can be defined. r,thr Once the strength falls below this level (or once it falls below this level for a specific number of workpieces, or below the appropriate operating average of the strength across multiple workpieces), the process switches to a new roughing contact zone. Correspondingly, an upper threshold finishing strength I is also defined. f,thr Once the stated intensity is exceeded (or exceeded for a specific number of workpieces, or exceeded by an appropriate operating average of the intensity across multiple workpieces), the process switches to a new finishing contact zone. Alternatively, when switching to a new roughing contact zone, a switch to a new finishing contact zone can always also be performed. Threshold I r,thr I f,thr For example, it can be determined empirically by measuring at a worm gear grinding wheel that is known to be worn.

[0145] Instead of grinding intensity, any other wear index associated with the wear state of the worm wheel can be considered, and correspondingly, a threshold for the wear index can be defined for roughing or finishing, switching to the next contact zone when the wear falls below the threshold. For example, the low-frequency (e.g., low-pass filtered) spectral component of a normalized power index can also be used as a wear index.

[0146] Once all roughing and finishing contact areas have worn, the worm wheel is re-dressed. In this way, the worm wheel is used optimally, as it is only dressed when it must be considered worn by objective standards.

[0147] Switching to a new contact area based on more complex standards.

[0148] exist Figure 11A and Figure 11B In the example, the determination of when a specific contact area has worn down is based on a comparison of a single scalar wear metric with a threshold. Alternatively, this can be determined using more complex criteria. Figure 12 Provide an example.

[0149] The attached figure schematically illustrates the roughing intensity I for a large number of workpieces. r (Vertical axis) and finish strength I f (Horizontal axis), the workpiece is machined by a roughing stroke and a finishing stroke, wherein all roughing strokes are performed through the same roughing contact area and all finishing strokes are performed through the same finishing contact area.

[0150] Initially, roughing strength is relatively high and finishing strength is relatively low, as deduced above. As the number of workpieces increases, roughing strength decreases and finishing strength increases. Figure 12 (The direction of the arrow in the image). When the roughing contact area is worn, the roughing intensity is below the threshold I. r,thr And the finishing strength exceeds the threshold I f,thr The decision to switch to a new, unused roughing or finishing contact area can now be made taking into account these two thresholds. For example, if the following two conditions: I r < I r,thr I f > I f,thr If a certain number of workpieces share the same requirement, then switching can be performed. This corresponds to... Figure 12 The region referred to as the "critical" region.

[0151] Here, any other wear index can be considered as an alternative to grinding strength. In this regard, grinding strength during roughing or finishing is merely one example of a wear index for roughing or finishing.

[0152] Normalization of the measured grinding power

[0153] The power index, or the grinding strength calculated from it, can be appropriately normalized to improve the comparability of the power index under different conditions.

[0154] Here, the normalization operation is preferably based on a model that describes the expected correlation between process power and the geometric parameters of the worm wheel and workpiece, as well as the set parameters of the generating gear grinding machine. The process power model can be particularly based on a force model that describes the expected correlation between the cutting force acting at the contact point between the worm wheel and the workpiece and the geometric parameters of the finishing tool, the geometric parameters of the workpiece, and the set parameters of the finishing machine.

[0155] Normalization operations may include, for example, multiplying a measured value of a power index, or a quantity derived therefrom, by a normalization factor. However, more complex normalization operations are also conceivable. The normalization factor may, in particular, be an inverse intensity quantity, or a quantity derived therefrom, which has been calculated based on a process power model for a specific processing condition. For example, the process power model used in WO2021048027A1 can be used.

[0156] Preferably, the normalization factor is recalculated after each trimming process, and the same normalization factor is applied to all values ​​of the power index or the quantities derived therefrom that have been obtained between the two trimming processes.

[0157] Closed-loop control of axial feed rate based on grinding power.

[0158] If the grinding power becomes too high, it can cause manufacturing deviations, such as shape defects and / or thermal damage to the edge areas of the workpiece teeth (so-called grinding burns). Furthermore, there is a risk of damage to the worm wheel, for example, by the appearance of a break at one or more worm helices. To avoid this, the grinding process can be controlled in a closed loop. Specifically, at least one grinding parameter, particularly the axial feed rate, can be controlled. Closed-loop control is achieved by adhering to a preset closed-loop control strategy. Different closed-loop control strategies can be used here. (Further details will follow...) Figure 13A and Figure 13B and Figure 14A and Figure 14B Two feasible closed-loop control strategies are described. However, other closed-loop control strategies can also be considered.

[0159] exist Figure 13A The exemplary diagram illustrates the time-varying power index, in the form of the power consumption P(t) of the tool spindle during the grinding stroke, when the axial feed rate is closed-loop controlled such that the (preferably normalized) power index does not exceed its maximum value (“maximum value closed-loop control”). Without closed-loop control of the grinding process, in this example, the power consumption would at least temporarily exceed the preset maximum value P. max (Dotted line). To avoid this, the axial feed rate can be... Closed-loop control is applied to ensure that the grinding power does not exceed the maximum value. Once the grinding power approaches the maximum value, the feed rate is reduced based on closed-loop control to ensure that the grinding power does not exceed the maximum value. For this purpose, any known closed-loop control principle can be used, such as PID closed-loop control. Feed rate The resulting change curve is in Figure 13B Explanation with Chinese diagrams.

[0160] exist Figure 14AThe example illustrates the time-varying power consumption of the tool spindle when the axial feed rate is closed-loop controlled such that the power consumption at least approximately follows a setpoint curve (“setpoint curve closed-loop control”). The setpoint curve, for example, describes the expected power consumption based on a model of process power expected under ideal conditions assuming a specific machining allowance distribution. For example, the model of process power already mentioned above can be used to calculate the setpoint curve. Deviation from the setpoint curve may occur, for example, by the workpiece having a machining allowance distribution that deviates from the assumed machining allowance distribution; however, other factors may also play a role, such as hardness distribution and the instantaneous wear state of the worm wheel. The power consumption variation curve is illustrated as a dotted line, as it would be at a constant axial feed rate. In this example, this variation curve deviates significantly from the setpoint curve. For example, the power consumption is initially lower during the feed phase compared to the setpoint curve, but significantly higher subsequently. The closed-loop control compensates for the deviation by increasing the feed rate when power consumption deviates negatively from the set curve and decreasing the feed rate when it deviates positively. This is in... Figure 14B Exemplary map solution illustration.

[0161] The closed-loop control strategy can also be modified and / or combined. Therefore, in Figure 14B For example, a maximum value for the feed rate can be additionally limited, which cannot be exceeded, or closed-loop control of the feed rate can be performed only during a phase of the grinding stroke in which a preset minimum power consumption is exceeded. Many other modifications to the closed-loop control strategy are conceivable.

[0162] Compared to using overlapping contact areas, this strategy for closed-loop control of the grinding process can be used with much greater predictability and reliability by using decoupled contact areas. As long as the contact areas overlap, the wear state of the worm wheel varies during each individual grinding stroke. Therefore, grinding power is strongly influenced by the variable wear of the worm wheel during each grinding stroke; that is, closed-loop control reacts strongly to the wear of the worm wheel and thus fails to optimally achieve its true purpose. With decoupling, the closed-loop control is now primarily influenced by the characteristics of the workpiece, especially its variable machining allowance, and possibly by the inhomogeneity of the worm wheel and the wear state within the relevant contact area itself, but is no longer affected by the variable wear states of other contact areas.

[0163] Therefore, decoupling of the contact area paves the way for intelligent closed-loop control of the grinding process.

[0164] Consideration of the effective area of ​​the teeth to be manufactured

[0165] Typically, when using a machined workpiece in a transmission device, not the entire left and right sides of the workpiece tooth surface engage with the mating teeth; rather, only specific areas of the relevant tooth surface engage with the mating teeth. This area is referred to in this document as the "effective area." This area can be determined, for example, by determining the contact pattern of the teeth under actual operating conditions.

[0166] exist Figure 15A The exemplary diagram illustrates the tooth surface 210 of a workpiece with teeth, which has an associated effective region 211. The effective region 211 extends along the width direction of the workpiece. The effective area does not extend across the entire width of the workpiece tooth surface. Furthermore, the effective area extends along the workpiece height direction. It does not extend over the entire height of the workpiece tooth surface.

[0167] Therefore, the goal of workpiece machining can be to obtain the highest possible quality of the teeth within the effective region, while the quality of the teeth outside the effective region can be secondary. In this case, it is advantageous to determine the area on the worm gear helical tooth surface that contacts the effective region during tooth machining, thereby determining the quality.

[0168] exist Figure 15B The image exemplarily illustrates such a quality-determining region 105 of the contact area 104 on the worm gear helical tooth surface 101. The quality-determining region can be determined, for example, through machining simulation as is well known to those skilled in the art, or through machining experiments.

[0169] If, instead of the entire contact area, only the quality-determining region of the contact area does not overlap with all other contact area arrangements, then the advantages described above can also be achieved. In this case, the wear state in the quality-determining region is not affected by the wear state of all other contact areas. Therefore, the following contact area can also be regarded as a "decoupled" contact area, in which only the quality-determining region does not overlap with all other contact area arrangements, and all the above considerations also meaningfully apply to such a decoupled contact area.

[0170] Exemplary construction of a generating gear grinding machine

[0171] exist Figure 16 The diagram exemplarily illustrates a generating gear grinding machine 1 suitable for performing the above-described method. The generating gear grinding machine is also referred to herein simply as a "machine tool".

[0172] Machine tool 1 has a base 11, on which a tool holder 12 is movably guided along a radial infeed direction X. The tool holder 12 carries an axial slide block 13, which is movably guided relative to the tool holder 12 along an axial feed direction Z. A grinding head 14 is mounted on the axial slide block 13, and the grinding head is pivotable about a pivot axis (so-called axis A) extending parallel to the X direction to match the helix angle of the tooth to be machined. The grinding head 14 also carries a tool slide block, on which a tool spindle 15 is movable relative to the grinding head 14 along a tool slide direction Y. A worm wheel 100 is clamped on the tool spindle 15. The worm wheel 100 is driven by the tool spindle 15 to rotate about a tool axis B.

[0173] Furthermore, the base 11 carries a pivotable workpiece carrier 20 in the form of a turret, which can pivot about a pivot axis C3 between at least three positions. Two identical workpiece spindles are mounted on the workpiece carrier 20, diametrically opposed to each other. Figure 16 Only one workpiece spindle 21 is visible. Workpieces can be clamped individually onto each workpiece spindle and can be driven to rotate about workpiece axes C1 or C2. Figure 16 The workpiece spindle 21, visible in the image, is in a machining position, in which the workpiece 200 clamped on the workpiece spindle can be machined using a worm gear grinding wheel 100. The workpieces are offset by 180° and... Figure 16 Another workpiece spindle, not visible in the center, is in the workpiece changing position, where the machined workpiece can be removed from this spindle and a new blank can be clamped. The dressing device 30 is installed at a 90° offset relative to the workpiece spindle.

[0174] Therefore, machine tool 1 has multiple movable components, such as sliders or spindles, which can be controlled to move by corresponding drive devices. These drive devices are often referred to in the art as "NC axes," "machine tool axes," or simply "axis." The name sometimes also includes components driven by the drive device, such as sliders or spindles.

[0175] Machine tool 1 also has multiple sensors. Figure 1 Only two sensors, 18 and 19, are schematically shown in the example. Sensor 18 is a vibration sensor for detecting vibrations of the housing of the grinding wheel spindle 15. Sensor 19 is a position sensor for detecting the position of the axial slider 13 relative to the tool holder 12 along the Z direction. However, the machine tool 1 also includes several other sensors. Among these sensors, there are in particular additional position sensors for detecting the actual position of linear axes, rotation angle sensors for detecting the rotational position of rotary axes, current sensors for detecting the drive current of axes, and additional vibration sensors for detecting the vibrations of driven components.

[0176] All driven axes of machine tool 1 are digitally controlled by machine tool control unit 40. Machine tool control unit 40 includes multiple axis modules 41, a control computer 42, and a control console 43. The control computer 42 receives operator commands from the control console 43 and sensor signals from various sensors of machine tool 1, and thereby calculates control commands for the axis modules 41. Furthermore, the control computer outputs operating parameters to the control console 43 for display. The axis modules 41 provide control signals to the machine tool axes at their outputs. In this way, all the movements mentioned above can be controlled, especially radial feed, axial feed, and traverse feed movements.

[0177] The monitoring device 44 is connected to the control computer 42.

[0178] The monitoring device 44 may be a separate hardware unit associated with the machine tool 1. The monitoring device may be connected to the control computer 42 via a known interface, such as the known Profinet standard, or via a network, such as the Internet. The monitoring device may be spatially part of the machine tool 1, or it may be spatially located away from the machine tool 1.

[0179] During machine tool operation, the monitoring device 44 receives multiple different measurement data from the control computer 42. The measurement data received by the control computer includes sensor data directly detected by the control computer 42 and data read by the control computer 42 from the axis module 41, such as data describing the desired positions of different machine tool axes and the desired current consumption in the axis module.

[0180] The monitoring device 44 may optionally have its own analog and / or digital sensor inputs to receive sensor data directly from other sensors as measurement data. These other sensors are typically those not directly used to control the actual processing, such as accelerometers for detecting vibrations or temperature sensors.

[0181] Alternatively, the monitoring device 44 may be implemented as a software component of the machine tool control device 40, which executes, for example, on the processor of the control computer 42, or the monitoring device may be configured as a software component of the business server 45, which is described in detail below.

[0182] The monitoring device 44 communicates directly or via the Internet and a network server 47 with the business server 45. The business server 45, in turn, communicates with a database server 46 that has a database DB. These servers can be located remotely from the machine tool 1. The server does not need to be the only physical unit. In particular, the server can be implemented as a virtual unit in the so-called "cloud".

[0183] The business server 45 communicates with the mobile terminal device 48 via the web server 47. The terminal device 48 can, in particular, run a web browser to visualize the received data and its evaluation. The terminal device does not need to meet special requirements for computing performance. For example, the terminal device can be a desktop computer, laptop computer, tablet computer, mobile phone, etc.

[0184] Workpiece machining

[0185] For completeness, the typical machining of a workpiece using machine tool 1 is described below.

[0186] To process an unprocessed workpiece (blank) 200, an automatic workpiece changer clamps the workpiece 200 onto the workpiece spindle in the workpiece changing position. The workpiece changing and the processing of another workpiece on the other workpiece spindle in the processing position are performed in parallel time. When the new workpiece 200 is clamped and the processing of the other workpiece is complete, the workpiece carrier 20 pivots 180° around axis C3, bringing the spindle with the new workpiece into the processing position. Before and / or during the pivoting process, a centering operation is performed using an associated centering probe. For this purpose, the workpiece spindle 21 is placed in rotation, and the orientation of the tooth grooves of the workpiece 200 is measured using the centering probe. The generating angle is determined based on this.

[0187] When the workpiece spindle carrying the workpiece 200 reaches the machining position, the workpiece 200 engages with the worm wheel 100 without collision by moving the tool carrier 12 along the X-axis. The workpiece 200 is now machined by the worm wheel 100 in generating engagement. This is done through one or more grinding strokes, such as one or more roughing strokes followed by one or more finishing strokes. During each grinding stroke, the worm wheel 100 is continuously fed relative to the workpiece 200 along the Z-axis with a constant or variable radial X-cut (so-called axial stroke). During the machining stroke, the tool spindle 15 can move slowly and continuously along the Y-axis. Alternatively or additionally, a jump cut can be made between two grinding strokes.

[0188] In parallel with the machining of the workpiece, the completed workpiece is removed from another workpiece spindle, and another blank is clamped on the same spindle.

[0189] If, after machining a certain number of workpieces, the worm wheel 100 becomes too dull and / or its tooth surface geometry becomes too inaccurate, then the worm wheel needs to be dressed. For this purpose, the workpiece holder 20 is pivoted ±90° so that the dressing device 30 is positioned opposite the worm wheel 100. The worm wheel 100 is then dressed using a dressing tool 33.

[0190] List of reference numerals

[0191] 100 worm grinding wheel

[0192] 101 worm gear helical tooth surface

[0193] 102 Contact Path

[0194] 103 contact area

[0195] 104 contact area

[0196] 104' contact area

[0197] 105 Quality Determines the Area

[0198] 110 reference tooth profile

[0199] 111 Worm Grinding Wheel Area for Rough Machining

[0200] 112 Worm Grinding Wheel Area for Finishing

[0201] 121 is the worm gear grinding wheel area used for the first batch of workpieces.

[0202] 122 is used for the worm gear grinding wheel area of ​​the second batch of workpieces.

[0203] 200 workpieces

[0204] 201 workpiece teeth

[0205] 210 Tooth surface of workpiece teeth

[0206] 211 effective area

[0207] 1. Generating Gear Grinding Machine

[0208] 11-frame

[0209] 12-tool holder

[0210] 13-axis slider

[0211] 14 grinding heads

[0212] 15-tool spindle

[0213] 18 vibration sensors

[0214] 19 position sensors

[0215] 20 workpiece brackets

[0216] 21 Workpiece Spindle

[0217] 41-axis module

[0218] 42 Control Computer

[0219] 43 Console

[0220] 44 monitoring devices

[0221] 45 business servers

[0222] 46 databases

[0223] 47 network servers

[0224] 48 mobile terminal devices

[0225] B worm shaft axis

[0226] C workpiece axis

[0227] X radial tangential direction

[0228] Y-axis cutting direction

[0229] Z-axis feed direction

[0230] Workpiece width

[0231] Worm Grinding Wheel Width

[0232] a p Cutting width

[0233] Tool shift value during roughing engagement

[0234] Tool shift value during finishing engagement

[0235] During rough machining, the contact path is along the width of the worm axis.

[0236] During rough machining, the contact path is along the width of the worm axis.

[0237] The amount of tool jump between roughing and finishing

[0238] The amount of tool runout between finishing and the next roughing operation

[0239] Power consumption of P-grinding wheel spindle

[0240] P max Maximum power consumption

[0241] I s Grinding strength

[0242] I r Grinding intensity during rough machining

[0243] I fGrinding intensity during finishing

[0244] I r,thr Threshold of grinding intensity during roughing

[0245] I f,thr Threshold of grinding intensity during finishing

[0246] t time

[0247] n number of workpieces

[0248] au (arbitrary units)

Claims

1. A method for machining a workpiece (200) with pre-formed teeth by continuous generating grinding using a worm wheel (100), the worm wheel rotating about a worm axis (B), wherein the method comprises: Dressing the worm wheel (100); and After the worm wheel (100) is dressed, the workpiece (200) is machined. Each workpiece (200) is machined by means of at least one grinding stroke, wherein during each grinding stroke, the worm wheel (100) moves relative to the associated workpiece (200) along the workpiece axis (C), and the workpiece (200) rotates about the workpiece axis (C) in machining engagement with the worm wheel (100). Each workpiece (200) has a workpiece tooth (201) with a workpiece tooth surface (210), wherein the workpiece tooth surface (210) is machined by means of at least one worm helical tooth surface (101) of the worm grinding wheel (100) during each grinding stroke. There are contact areas (104, 104') on the worm helical tooth surface (101), in which the worm helical tooth surface (101) contacts the workpiece tooth surface (210) during the relevant grinding stroke. Each workpiece tooth surface (210) has an effective region (211), and the contact region (104, 104') has a quality-determining region (105) that contacts the effective region (211) of the workpiece tooth surface (210). Its features are, Multiple decoupled contact areas (104, 104') exist on the worm helical tooth surface (101), and the quality-determining region (105) of the decoupled contact areas is substantially not overlapped with all other contact areas (104, 104') of the worm helical tooth surface (101), and multiple grinding strokes are performed separately by means of the decoupled contact areas before the worm grinding wheel (100) is re-dressed.

2. The method according to claim 1, wherein, prior to the worm wheel (100) being re-dressed, a predetermined number of grinding strokes are performed by means of each of the decoupled contact areas.

3. The method according to claim 2, wherein the predetermined quantity is at least 3, preferably at least 5, and especially at least 10.

4. The method according to claim 1, For each of the decoupled contact areas, a wear characteristic parameter (I) is determined to characterize the wear state of the worm grinding wheel in the corresponding contact area (104, 104'). r ;I f ),and Wherein according to the wear characteristic parameter (I) r ;I f It automatically makes a decision: for subsequent grinding strokes, whether to re-engage in the same contact area (104, 104') or whether the contact area (104, 104') will no longer be used for further grinding strokes until the worm wheel is dressed.

5. The method according to claim 4, The wear characteristic parameter (I) r ;I f () is a strength measure used for grinding strength. In particular, the maximum value, integral, average value, or low-pass filtered spectral component of the power index (P(t)) used for instantaneous grinding power.

6. The method according to claim 4 or 5, wherein the wear characteristic parameter (I) is used to determine the wear characteristic parameter (I) r ;I f ) or parameters and thresholds derived therefrom (I r,thr ;I f,thr The decision was made based on a comparison of the above.

7. The method according to any one of the preceding claims, wherein a plurality of workpieces (200) are machined by means of each of the decoupled contact areas (104, 104') before the worm wheel (100) is re-dressed.

8. The method of claim 7, wherein only one selected grinding stroke, particularly a roughing stroke or a finishing stroke, is performed for each workpiece by means of each decoupled contact area (104, 104').

9. The method according to any one of the preceding claims, wherein all contact areas (104, 104') used on the worm wheel (200) are arranged relative to each other such that the quality-determining regions of all contact areas substantially do not overlap with all other contact areas (104, 104').

10. The method according to any one of the preceding claims, wherein for each workpiece (200), during a grinding stroke of the associated workpiece (200) by means of one of the decoupled contact areas (104, 104'), the worm wheel (100) moves continuously relative to the workpiece (200) not only along an axial feed direction (Z) having a component parallel to the workpiece axis (C) but also along a traverse direction (Y), such that during the associated grinding stroke, during the machining engagement, the worm wheel (200) generally travels a traverse distance along the traverse direction (Y) and a travel distance along the axial feed direction (Z) relative to the workpiece (100).

11. The method according to any one of the preceding claims, wherein the worm grinding wheel does not have the modification of the worm helical tooth surface (101) in at least a portion of the decoupled contact area (104, 104'), the modification being different along different contact paths within the associated contact area.

12. The method according to any one of the preceding claims, The worm wheel occupies the first cutting position at the end of the machining engagement during the first grinding stroke. The worm wheel occupies the second traverse position at the start of the machining engagement in the subsequent second grinding stroke. The following distance exists between the first cutting tool position and the second cutting tool position: The spacing corresponds at least to the width of the contact path on the worm grinding wheel tooth surface (101) measured along the worm axis (B), and The worm wheel (100) occupies only the traverse position (Y) outside the area between the first traverse position and the second traverse position during all grinding strokes.

13. The method according to claim 12, Wherein the distance between the first cutting tool position and the second cutting tool position ( The following relationship must be satisfied: ， Where a p,0 The width of the contact path is measured perpendicular to the contact path extension, and α represents the engagement angle of the worm wheel (100) in the axial section.

14. The method according to any one of the preceding claims, wherein during machining using the decoupled contact area (104, 104'), a power index (P(t)) characterizing instantaneous grinding power is measured, and at least one grinding parameter is closed-loop controlled based on the power index (P(t)).

15. A gear-making machine for processing pre-formed teeth on a workpiece (200) by continuous generating milling, the gear-making machine comprising: A tool spindle (15) is configured to clamp a worm wheel (100) thereon so as to drive the worm wheel (100) to rotate about the worm axis (B); Workpiece spindle (21), which is configured to clamp a pre-made toothed workpiece (23) thereon so as to drive the pre-made toothed workpiece (23) to rotate about the workpiece axis (C). An axial slider (13) is configured to generate an axial feed motion of the tool spindle (15) relative to the workpiece spindle (21) along an axial feed direction (Z), the axial feed direction having a component parallel to the workpiece axis (C). The tool slide block is configured to generate a tool movement relative to the workpiece spindle (21) in a tool movement direction (Y) that extends parallel to the worm axis (B); Adjustment device (30); and Machine tool control device (40). Its features are, The machine tool control device (40) constitutes a method for implementing any one of the preceding claims.

Images