Method for detecting wear of a tool

Abstract

The invention relates to a method for detecting wear of a tool (2) during the machining of a workpiece (7) in a machine tool (1), comprising the following steps: during a starting phase at the beginning of the machining using a new tool (2), capturing at least one machining parameter (B1) over a predefined starting period (T1); forming a starting average value (D1) of the machining parameter (B1); during the further machining of the workpiece (7) after the starting time period (T1), comparing each current value (D2) of the machining parameter (B1) with the starting average value (D1); if the current value (D2) of the machining parameter (B1) lies within a predefined interval (I) around the starting average value (D1), determining that the tool (2) is in order and continuing the machining using the tool (2), and if the the current value (D2) of the machining parameter (B1) lies outside the predefined interval (I) around the starting average value (D1), determining that the tool (2) is worn and replacing the tool (2) or terminating the machining.

Description

[0001] HOEFER & PARTNER

[0002] ROE250801 PCT

[0003] August 26, 2025

[0004] Applicant:

[0005] P&L GmbH & Co. KG Scheibenstrasse 6 29614 Soltau

[0006] Method for detecting tool wear

[0007] Description

[0008] The present invention relates to a method for detecting the wear of a tool during the machining of a workpiece in a machine tool and to a machine tool that is set up to carry out the method.

[0009] When machining workpieces in a machine tool, the tool used is subject to wear over time. Tool wear is often determined by an experienced machine operator who, upon removing the tool, observes irregularities on its surface, such as chipping on the cutting edges, scoring, or similar defects. The operator then determines that the tool is worn and needs to be replaced. In automated machining processes, tool replacement is frequently performed after a predetermined number of machining operations, exceeding a specified maximum machining time, or reaching a predetermined tool life.The risk here is that a tool is replaced too late, resulting in a large number of defective parts being produced, or that a tool breaks, or that a tool is replaced too early, even though it could still have been used for a certain number of machining operations.

[0010] The object of the present invention is to provide an automated method for detecting the wear of a tool during the machining of a workpiece in a machine tool, which enables a reliable determination of tool wear.

[0011] This problem is solved by a method with the features of claim 1 and a machine tool with the features of claim 22. The dependent claims describe preferred embodiments of the invention. The inventive method for detecting wear of a tool with the features of claim 1 has the advantage over the prior art that reliable, automated wear detection of the tool is possible. Thus, a tool, particularly in the automated machining of workpieces, can be used precisely long enough to prevent the production of defective parts, tool breakage, and premature tool changes. This allows the tool to be used for machining workpieces over an optimal service life. This results in significant cost advantages, as it optimizes tool utilization and prevents tools from being replaced too early or too late.

[0012] According to the invention, this is achieved by the method, as a first step during a start-up phase at the beginning of machining with a new tool that does not yet show any wear, recording at least one machining parameter B1 over a predetermined start period T1. Subsequently, a starting average value D1 of the machining parameter B1 is calculated. Calculating the starting average value prevents outlier values ​​of the machining parameter from distorting the tool's condition at the start of machining with the new tool. In particular, if the machining parameter B1 is recorded with very high temporal resolution, for example, every 10 milliseconds, the machining parameter B1 can fluctuate considerably due to the machining process and the varying material removal rates of the tool on the workpiece, i.e., it can oscillate between maximum and minimum values.When calculating the initial average value D1, the fluctuation of the machining parameter B1 is preferably averaged out using a sufficient number of values ​​during the initial period T1, and a suitable average value is calculated. In a subsequent step, during further machining of the workpiece after the initial period T1, the current value D2 of the machining parameter B1 is compared with the initial average value D1. If the current value D2 of the machining parameter B1 lies within a predetermined interval around the initial average value D1, it is determined that the tool is in good working order and machining continues with the tool. If the current value D2 lies outside the predetermined interval around the initial average value D1, it is determined that the tool is worn and either the tool is replaced or machining is aborted.

[0013] The predetermined interval around the initial average value D1 is preferably determined from historical values. The predetermined interval is preferably symmetrical around the initial average value D1 and is particularly ±40 percent of the initial average value D1. The machining parameter B1 can preferably be a total machining force F1 of the tool on the workpiece, or a radial force FR1 acting radially on the tool, or an axial force FA1 acting axially on the tool.

[0014] The machining parameter B1 is preferably a vibration of the tool, which in particular represents the vibration of the tool during machining and / or a torque M1 of a main spindle of the machine tool and / or a structure-borne sound K1 resulting from machining and / or a sound SA1 resulting from machining, which is detected in the working space, in particular by means of a microphone.

[0015] Preferably, several different machining parameters B1 are monitored in parallel. In particular, if at least one of the multitude of monitored machining parameters B1 lies outside the corresponding interval, it is determined that a tool change or a machining abort is necessary.

[0016] The method is preferably implemented such that the current value D2 is an average of several successive values ​​of this processing parameter B1. Particularly preferably, the current value is a moving average of several successive values ​​of the processing parameter B1. A moving average is understood to be a value that comprises several values, whereby when a new value is added, the oldest value in the series is discarded. Preferably, the moving average comprises at least ten previous values ​​of the current value and the current value D2. With a very fine temporal resolution of the processing parameter B1, the moving average is preferably calculated over a larger number of values, for example, 100 or 1000, of the processing parameter B1 in order to eliminate the fluctuating behavior of the processing parameter B1.When averaging the current value D2 with so many values, the predetermined interval I for wear detection can be chosen much narrower, for example with + / - 10% of the initial average value D1.

[0017] Furthermore, when calculating the average for the current value D2 and / or the initial average value D1, a value that exceeds or falls below a predetermined threshold is ignored for the averaging process. This prevents outliers in the values ​​of the processing parameter from distorting the average.

[0018] Preferably, a waiting phase W precedes the start phase, in which the start average value D1 is determined. This ensures that distorting values, which are particularly relevant at the start of machining when the tool first comes into contact with the workpiece and the cutting edges are still very sharp, are disregarded and do not distort the start average value. The duration of the waiting phase W can be determined from historical data and empirical values ​​and is preferably hard-coded into the control system of a machine tool.

[0019] A further refinement of the method according to the invention can be achieved if, when comparing the current value D2 of the machining parameter B1 with the initial average value D1, this comparison is only performed if the tool speed and / or feed rate remains constant for a predetermined period, in particular 3 seconds. This also ensures that individual values ​​do not distort the comparison result and that the tool is not replaced prematurely, even though it is not yet fully worn. Similarly, it can be specified that the initial average value D1 is only calculated if the tool speed and / or feed rate is constant.

[0020] Furthermore, when machining with a changing feed rate, a correction calculation is preferably performed for the current value D2. Thus, at lower feed rates than in the start-up phase, the value of the current value D2 can be increased before comparison with the start-up average value D1, and at higher feed rates than in the start-up phase, the value of the current value D2 can be decreased before comparison with the start-up average value D1.

[0021] Furthermore, when calculating an average value for the current value D2, the values ​​for the machining parameter B1 included in the average calculation are preferably increased or decreased depending on the feed rate.

[0022] Preferably, the tool is reused once the machining of a workpiece is complete and no wear has yet been detected. Since the tool is therefore not yet worn, it can be used to machine the next workpiece. For this purpose, the starting average value D1 of the first machining operation on the first workpiece is preferably used to compare the current value D2 of the machining parameter B1 with this tool.

[0023] Preferably, the machining parameter B1 is only determined when the tool is engaged with the workpiece. This ensures that idle movements, for example during translation movements, are detected, especially since vibrations, which are recorded as machining parameters, correspond to an idle run of the main spindle of the machine tool and / or the machining forces are zero and / or the torque of the main spindle is minimal, corresponding to an idle run.

[0024] Preferably, after replacing a worn tool with a sister tool, the starting average value D1 of the worn tool is used. This can, in particular, reduce the computational load of the machine tool. Alternatively, a new starting average value D1 is determined for the sister tool. Preferably, the starting average value of the worn tool is then compared with the starting average value of the new sister tool. If a deviation above or below a predetermined threshold is detected, the sister tool is replaced or the machining operation is aborted. This can be understood as a precautionary measure to avoid damage to the next workpiece to be machined and / or the newly inserted sister tool.

[0025] Particularly preferred is the storage and reuse of starting average values ​​for certain tool types, for example milling cutters with different tooth configurations and / or different diameters, for different machining operations, so that when machining with a specific tool type, the determination of the starting average values ​​D1 can be omitted.

[0026] Preferably, a machining simulation is performed in parallel with the machining of the workpiece. In this simulation, the engagement conditions for the tool are calculated for each position of the tool during machining, and the determination of the initial average value D1 and the comparison of the current value D2 with the initial average value D1 are only carried out if the engagement conditions lie within a predefined interval.

[0027] Preferably, the engagement conditions calculated in the simulation for each position of the tool along a toolpath correspond either to a material volume that is removed while the tool travels a defined path increment along a toolpath and / or to a plunge depth AZ, which is calculated from a difference between a highest and a lowest point of contact of the tool with the workpiece material during machining at the respective position and / or to a wrap-around, which indicates the angular range over which the cutting edge of the tool is in engagement with the workpiece material during one revolution of the tool and / or to the size of a contact area A, which can be calculated from a tool envelope created by rotation of the tool and the workpiece material with which this envelope is in contact during machining at the respective tool position.

[0028] Preferably, several intervals are created for different engagement conditions calculated in the simulation, for each of which initial average values ​​D1 and current values ​​D2 are determined and evaluated separately. Preferably, the initial average values ​​D1 for a given tool type are stored for different intervals of engagement conditions, so that the initial average values ​​D1 for these intervals do not need to be recalculated when machining with a new tool of the same tool type.

[0029] It is preferable to calculate the initial average values ​​D1 for different intervals of engagement conditions from initial average values ​​D1 already available for other intervals of engagement conditions, particularly by interpolation. Thus, it is possible, for example, for a tool type, to calculate the initial average value D1 for an interval for that tool type from already known initial average values ​​D1 for other intervals.

[0030] Furthermore, it is preferably also preferred that a feed rate and / or a rotational speed of the tool be taken into account when simulating the engagement conditions.

[0031] Preferably, the engagement intervals are determined based on two parameters typical of engagement conditions, such as the insertion depth AZ and the contact area A, which is calculated from the tool's outer body with which it is in contact with the workpiece. This creates a two-dimensional array of engagement intervals based on two parameters. Each interval is defined by a range for a first parameter and a range for a second parameter of the engagement conditions. If both parameters lie within an interval, the initial average value D1 or the current value D2 for that interval is determined and applied for wear control. Of course, more than two parameters can also be used for the engagement conditions, resulting in a multi-dimensional array of intervals.

[0032] It is also preferable to exclude individual intervals for the engagement conditions from wear monitoring, for example if it turns out that certain engagement conditions are not so suitable for wear monitoring as described above, for example because the course of the machining parameter B1 or the machining parameter B1 with increasing wear is not so clear.

[0033] Alternatively, the engagement conditions occurring during machining can be calculated in a simulation prior to machining, for example, in a CAD / CAM system, and thus made available for the process. To ensure that the continuously changing engagement conditions during machining are known for each tool position in the control system, they can be stored in a suitable format within the machining program, which contains the toolpaths for material removal on the machine, assigned to the individual tool positions during machining. Thus, the machining program contains not only the coordinates for controlling the machine during the material removal process, but also the corresponding engagement conditions for each tool position during machining.Thus, the machine control system can use the path information from the machining program, such as X, Y, and Z coordinates for the tool movement relative to the workpiece, to control the machine and achieve the desired material removal rate. Simultaneously, the machine control system uses the information about the engagement conditions from the machining program for each tool position to perform wear detection as described previously. With this approach, it is no longer necessary to calculate the simulation in parallel with the machining process within the control system.

[0034] The invention further relates to a machine tool configured to carry out a method according to the invention. The machine tool preferably comprises a control unit, in particular with a memory in which program sequences according to the method according to the invention are stored. Preferably, the machine tool has an input panel with which, in particular, threshold values ​​can also be entered manually.

[0035] The machine tool is preferably a milling machine, in particular with an automatic tool changing system in which new sister tools and / or different tool types are stored and an automated tool change can be carried out.

[0036] The invention is described in detail below with reference to the accompanying drawing. The drawing shows:

[0037] Fig. 1 shows a schematic, perspective view of a machine tool for carrying out the method according to the invention.

[0038] Fig. 2 shows a curve of the processing parameter B1 over a few seconds, in which, due to the fine temporal resolution of the acquisition of the processing parameter B1, a strongly fluctuating curve for the processing parameter B1 is recorded.

[0039] Fig. 3 shows a schematic diagram illustrating the radial force FR1 occurring on the tool during machining of a workpiece over time t and a determination of the current value D2 using averaging.

[0040] Fig. 4 is a schematic diagram showing the radial force FR1 occurring on the tool during machining of a workpiece over time t and a determination of the current value D2 without averaging. Fig. 5 is a schematic diagram showing vibrations S1 of the tool during machining over time t.

[0041] Fig. 6 is a schematic diagram showing the torque M1 of a main spindle during machining over time t.

[0042] Fig. 7 shows a schematic sequence of the method according to the invention,

[0043] Fig. 8 shows a schematic representation for the assignment of the radial force FR1 to different intervals for the engagement conditions, here the contact area A between tool and workpiece with an interval for the starting average values ​​D1 ,

[0044] Fig. 9 shows a schematic representation for the assignment of the radial force FR1 to different intervals for the engagement conditions, here the contact area A between tool and workpiece with several intervals for the star average values ​​D1 ,

[0045] Fig. 10 shows a schematic representation for the assignment of the radial force FR1 to different intervals for the engagement conditions, wherein the intervals depend on two parameters for the engagement conditions and the initial average values ​​D1 were determined only for one interval, and

[0046] Fig. 11 shows a schematic representation for the assignment of the radial force FR1 to different intervals for the engagement conditions, where the intervals depend on two parameters for the engagement conditions and the starting average values ​​D1 were determined for several intervals.

[0047] Below, with reference to Figures 1 to 11, a machine tool 1, in particular a milling machine, and a method according to the invention for wear detection are described in detail.

[0048] Fig. 1 schematically shows the machine tool 1 for carrying out the method according to the invention. The machine tool 1 is a milling machine and comprises a main spindle 3 on which a tool 2 is clamped. The reference numeral 10 denotes a control unit of the machine tool.

[0049] Furthermore, the machine tool 1 has a multitude of sensors, of which only a selection is shown schematically in Fig. 1. The machine tool 1 includes, in particular, a structure-borne sound sensor 4 for detecting structure-borne sound K1 and a torque sensor 5 for detecting a torque M1 of the main spindle 3. Alternatively, the torque of the main spindle 3 can also be calculated from the motor current of the main spindle 3. Furthermore, a vibration sensor 6 for detecting vibrations S1 of the tool 2 is arranged on the main spindle 3.

[0050] Reference symbol 8 designates a microphone that detects sound SA1 in the workspace.

[0051] Furthermore, a large number of force sensors are arranged, particularly on the main spindle, to detect the total machining force F1 of the tool 2 on a workpiece 7, as well as a radial force FR1 acting radially on the tool and an axial force FA1 acting axially on the tool. For clarity, the force sensors are not shown in Fig. 1. Alternatively, the machining forces (total force, axial forces, and radial forces) can also be calculated from the feed forces in the machine axes, as described, for example, in DE 10 2021 132 300.

[0052] Thus, the machine tool can detect various machining parameters B1 during the machining of a workpiece 7. All sensors are connected to the control unit 10 of the machine tool 1.

[0053] Fig. 2 shows the curve of the machining parameter B1, in this case the radial force FR1 in Newtons, during the machining of a workpiece 7 with a tool 2 over a time t of approximately 30 seconds. The radial force FR1 is recorded with a very fine temporal resolution. The fluctuating behavior of the radial force FR1 is evident. Due to the machining process, the radial force is sometimes in the range of 400 N to 450 N and sometimes, for example during small, slower feed movements, in the range of 25 N to approximately 250 N. The curve of the current value D2, calculated here as a moving average, is also shown as a lighter line. Despite the fluctuations in the radial force FR1, the current value D2 shows a relatively smooth curve.

[0054] Fig. 3 schematically shows a diagram illustrating the radial force FR1 during the same machining operation of workpiece 7 with tool 2 in Newtons over time t in seconds as in Fig. 2, but over a significantly longer period of approximately 8000 seconds. The inventive method is schematically explained using the radial force FR1 as an example. The inventive method essentially proceeds in the same way for the other machining parameters B1.

[0055] The method according to the invention is explained in detail below with reference to Figures 3 and 6. It is assumed that a new, wear-free tool 2 is used, with wear monitoring being carried out on the basis of the radial force FR1.

[0056] At the start (step S1 in Fig. 6), the tool initially travels several idle strokes or undergoes a warm-up phase before engaging the workpiece. This phase is labeled L as the idle phase in Fig. 3. The radial force FR1 is zero. At time tO, the tool engages the workpiece, and the radial force increases. Since the radial force FR1 between the tool and workpiece is significantly lower immediately after the start of machining, a waiting phase W is inserted in step S2. The waiting phase W extends from time tO (the start of material removal from the workpiece) until time t1. The length of the waiting phase W is preferably determined based on historical data for the tool type currently in use. Since Fig. 3 depicts a relatively long period of approximately 8000 seconds, and as shown in Fig.As shown in Figure 2, where the radial forces FR1 fluctuate relatively strongly, the radial forces FR1 in Figure 3 look like a dark area where the maximum values ​​of the fluctuations of the radial force FR1 form the upper limit of the area and the minimum values ​​of the fluctuations of the radial force FR1 form the lower limit (near the zero line).

[0057] As can be seen in Fig. 3, the waiting phase W is followed by the predetermined start period T1. The start period T1 covers the time from t1 to t2, which in the example of Fig. 2 can be, for example, 800 s long. During the start period, a large number of individual values ​​are recorded (step S2), in this example the radial force FR1.

[0058] After the initial period T1 has elapsed, an initial average value D1 of this parameter, i.e., the radial force FR1, is calculated (step S3). The initial average value D1 for the radial force FR1 naturally lies in the middle of the range between the maximum and minimum values ​​of the radial force FR1, in this case rather towards the upper end, since values ​​for the radial force close to the maximum values ​​occur more frequently than values ​​close to the minimum values, as can also be seen in Fig. 2.

[0059] In step S4, a current value D2 of this machining parameter B1, i.e., the radial force FR1, is recorded as a moving average, similar to Fig. 2. Approximately 7000 values ​​of the radial force FR1 are considered for calculating the moving average for the current value D2 in order to smooth out the significant fluctuations of the radial force FR1. The resulting current value D2 is shown as a light line in the diagram in Fig. 3.

[0060] In step S5, a comparison is made between the current value D2 of the machining parameter B1 and the starting average value D1. If the comparison in step S5 shows that the current value D2 lies within a predetermined interval I around the starting average value D1, it is determined that the tool is OK (J in Fig. 5) and the procedure returns to step S4 and the next current value D2 is determined.

[0061] Figure 3 schematically depicts an interval I around the average value D1. If, in step S5, it is determined that the current value D2 of the machining parameter B1 lies outside the predetermined interval I around the starting average value D1, it is determined that the tool is worn (N in Figure 5). The process then proceeds to step S6, in which the machining of the workpiece is stopped.

[0062] In step S7, the tool is automatically exchanged using an automatic tool exchange system not shown in Fig. 1.

[0063] For further processing, an identical new sister tool is then used. Preferably, the machining process continues immediately, with the value of the worn predecessor tool being used as the starting average value D1 for the sister tool. Alternatively, the described start-up phase with the determination of a new starting average value D1 can also be performed for the sister tool. Then, preferably, a comparison is also carried out between the starting average value of the sister tool and the starting average value of the previous worn tool. If the deviation exceeds a predetermined value, it is determined that the sister tool is defective and machining of the workpiece with the sister tool is discontinued. If necessary, the sister tool is then replaced with a second new sister tool.

[0064] Figure 3 schematically illustrates the waiting phase W and the predetermined start period T1. Over a further period T2, extending from point t2 to point t3, essentially constant current values ​​D2 are determined as a moving average of the radial forces FR1. This means that during period T2, the tool is not worn and machining continues uninterrupted. A predetermined interval is maintained around the starting average value D1, so that minor deviations from the starting average value do not immediately indicate tool wear.

[0065] At time t3, the radial forces FR1 increase sharply. This causes the current value D2 to lie outside the interval around the initial average value D1. The machine tool's control unit 10 then determines that the currently used tool is worn and immediately stops machining to prevent damage to the workpiece. The worn tool is then replaced with a new, identical tool.

[0066] Figure 3 illustrates the short period T3, showing that the radial forces FR1 increase sharply with increasing time t. This would lead to significant damage to the workpiece if the tool is not changed. Figure 4 shows a diagram for the same machining operation as in Figure 3, with a time course over approximately 8000 seconds and the radial force FR1 curve. After the idle phase L and the waiting phase W, the starting average value D1 is determined during the start-up phase T1. Unlike Figure 3, Figure 4 does not calculate a moving average for the current value D2, but uses the radial force FR1 values ​​directly. To ensure wear detection, the predetermined interval I around the starting value D1 is significantly larger. At time t3, the maximum values ​​of the radial force FR1 exceed the predetermined interval I, and the tool is recognized as worn.A disadvantage of this approach without calculating a moving average for D2 is that individual outliers in the radial force FR1, e.g., during machining, can lead to premature wear detection, even though the tool is not yet worn. This can also be countered by ignoring individual outliers.

[0067] Figure 5 schematically depicts the vibrations S1 of the tool during the machining process over time t. As can be seen in Figure 5, the procedure is fundamentally the same as for the radial force FR1 when using the vibrations S1 as the initial average values ​​D1 and the current values ​​D2. However, as shown in Figure 5, a significantly larger number of outliers are visible in the vibrations S1, with one outlier marked with the reference symbol S'. Therefore, when using the vibrations S1 as the machining parameter B1, it is even more advantageous to calculate the current values ​​D2 of the vibrations as an average over several values ​​for the vibrations S1. This prevents such outliers from creating a false impression of the actual wear situation of the tool.A moving average is particularly preferred here, in which, when a new current value D2 is recorded, the oldest value of the vibrations S1 is replaced by the newest value of the vibrations S1 during the averaging process. Figure 5 also shows that the main spindle was only switched on after the idle phase L, since the vibrations are zero beforehand. After switching on the main spindle, the minimum values ​​for the vibrations S1 are approximately 10 m / s. 2 , which corresponds to the idle vibrations of the main spindle. The maximum vibration values ​​fluctuate considerably during period T2, between approximately 40 m / s². 2 and 50 m / s 2 . When using the vibrations S1 as a machining parameter for wear detection, it is advantageous to ignore outliers such as S' or to use a moving average when calculating the current value D2.

[0068] Fig. 6 shows another schematic diagram of a machining parameter B1. In Fig. 6, the machining parameter B1 is a torque M1 of the main spindle 3 of the machine tool 1. The torque M1 is again plotted against time t. As in the diagrams of Figs. 3 to 5, an idle phase L, a waiting phase W, a start period T1, a continuous machining period T2, and a period T3, before which the machining of the workpiece should actually be stopped to avoid damage to the workpiece and / or the machine tool, are shown. As can be seen in Fig.As can be seen in Figure 6, the torque of the main spindle 3 is very well suited for inferring tool wear during this machining operation, since the current values ​​D2 during the machining period T2 show only minimal changes and only small outliers, both when calculated as a moving average of the torque values ​​M1 and when the individual torque values ​​M1 are used as current values ​​D2 without averaging. It can be seen that the minimum values ​​for torque M1 form a lower limit of the measured torque range, corresponding to the no-load torque of the main spindle 3. The maximum values ​​for torque M1 occur during the tool's engagement with the workpiece material and form the upper limit of the measured torque range during machining.

[0069] The method can therefore be carried out by selecting a single machining parameter B1 and monitoring it as described above, or alternatively, several, in particular all, detectable machining parameters B1 can be used for wear monitoring. With multiple machining parameters, wear can be determined if only one of the multitude of machining parameters B1 lies outside an interval I defined around that machining parameter.

[0070] Fig. 8 shows the procedure when the engagement conditions between tool and workpiece are considered for wear detection. In this case, the engagement parameter is the size of a surface (contact area A), which can be calculated from a tool envelope created by the tool's rotation and the workpiece material with which this envelope is in contact at the respective tool position during machining. The range for a surface area (contact area A) between 4.5 mm 2 and 5.0 mm 2 A starting average value D1 for the radial force of 500 N was calculated. Starting average values ​​D1 for other intervals of engagement conditions, specifically the contact area A between the tool and the workpiece, have not yet been determined. If, during machining, engagement conditions with a contact area A between 4.5 mm² occur again... 2 and 5.0 mm 2If current values ​​D2 occur, they can be compared with the initial average value D1 for that interval. If engagement conditions with larger or smaller contact areas A occur, the initial average values ​​D1 for the intervals in which the engagement conditions fall must first be determined.

[0071] Fig. 9 shows the same machining process as in Fig. 8 in a more advanced stage. Meanwhile, starting average values ​​D1 for the radial force FR1 have also been determined for other intervals of engagement conditions (size of the contact area A), so that if engagement conditions occur again that fall within these intervals, the current values ​​D2 can also be compared with the respective starting average value D1. This can be seen for the interval for the size of the contact area A of 5.0 mm. 2 up to 5.5 mm 2No initial average value D1 could yet be determined. According to the invention, this is possible by interpolation with the two adjacent intervals (interval 4.5 mm). 2 up to 5.0 mm 2 and interval 5.5 mm 2 up to 6.0 mm 2 ) by interpolation a starting average value D1 for the interval of 5.0 mm 2 up to 5.5 mm 2 to calculate. It is understood that in this example the width, i.e., the range of values, of the intervals was chosen arbitrarily to illustrate the procedure. In practice, the intervals can be chosen to be considerably narrower, i.e., with a smaller range of values, so that considerably more intervals are created.

[0072] Fig. 10 shows the method when two parameters are considered for the engagement conditions. In this case, these are the size of a surface (contact area A), which can be calculated from the tool's envelope created by rotation and the workpiece material with which this envelope is in contact at the respective tool position during machining, and the plunge depth AZ, which is calculated from the difference between the highest and lowest points of contact between the tool and the workpiece material during machining at the respective position. This results in a two-dimensional array of intervals, and each interval is assigned a range of values ​​for the first parameter, the size of the contact area A, and a range of values ​​for the second parameter, the plunge depth AZ. The average star value D1 could only be determined for one interval up to the machining time represented by the diagram.

[0073] Figure 11 shows a diagram for the same machining process as Figure 10, but at a later stage during the machining process. Meanwhile, initial average values ​​D1 were calculated for several intervals of contact area A and immersion depth AZ. The bars in the diagram are narrower for clarity only. According to the invention, the intervals are directly adjacent to one another, analogous to Figures 8 and 9. A slightly diagonal pattern of the intervals with calculated values ​​for the initial average values ​​D1 is evident; that is, as the immersion depth AZ increases, so does the contact area A. Similar to Figures 8 and 9, the number of intervals can be considerably higher, and thus the range of values ​​for the individual intervals can be considerably smaller; it has only been kept small for clarity.The comparison of the current values ​​D2 with the respective starting average value D1 is carried out for all intervals for which calculated starting average values ​​D1 were calculated, provided that during further processing the values ​​for immersion depth AZ and contact area A fall back into such an interval.

[0074] Thus, a tool 2 can be used optimally until actual wear occurs and a tool replacement becomes necessary. In particular, time-based or tool life estimates are no longer required to determine a tool's service life, as these methods always carry the risk of replacing the tool too early or too late.

[0075] In addition to the above written description of the invention, explicit reference is hereby made to the graphic representation of the invention in the figures for its supplementary disclosure.

[0076] Reference symbol list

[0077] 1 machine tool

[0078] 2 tools

[0079] 3 Main spindle

[0080] 4 structure-borne sound sensor

[0081] 5 Torque sensor

[0082] 6 vibration sensor

[0083] 7 workpiece

[0084] 8 microphones

[0085] 10 Control

[0086] A touch surface

[0087] B1 Processing parameters

[0088] D1 starting average value

[0089] D2 current value of the processing parameter

[0090] F1 Total machining force

[0091] FA1 Axial force

[0092] FR1 radial force

[0093] I interval

[0094] K1 Structure-borne sound

[0095] L Idle phase

[0096] M1 Main spindle torque

[0097] S1 Vibration of the tool

[0098] SA1 sound

[0099] T1 Start period

[0100] T2 processing time

[0101] T3 Period in which work is carried out with worn tools

[0102] Waiting period

[0103] AZ Immersion depth

Claims

Claims 1. Method for detecting wear of a tool (2) during the machining of a workpiece (7) in a machine tool (1), comprising the steps: - during a start-up phase at the beginning of processing with a new tool (2) recording at least one processing parameter B1 over a predetermined start period T1 , - Calculating a starting average value D1 of the processing parameter B1 , - during further processing of the workpiece (7) after the start period T 1, compare a current value D2 of the processing parameter B1 with the start average value D1 , - wherein, if the current value D2 of the machining parameter B1 is within a predetermined interval I around the starting average value D1, determine that the tool (2) is OK and continue machining with the tool (2), and - if the current value D2 of the machining parameter B1 is outside the predetermined interval I around the starting average value D1, determine that the tool (2) is worn and replace the tool (2) or abort the machining.

2. Method according to claim 1, wherein the machining parameter B1 is a total machining force F1 of the tool (2) on the workpiece (7) and / or a radial force FR1 acting radially on the tool (2), and / or an axial force FA1 acting axially on the tool (2).

3. Method according to one of the preceding claims, wherein the machining parameter B1 is a vibration S1 of the tool (2) and / or a torque M1 of a main spindle (3) of the machine tool (1) and / or a structure-borne sound K1 resulting from the machining and / or a sound SA1 resulting from the machining which is detected in the working space.

4. Method according to one of the preceding claims, wherein the current value D2 is an average of several successive values ​​of the processing parameter B1.

5. Method according to claim 4, wherein the current value D2 is a moving average of several successive values ​​of the processing parameter B1 with the current value and in particular at least two previous values ​​of the current value.

6. Method according to claim 4 or 5, wherein, when forming the mean value for the current value D2 or the mean value for the starting average value D1, values ​​of the processing parameter B1 that exceed or fall below a predetermined threshold are ignored for the purpose of averaging.

7. Method according to one of the preceding claims, wherein a waiting phase W is placed before the start period T 1 and / or, in particular, an idle phase L is placed before the waiting phase W.

8. Method according to one of the preceding claims, wherein the comparison of the current value D2 of the machining parameter B1 with the starting average value D1 is only performed if the rotational speed of the tool (2) and / or the feed rate of the tool (2) is constant over a predetermined period, in particular 3s.

9. Method according to one of the preceding claims, wherein, in the case of machining with changing feed rates, a correction calculation is performed for the current value D2, such that, for feed rates smaller than in the start period T1, the value of the current value D2 is increased before comparison with the start average value D1, and for feed rates larger than in the start period T1, the value for the current value D2 is decreased before comparison with the start average value D1.

10. Method according to claim 9, wherein, when calculating the mean value for the current value D2, the values ​​for the machining parameter B1 included in the mean value calculation are increased or decreased depending on the feed rate.

11. Method according to one of the preceding claims, wherein the current values ​​D2 for several different machining parameters B1 are monitored simultaneously and if one of the current values ​​D2 for the monitored machining parameters B1 lies outside the predetermined interval I around the respective starting average value D1, determine that the tool (2) is worn and replace the tool (2) or terminate the machining.

12. Method according to one of the preceding claims, wherein, when the machining of the workpiece (7) is completed and no wear of the tool (2) has yet been detected, the tool (2) is used for further use in the machining of a next workpiece (7), wherein the starting average value D1 from the machining of the first workpiece (7) machined with the tool (2) is used to compare the respective current value D2 of the machining parameter B1.

13. Method according to one of the preceding claims, wherein the machining parameter B1 is determined only when the tool (2) is engaged with the workpiece (7).

14. Method according to one of the preceding claims, wherein, after replacing a tool (2) with a sister tool (2), the starting average value D1 of the worn tool (2) is used as the starting average value D1, or a new starting average value D1 is determined and, in particular, compared with the starting average value D1 of the worn tool (2), and if there is a deviation above a threshold value, the sister tool (2) is replaced or the machining is aborted.

15. Method according to one of the preceding claims, wherein the starting average values ​​D1 are stored for certain tool types and are reused for different machining operations, so that the determination of the starting average values ​​D1 can be omitted when machining with a certain tool type.

16. Method according to one of the preceding claims, wherein a simulation of the machining is performed in parallel with the machining of the workpiece (7) or a simulation of the machining is calculated before machining the workpiece (7) and the results of the simulation are stored in the machining program and assigned to the travel paths, wherein the engagement conditions for the tool (2) are calculated for each position of the tool (2) during machining and the determination of the starting average values ​​D1 and the comparison of the current value D2 with the starting average value D1 only takes place if the engagement conditions are within a predetermined interval.

17. Method according to claim 16, wherein the engagement conditions calculated in the simulation for each position of the tool (2) along a toolpath either correspond to a material volume which is removed while the tool (2) travels a defined path increment along the toolpath, and / or correspond to an immersion depth AZ, which is calculated from a difference between the highest and lowest point of contact of the tool (2) with the material of the workpiece (7) during machining at the respective position, and / or correspond to a wrap-around, which indicates over what angular range during one revolution of the tool (2) the cutting edge is in engagement with the material of the workpiece (7), and / or correspond to a size of a contact area A, which can be calculated from a shell of the tool (2) which is created by rotation of the tool (2), and the material of the workpiece (7) with which this shell is in contact during machining at the respective position of the tool (2).

18. Method according to claim 16 or 17, wherein several intervals are formed for different engagement conditions calculated in the simulation, for which starting average values ​​D1 and current values ​​D2 are determined and evaluated separately.

19. Method according to claim 18, wherein the starting average values ​​D1 for a tool type are stored for different intervals of engagement conditions, so that they do not have to be recalculated when machining with a new tool (2) of this tool type.

20. Method according to one of the preceding claims, wherein the starting average values ​​D1 for different intervals of engagement conditions are calculated from starting average values ​​D1 already available for other intervals of engagement conditions, in particular by interpolation.

21. Method according to one of the preceding claims, wherein the intervals for the engagement conditions are defined as a function of two or more parameters of the engagement conditions, for example the immersion depth AZ and the contact area A, such that a two- or more-dimensional field of intervals is created for the assignment of the initial average values ​​D1 and the current values ​​D2 to the respective intervals for the engagement conditions.

22. Method according to one of claims 16 to 21, wherein the feed rate and / or rotational speed are taken into account when simulating the engagement conditions.

23. Machine tool configured for carrying out a method according to one of the preceding claims.

Images