JPH0616981B2 - Tool loss detector - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial field of application) The present invention reliably detects an abnormality such as a tool defect without being affected by the surface shape of a workpiece, a change in cutting depth, a position of a tool, or the like. The present invention relates to a tool loss detection device that can perform.

(Prior Art) In a milling machine or the like, numerical control or copying control is performed for machining. In the automatic machine tools that perform these numerical control and copying control, in order to improve the processing accuracy of the work piece, this is detected when the tool is damaged,
It is necessary to stop the machine tool or issue an alarm by this detection signal. Many proposals are currently made for this purpose. Particularly in recent years, with the development of various detection and signal processing techniques, there has been an increasing interest in detection during milling, that is, in-process detection methods and devices.

Conventional tool loss detection focuses on the fact that cutting force, cutting temperature, vibration, spindle motor current, acoustic emission, etc. change due to tool loss, and by measuring and processing these signals, A method of indirectly detecting a tool loss is adopted.

However, particularly in the automatic cutting process of a die, the process is a single item process having a different shape each time, and moreover, it is necessary to perform a complicated curved surface process in which curved surfaces are continuously formed in multiple layers. Therefore, it is fundamentally impossible to adopt a method of memorizing the machining conditions and the cutting phenomenon last time, comparing them with the cutting state of this time, and extracting the abnormal phenomenon.

The present inventors have already detected the cutting load force and compared its output value with a steady load value having an upper limit value and a lower limit value set in advance to control the tool feed speed. On the other hand, the inventor has invented a tool loss detection device that compares the abnormal load reference value that is set to be larger than the steady load value and stops the machine tool when the abnormal load reference value is exceeded. Also,
Focusing on the unique torque waveform at the time of tool loss, it detects the differential value with respect to the rise and fall times of its output signal, and has a waveform operation circuit that has a pre-stored waveform operation circuit that determines whether or not it is unique Invented the device.

In these previously proposed inventions, attention is paid to the peculiar torque waveform generated when the tool is broken, and the torque waveform is faithfully
It is a device that detects a tool defect by quickly extracting it and comparing it with a preset reference value, and can operate with high accuracy, that is, without malfunction, and is a device excellent in performance. However, since it is necessary to use a high-level arithmetic circuit and a processing circuit, there is a problem that a high level of technical skill is required for circuit design at the time of manufacturing the device, and the cost becomes high.

(Object of the Invention) The present invention eliminates such obstacles and can be applied to various types of machine tools, and can respond to various cutting conditions that differ depending on the shape of a work piece. It is an object of the present invention to provide a tool loss detection device that is simple and inexpensive.

(Principle of the Invention) Various information is included in the load current detected from the spindle motor of the machine tool. The inventors of the present invention have achieved the present invention as a result of performing frequency analysis of the load current and examining the correspondence with the tool loss for the above purpose.
According to the examination result, when the milling is performed by the automatic cutting method, when the load current value is converted into torque and the torque fluctuation is seen, the inclination of the machined surface with respect to the reference plane installed on the workpiece, That is, when the tool is automatically fed, the amount of torque fluctuation that occurs when moving up and down an inclined surface, when moving a flat part horizontally, when biting into a groove, and the amount of cut change greatly depending on the dimensional accuracy of the rough material. It was found that the fluctuations caused by rough cutting, etc., were of relatively low frequency. On the other hand, it has been found that the torque fluctuation that occurs when the cutting edge separates the cutting chips from the work piece and when the tool is broken is at a relatively high frequency. Therefore, it is possible to detect the occurrence of a tool defect by detecting the torque fluctuation based on the motor load current value, extracting the signal component of the frequency band corresponding to the tool defect included in the detection signal, and monitoring this. I understood it.

The results of such an examination will be described by using mathematical expressions: T = T ₁ + T ₂ + T ₃ + T ₄ + T ₅ where T: measured value (original waveform output) T ₁ : torque due to inclination of the machined surface of the workpiece with respect to the reference surface variation T _2: the depth of cut (stock removal amount) variation T 3 due to a change _in: variation T 4 when the cutting edge to produce a _chip: sudden output levels of the abnormal phenomena of the tool defects such as T _5: rotation of the tool The resistance values T ₁ and T ₂ of the bearing and the like of the main shaft portion of the processing machine are low frequency fluctuation components, and T ₃ , T ₄ ,
T ₅ is a high-frequency fluctuation component, and T ₄ > T ₃ > T ₅ and
Since T ₅ is much smaller than T ₄ , it is a negligible value. When T ₄ = T− (T ₁ + T ₂ + T ₃ + T ₅ ) ≧ 0, it is possible to detect the tool loss. When the T _th and tool defect detection set value, _T 4 when the _{-T th} ≧ 0, or if _{T th = n (T 1 +} T 2 + T 3 + T 5), T 4 -n (T 1 + T 2 + T ₃ + T ₅ ) ≧ 0, and even when T− (n + 1) (T ₁ + T ₂ + T ₃ + T ₅ ) ≧ 0, the tool defect detection can be performed.

The above explanation is a phenomenon during automatic cutting in milling, but torque fluctuations can be similarly analyzed in turning and drilling, and at that time, the above-mentioned T ₁ (tilt with respect to the reference plane of the work piece is inclined). Torque fluctuation) is omitted,
It suffices to focus only on T ₂ , T ₃ , T ₄ , and T ₅ .

FIG. 1 shows the relationship between the cutting time and the load current value of the spindle motor converted into torque in the cutting machine.
In numerical control or copying control machining, the power or load current value of the spindle motor in the cutting process under normal conditions fluctuates and is unstable, and when a loss occurs, the increase shows a characteristic behavior. It is a thing. The waveform A in FIG. 1 shows this state, and the output waveform at the time of normal cutting of the part left of the point P of the arrow shows that the machined surface of the workpiece has various inclinations with respect to the reference surface. Also, due to non-uniform cutting depth and the like, the fluctuation is severe and unstable.

On the other hand, the arrow P in FIG. 1 indicates the state when the tool is missing. Here, the torque sharply rises or falls and the low frequency fluctuation level (waveform B in FIG. 1) is used as a reference. Since the peak value of waveform A temporarily rises rapidly,
It can be distinguished from the situation before that. When a tool is broken, the characteristic torque changes in this way because the broken part is sandwiched between the workpiece and the tool, or the torque is temporarily increased during the cutting process due to an overload condition. Is caused by a cutting resistance which sharply rises, and then sharply descends with the idle cutting which occurs in association therewith. Therefore, the waveform B is obtained by removing the high frequency component, that is, the torque components T ₃ , T ₄ and T ₅ described above, from the waveform A through an averaging circuit having a low-pass filter (for example, 2 Hz) built therein. Note that the waveform B is displayed with its level lowered for convenience of explanation and signal processing. This waveform B
Includes a variation due to the inclination of the machined surface of the workpiece with respect to the reference plane and a variation due to non-uniform cutting. By comparing the waveform B with the waveform A, the waveform A is Characteristic components such as tool defects included will clearly appear. The comparison result is shown as a waveform C as shown in FIG. 5 as an example. As is clear from the waveform C, the level of fluctuation caused by the generation of chips is lower than the level of fluctuation at the time of tool loss,
Therefore, only the tool loss can be detected by setting and comparing the reference that can detect only the variation due to the tool loss.

(Structure and Effect of the Invention) In order to achieve the above-mentioned object, the tool loss detection device of the present invention is made based on the logic of frequency analysis of torque as described above. From the load current, the fluctuation of the signal of relatively low frequency such as the fluctuation due to the inclination of the work surface of the work piece with respect to the reference surface, the fluctuation due to the change of the cutting depth, the fluctuation when the cutting edge produces chips, the tool loss Sudden changes in output due to abnormal phenomena such as
A tool load detection circuit that detects a tool load fluctuation signal that includes fluctuations in a signal with a relatively high frequency such as fluctuations due to the resistance of the processing machine spindle accompanying the rotation of the tool, and a relatively high frequency from the tool load fluctuation signal. An output circuit that extracts and outputs the fluctuation of the signal, and if the level value of the fluctuation of the signal of a relatively high frequency output from the output circuit exceeds the set value for tool loss detection, an error such as tool loss occurs. And a signal processing circuit having a comparison circuit for outputting a signal indicating

The present invention detects the tool load by detecting the power or load current of the spindle motor of the processing machine in this way, and wave the detection signal at a specific cutoff frequency, and the waved signal and a relatively high frequency. By comparing with the comparison signal for detecting the part with a large peak value in the fluctuation amount of, by extracting the fluctuation amount of the signal of relatively high frequency including the information of the characteristic waveform indicating the occurrence of tool loss , It is not affected by the fluctuation of the signal of the relatively low frequency which is the unstable fluctuation of the level. That is, it is possible to reliably detect the tool loss without being affected by the cutting conditions that differ according to the shape of the work piece, for example, the fluctuations that appear as a relatively low frequency component due to the changes in the tool feed direction and the cutting depth. . Furthermore, since the structure is simple, it can be manufactured in a small size, at low cost, and easily.

According to the first aspect of the present invention, the tool load detection circuit constantly detects the load by detecting the power or load current of the spindle motor of the processing machine. Further, the output circuit of the signal processing circuit includes an averaging circuit that extracts only a fluctuation component of a signal having a relatively low frequency from the tool load fluctuation signal, and a relatively low value that is extracted from the tool load fluctuation signal by the averaging circuit. And a subtraction circuit for subtracting the variation of the frequency signal and outputting the level value of the variation of the relatively high frequency signal. In this mode, since the output of the subtraction circuit is compared with the reference set value for tool loss detection, it is possible to accurately detect the tool loss by discriminating the variation due to the chip generation, and further, in the numerical control or the scanning control. It is possible to reliably detect the collision between the tool and the workpiece or the collision between the tool and the processing machine, that is, the occurrence of the abnormal torque, which is caused by the input of the tool trajectory.

According to the second aspect of the present invention, the output circuit of the signal processing circuit is configured by a high-pass filter that extracts only the fluctuation component of the signal having a relatively high frequency from the tool load fluctuation signal. In this aspect, since the signal processing circuit is composed of the high-pass filter and the comparison circuit, it is possible to detect the tool loss with a simple circuit. When noise with a high frequency is mixed, a bandpass filter that blocks the noise and other noise countermeasures are required.

According to a third aspect of the present invention, in the tool loss detection device of the first aspect, the cutting edge is integrated by integrating the output signal of the subtraction circuit between the subtraction circuit and the comparison circuit at a set integration time. An integrator circuit is provided for averaging fluctuations in a signal having a relatively high frequency when cutting chips are generated. This integration circuit averages the fluctuations due to the generation of chips among the fluctuations due to only the cutting edge phenomenon, so that the discriminability of the waveform portion at the time of tool loss is improved.

According to a fourth aspect of the present invention, in the tool loss detecting device according to the third aspect, the first and second outputs of the integrating circuit are alternately sampled and held between the integrating circuit and the comparing circuit. A sample and hold circuit and an integral value subtraction circuit that performs subtraction between adjacent sample values output from the first and second sample and hold circuits are provided. As a result, the subtraction is performed between the adjacent sample values of the output of the integrator circuit, so that it is possible to remove the averaging level value due to the chip generation that occurs steadily, and the discriminability of the waveform portion when the tool is missing It can be further improved.

(Examples) Various examples of the present invention will be described below with reference to the drawings.

FIG. 2 shows a basic configuration common to each embodiment. In FIG. 2, a work piece 1 is installed on a surface plate 2, and a tool 3 is provided on an upper part of the work piece 1 and a lower end (a cutting edge) thereof.
Is held by the tool holding part 4 of the processing machine in a state of being brought into contact with the surface (working surface) of the workpiece 1. The tool holder 4 is
It can be moved up and down, front and back, and left and right by the support mechanism. Then, while moving in this way, the tool 3
The workpiece 1 can be processed into a desired shape by rotating. The tool 3 is driven by the spindle motor of the processing machine, and the tool load (driving force) of the spindle motor, that is, the supplied power or load current is constantly detected by the tool load detection circuit 5. The output of the tool load detection circuit is guided to the signal processing circuit 6, and the signal corresponding to the tool loss included in the output of the tool load detection circuit 5 is extracted. The signal processing circuit 6 pays attention to the fact that the variation T ₄ generated in response to the tool loss of the driving force (torque) has a relatively high frequency and a high peak value from the average value level, It is configured to detect its presence.

FIG. 3 shows the configuration of the first embodiment of the present invention, in which the signal processing circuit 6 passes only the low frequency component of the output which is the output of the tool load detection circuit 5 detected from the spindle motor of the machine tool. Averaging circuit 8 for averaging and its averaging circuit 8
Is compared with the output of the tool load detection circuit 5, and a difference circuit 9 for amplifying and outputting the difference, a threshold value setting circuit 10 for setting a threshold value T _th , and an output of the subtraction circuit 9 are used. It comprises a comparison circuit 11 which compares the output of the threshold value setting circuit 10 and extracts a portion having an amplitude larger than the threshold value T _th . FIG. 4 shows the signal processing circuit 6 in FIG. 3 more specifically. An overcurrent prevention circuit 12 is added to this circuit. The overcurrent prevention circuit 12 connects a pair of diodes 122 and 123 connected in parallel in opposite directions to the input section of the operational amplifier 121 to limit the amplitude of the input signal. This protects the signal processing circuit against a sudden abnormal input. The averaging circuit 8 includes operational amplifiers 81, 82,
It is composed of diodes 83 and 84, a capacitor 85 and a resistor, and is configured as a low-pass filter that removes high frequencies after full-wave rectifying an input signal. The cutoff frequency of the low-pass filter is relatively large, such as a variation T ₃ when the cutting edge produces chips, a variation T ₄ due to a tool loss, and a variation T ₅ due to the resistance value of the bearing of the processing machine spindle. The high frequency component is removed, and the fluctuation T ₁ due to the inclination of the machined surface with respect to the reference surface and the fluctuation T ₂ due to the change in the cut amount
A frequency (for example, 2 Hz) capable of passing a relatively low frequency component such as is selected. When the waveform A in FIG. 1 passes through the averaging circuit 8, an output like the waveform B in FIG. 1 is obtained. The subtraction circuit 9 includes an operational amplifier 91 and an input resistor 9
2, 93, which subtracts the signal shown by the waveform B output from the averaging circuit 8 from the signal shown by the waveform A in FIG. A signal having a waveform C as shown in FIG. 5 is obtained. The output of the subtraction circuit 9 is compared with the threshold value T _th set by the threshold value setting circuit 10 for detecting the tool loss by the comparison circuit 11, and when the threshold value T _th is _exceeded , a tool loss occurs. A signal indicating that is output. That is, in the vicinity P of the tool loss occurrence in FIG.
The threshold value is exceeded and the tool loss detection signal S as shown in FIG. 6 is generated.

A high frequency component such as the waveform C shown in FIG. 5 which is the output of the subtraction circuit 9 can also be obtained by passing the output of the tool load detection circuit 5 through a high pass filter. Therefore,
In the circuits of FIGS. 3 and 4, a high pass filter may be used instead of the combination of the averaging circuit 8 and the subtracting circuit 9. However, the combination of the averaging circuit and the subtracting circuit is more advantageous in terms of noise generation in the filter used. That is, when a high-pass filter is used, noise may be mixed in the output waveform of the high frequency component, but in the case of the combination of the averaging circuit and the subtraction circuit, the low-pass filter is used.
The filter is advantageous because it does not generate high frequency noise components. A bandpass filter can also be used.

FIG. 7 shows a second embodiment of the present invention, in which parts having the same functions as those in FIG. 3 are designated by the same reference numerals. In this embodiment, an integrating circuit 14 to which an integration time setting circuit 13 is connected is inserted between the subtracting circuit 9 and the comparing circuit 11 of the embodiment shown in FIG. That is, the output of the subtraction circuit 9 is integrated by the integration circuit 14 within a very short time, so that the fluctuation T ₃ due to the generation of chips is averaged to clarify a sharp increase in output when a tool is lost. In FIG. 8, the time width obtained from the integration time setting circuit 13 is set to the waveform of only the cutting edge phenomenon level output from the subtraction circuit 9 which is the same as the waveform C of FIG.
FIG. 9 shows an averaged output within the time width obtained from the integration time setting circuit 13 for the waveform C having only the cutting edge phenomenon level shown in FIG. It shows a waveform in which a sharp output when a tool is broken is clarified. The output value clarified as shown in FIG. 9 is compared with the set value for tool loss detection set by the threshold setting circuit 10 in the comparison circuit 11, and if it exceeds the set value, an abnormality such as tool loss occurs. Output a signal that represents.

FIG. 10 shows a third embodiment of the present invention, in which parts having the same functions as those in FIG. 7 are designated by the same reference numerals. This embodiment corresponds to the integrating circuit 14 and the comparing circuit of the second embodiment (FIG. 7).
The first and second sample and hold circuits 16 and 17 in which the integration time control circuit 15 is connected between
8 is inserted. That is, in the subtraction circuit 9, the fluctuation level of only the cutting edge phenomenon is integrated by the integration circuit 14 in a minute time width to average the fluctuation amount due to the chip generation, and the integration time control circuit 15 is further connected to the first and second circuits. By sequentially shifting the output value integrated in a minute time width with the sample and hold circuits 16 and 17 and inputting it to the integrated value subtraction circuit 18 for subtraction, by removing the fluctuation due to chip generation,
It is possible to clarify the sudden increase in output when a tool is lost.

FIG. 11 shows a waveform of only the cutting edge phenomenon level obtained by subtracting the output of the averaging circuit 8 which is a low-pass filter from the output waveform of the tool load detection circuit 5 in the subtraction circuit 9 and the integration time setting circuit 13 It shows the integrated output value integrated by the integration circuit 14 at the time of acquisition, and the result obtained by averaging the fluctuations at the chip phenomenon level is obtained. In FIG. 12, the first and second integrated values, which are held by the first and second sample and hold circuits 16 and 17 and are separated by a minute time, are alternately input to the integrated value subtraction circuit 18. It is a figure which shows the value which was subtracted. As shown in this figure, since the variation at the chip generation level can be attenuated by the integral subtraction processing, the variation due to the tool loss can be further clarified. The comparison circuit 11 detects a tool loss when the input value exceeds the tool loss detection setting value T _th set by the threshold setting circuit 10.

[Brief description of drawings]

FIG. 1 shows an example of a torque fluctuation waveform of a tool in a cutting machine and a waveform obtained by averaging the waveform. FIG. 2 is a diagram showing the basic configuration of the present invention. 3 and 4 are diagrams showing a first embodiment of the present invention. FIG. 5 is a diagram showing a waveform C (output of the subtraction circuit 9 in FIG. 3) obtained by removing a relatively low frequency component from the torque signal. FIG. 6 is a diagram showing a signal (output signal of the comparison circuit 11 in FIG. 3) output when the waveform C in FIG. 5 exceeds the threshold value T _th . FIG. 7 is a block diagram showing the second embodiment of the present invention. 8 and 9 are waveform charts for explaining the second embodiment. FIG. 10 is a block diagram showing a third embodiment of the present invention. 11 and 12 are waveform charts for explaining the third embodiment. 1 ... Workpiece, 2 ... Surface plate, 3 ... Tool, 4 ... Tool holding part, 5 ... Tool load detection circuit, 6 ... Signal processing circuit, 7 ... Machine control panel, 8 ... Averaging circuit, 9 ...
Subtraction circuit, 10 ... Threshold setting circuit, 11 ... Comparison circuit,
12 ... Overcurrent prevention circuit, 13 ... Integral time setting circuit, 14 ...
… Integrator circuit, 15 …… Integral time control circuit, 16,17 …… Sample and hold circuit, 18 …… Integral value subtraction circuit.

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Masahiro Hayashi Inventor Masahiro Nagakute-cho, Aichi-gun, Aichi Prefecture 1 1st 41st Yokomichi Yokouchi Central Research Institute Co., Ltd. (72) Inventor Kazutaka Ogo Nagakute-cho, Aichi-gun Aichi Prefecture 1 in 41 Nagamichi Yokomichi Toyota Central Research Institute Co., Ltd. (72) Inventor, Daito Uematsu 4-26, Onda-cho, Kariya city, Aichi prefecture

Claims (5)

[Claims] 1. A fluctuation amount of a signal of a relatively low frequency such as a fluctuation due to a tilt of a machining surface of a workpiece with respect to a reference plane, a fluctuation due to a change in a cutting amount, or the like, from power or load current of a spindle motor of a machining machine. The fluctuations of relatively high frequency signals such as fluctuations when the cutting edge produces chips, sudden output fluctuations due to abnormal phenomena such as tool loss, fluctuations due to the resistance of the machine tool spindle accompanying tool rotation, etc. A tool load detection circuit for detecting a tool load fluctuation signal including: an output circuit for extracting and outputting a fluctuation component of the signal of the relatively high frequency from the tool load fluctuation signal; and the comparison output from the output circuit. And a signal processing circuit having a comparison circuit that outputs a signal indicating the occurrence of an abnormality such as a tool loss when the level value of the fluctuation of the signal of a relatively high frequency exceeds the set value for tool loss detection. Features Tool loss detection device that. 2. An output circuit of the signal processing circuit, an averaging circuit for extracting only the fluctuation component of the signal of the relatively low frequency from the tool load fluctuation signal, and the averaging circuit from the tool load fluctuation signal. And a subtraction circuit for subtracting the variation of the signal of the relatively low frequency extracted in the circuit and outputting the level value of the variation of the signal of the relatively high frequency. The tool loss detection device according to item (1). 3. The output circuit of the signal processing circuit comprises a high-pass filter for extracting only a fluctuation component of the relatively high frequency signal from the tool load fluctuation signal. The tool loss detection device described in 1). 4. The relatively high frequency when the cutting edge produces chips by integrating an output signal of the subtraction circuit between the subtraction circuit and the comparison circuit for a set integration time. The tool loss detection device according to claim (2), characterized in that an integration circuit for averaging the fluctuation amount of the signal of (1) is provided. 5. First and second sample and hold circuits for alternately sampling and holding the output of the integrator circuit between the integrator circuit and the comparator circuit, and the first and second sample and hold circuits. The tool loss detection device according to claim (4), further comprising an integrated value subtraction circuit for performing subtraction between adjacent sample values output from the hold circuit.