JPH04372340A - Abnormality prediction device for rotating tools - Google Patents

Abstract

PURPOSE:To provide a rotating tool abnormality predicting device capable of predicting abnormality correctly taking account of the internal fatigue of a rotating tool. CONSTITUTION:The cutting force of a rotating tool is detected by a cutting force detector 2, and the fluctuation element of the cutting force demodulated by a demodulator 4 is taken out by a high-pass filter 18. The fluctuation element is integrated at a CPU 11 so as to set the threshold value as the function of the integrated value, and an abnormality signal is outputted when the cutting force exceeds this threshold value.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an abnormality prediction device for rotating tools such as drills, taps, end mills, etc., which detects in advance the occurrence of abnormal conditions such as breakage, chipping, wear, etc.

0002

[Prior Art] When cutting a workpiece using a rotating tool such as a drill, tap, or end mill, the rotating tool may break or break.
If defects, wear, etc. occur, not only will it be impossible to perform the intended machining on the workpiece, but if operation continues in such abnormal conditions, the workpiece will be damaged and the machine tool will be damaged. may cause damage.

[0003] As a result, rotary tools actually break, chip, or
It is necessary to predict the occurrence of these abnormal conditions before wear or the like occurs, and various methods have been proposed for this purpose. That is, one method is to detect the drive current of a motor that drives the main shaft of a mounting device for mounting a rotary tool, and to predict the occurrence of an abnormal state of the rotary tool by an increase in this drive current. Another method is to detect the vibration of the rotary tool with a vibration detector such as an AE sensor, and predict the occurrence of an abnormal state of the rotary tool based on an increase in the vibration value. Furthermore, a third method was proposed in JP-A-62
- The method described in Publication No. 166948 detects the cutting force of a rotary tool with a cutting force detector, and compares the cutting force detected by the cutting force detector or its average value, or the previously detected cutting force and the currently detected cutting force. The abnormal state of the rotary tool can be determined by comparing the difference between the cutting force or the average value of the cutting force for a predetermined number of successive times and the difference between the average value of the cutting force for a predetermined number of successive times with a predetermined threshold value. This is a method of predicting occurrence.

FIG. 11 shows the configuration of the device disclosed in Japanese Patent Application Laid-Open No. 166948/1983.

In FIG. 11, reference numeral 1 denotes a numerical control (NC) device, which also includes a spindle for rotating a rotary tool. 2
is a tool holder type cutting force detector. A rotary tool such as a drill is attached to this cutting force detector 2 to perform cutting. The cutting force (thrust and torque) at this time is output as a cutting force signal from the FM transmitter in the cutting force detector 2, and this signal is input to the demodulator 4 via the external receiving antenna 3. The signal demodulated by the demodulator 4 is converted into a digital signal by the A/D converter 5, and the computer 6 performs the various determinations described above. If it is determined that the cutting force value is abnormal as a result of this determination, a signal is sent from the computer 6 to the alarm device 8 via the D/A converter 7 to activate the alarm device 8, and the remote buffer 9 A signal to avoid breakage is sent to the NC device 1 via the NC device 1, and the cutting is stopped immediately.

A signal as shown in FIG. 12 is input to the computer 6. (a) and (b) are aluminum materials cut with a drill with a diameter of 2 mm, (c) and (d) are steel materials S5
This is the signal when 0C is cut with a 2mm diameter drill. Threshold values of A, B, A', and B' are set for these, and when these are exceeded, a signal is sent to the alarm device 8 and NC device 1. . Here, the threshold value is set according to various cutting conditions, and may be a value determined through various experiments, or a predetermined multiple of the average value after a steady state is reached after the start of cutting.

Furthermore, as shown in FIG. 12(b), the magnitudes S1, S2,...Sn of the spike-like cutting force fluctuation components and the number of times n thereof are counted, and the magnitude S1 of the cutting force fluctuation components is calculated.
, S2,...Sn are compared with a threshold value S set for each drill used, and the number of occurrences n of cutting force fluctuation components exceeding a predetermined threshold value is compared with a preset number N, and Si(i=1,2
...n) Sends a signal to the alarm device 8 and the NC device 1 when >S or n>N.

[0008]

[Problem to be Solved by the Invention] This proposed method of determining threshold values for cutting force and cutting force fluctuation components compares the instantaneous magnitude of stress on the drill at the time of measurement with a threshold value. This does not take into account fatigue caused by external forces previously applied to the drill. Regarding this point, the same applies to the other conventional methods mentioned above. However, if the rotary tool has accumulated fatigue due to external force, the rotary tool may break even with a slight external force. It is necessary to detect the accumulation of external forces.

[0009] In this regard, the method disclosed in the above publication also discloses a method of issuing an alarm by counting the number of times the fluctuation component exceeds a threshold value, as described above, and although this method can also detect the accumulation of fatigue to a certain extent. Even if a fluctuation component can be counted as one fluctuation, the time and magnitude of the external force that causes this one fluctuation vary, and it is impossible to accurately predict abnormalities by simply counting the number of fluctuations. It is.

The present invention has been made in order to eliminate the above-mentioned drawbacks of the conventional rotating tool abnormality prediction device, and it is a rotating tool that can accurately predict abnormalities by taking into account the internal fatigue of the rotating tool. The purpose is to provide an abnormality prediction device.

[0011]

[Means for Solving the Problems] An abnormality prediction device for a rotating tool according to the present invention includes a detection means for detecting a cutting force of the rotating tool;
an extraction means for extracting only the variation component of the cutting force detected by the detection means; an integration means for integrating the variation component extracted by the extraction means; and a threshold value set as a function of the integral value of the integration means. and a comparison means that outputs an abnormal signal when the cutting force detected by the detection means exceeds the threshold set by the threshold setting means.

0012

[Operation] In the rotating tool abnormality prediction device of the present invention,
By extracting only the fluctuating component of the cutting force of the rotary tool, setting a threshold value as a function of the integral value of the fluctuating component, and comparing the threshold value with the above fluctuating component, abnormal conditions are predicted, so the load of external force can be predicted. This makes it possible to detect the accumulation of variable components appearing in the cutting force of the rotary tool, and when the accumulation is large, an abnormal state can be detected even with the occurrence of a slight variable component, making it possible to accurately predict abnormalities.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

First, the configuration of the embodiment shown in FIG. 1 will be explained.

A cutting force detector 2 is attached to the main shaft of the NC device 1. The demodulator 4 receives F from the cutting force detector 2.
It is arranged to receive a torque/thrust signal (cutting force signal) indicating M-modulated torque and thrust via the antenna 3. A low-pass filter 10 is connected to the output side of the demodulator 4, and an A/D converter 5 and a high-pass filter 18 are connected to the output terminal of the low-pass filter 10.
are connected in parallel. An A/D converter 19 is connected to the output terminal of the high-pass filter 18.

The CPU 11, which functions as an integrating means, a threshold value setting means, and a comparing means, is connected to the A/D converters 5 and 19 via a common bus 12, as well as a RAM 13, a display 14, and input/output for external output. An alarm input/output section (I/O) 15, a keyboard 20, and an alarm input/output section (I/O) 16 are connected, and an alarm device 17 is connected to the alarm input/output section (I/O) 16.

An example of a specific configuration of the cutting force detector 2 is shown in FIG.
Shown below.

The cutting force detector shown in FIG.
Two types of detection parts having different functions, such as a thrust detection part, a torque detection part, a thrust detection part, and a torque detection part, are arranged alternately in the circumferential direction in the ten cylindrical parts. The cylindrical portion of this cylindrical member 110 is divided into four parts by four vertically dividing grooves 111 provided with a phase difference of 90°, and slits 112 opening on both upper and lower end surfaces are bored in the circumferential direction in the two opposing cylindrical portions. Spring-like elastic support structure parts A and A are arranged radially on both sides for torque detection, and the remaining two opposing cylinder parts have left and right vertical dividing grooves 111, and slits 113 that open at the end faces of 111. A parallel spring-like elastic support structure for thrust detection is provided on both sides of the hole in the longitudinal direction.
It forms B. Flanges 114 and 115 are screwed onto the upper and lower ends of the cylindrical member 110 in the following manner, respectively.The flange 114 is provided with an internal thread 116, and the screw 118 that engages with the internal thread 116 is as described above. Cut only the upper side of the torque sensor, and leave no such screw on the upper side of the thrust sensor. On the other hand, flange 11
The screw 119 that engages with the internal thread 117 of No. 5 is cut only on the lower side surface of the thrust detection section, and such a screw is not provided on the lower side surface of the torque detection section. It is desirable to securely fix any screwed parts by welding later. The flange 115 is fixed, for example, to a rotating shaft such as a main shaft of a machine tool.

If the torque shown by arrow T and the thrust shown by arrow P are applied to the flange 114 side opposite to the flange 115 fixed to the rotating shaft, the torque T will cause the screws 116 and 118 to Then, the torque is transmitted to the portion of the elastic support structure A of the torque sensing portion that is on the outside with respect to the slit 112, and a strain is generated between it and the portion that is on the inside of the slit 112. Further, the thrust P is transmitted through the slit 1 of the elastic support structure portion B of the thrust detection section through the inner part of the cylindrical member 110.
The strain is transmitted to the part that is on the inside with respect to 13, and a strain is generated between it and the part that is on the outside that is received by the fixed screws 119 and 117. Any strain is extracted as an electrical signal by a strain gauge fixed to the strained portion, and output as a cutting force signal from the FM transmitter in the cutting force detector 2. Of course, it is also possible to utilize the principle of a potentiometer, an actuating transformer, or a magnetic encoder, or to extract the distortion using a piezoelectric element.

FIG. 3 shows another example of the cutting force detector 2. In this example, the torque detection part and the thrust detection part are arranged concentrically, and the thrust P and torque T transmitted from the rotating tool are transmitted to the outer flange (not shown) via the elastic support structure part A and the elastic support structure part B. It is transmitted to At this time, the elastic support structure part A has low rigidity against the action of torque, so it is greatly distorted, and the elastic support structure part B has low rigidity against the action of thrust, so it is greatly distorted. Therefore, by arranging a means for detecting strain, such as a strain gauge, at an appropriate position on the elastic support structure, the torque and thrust acting on the rotary tool can be detected.

Next, the operation will be explained. As described above, the cutting force detector 2 FM modulates the torque thrust signal indicating the torque thrust and sends it to the antenna 3 and demodulator 4. The FM modulation signal is demodulated by a demodulator 4, high frequencies such as noise are separated by a low-pass filter 10, and the above-mentioned torque and
Thrust signal is obtained. Further, the signal passed through the high-pass filter 18 after passing through the low-pass filter 10 contains only the cutting force fluctuation component. The torque/thrust signal and the cutting force fluctuation component are converted into digital signals by the A/D converters 5 and 19 and taken into the CPU 11.
11 is compared with a predetermined threshold value set by
Alarm I/O when the thrust signal value exceeds a predetermined threshold
The alarm signal is output to the alarm device 17 via the alarm device 16.
7 notifies the alarm by display or sound.

Next, examples of thrust signals and fluctuation components detected by the CPU 11 are shown in FIGS. 4 to 5, and function patterns for determining threshold values are shown in FIGS. 6 to 8. Figures 4 and 5 show an 8mm hole drilled in S45C tempered steel with a 1.5mm diameter drill at a rotation speed of 18mm.
00r. p. m. , feed 0.05mm/rev. Fig. 4(a) shows the thrust signal characteristics when drilling with a new drill, and Fig. 4(b) shows the torque or thrust signal when drilling with a new drill. It is a characteristic diagram of the fluctuation component after passing. Figure 4
In (a), a low fluctuation component in the second half frequency of the cutting moving body appears, but because the frequency setting of the high-pass filter 18 is high, it is not detected in FIG. 4(b). As a result, a substantially constant and small fluctuation component is detected in the output signal of the high-pass filter 18.

Next, FIG. 5 shows the thrust signal at the time of breakage, and although the fluctuation component increases just before the time of breakage t1, the thrust itself does not increase much. Therefore, the diameter of this example is 1
．． When setting the threshold value S (1.5) for the cutting force for 5 mm drilling, the maximum thrust force S that can be generated even with a new drill
(L) and the maximum thrust force S(H) generated before breakage. Then, the maximum thrust force S (L
) and S(H), the threshold value S(1.
5) will be very narrow, or the range of threshold S(1.
5) may not be set and breakage prediction may not be possible. Further, although the thrust fluctuation component also becomes a large value just before breakage, it is dangerous to simply set a constant threshold value, considering the time required to stop the feeding of the machining center.

Therefore, as shown in FIG. 6, the threshold value S'(1
．． 5) is a function of the integral value Ci of the thrust fluctuation component: S
'(1.5)=f(Ci, 1.5).

Although this embodiment is defined using the function S'(1.5)=aCi+Smax(1.5), it may also be an inverse function, natural logarithm, etc., and the function form, constant a (a is negative), Smax( 1.5) may be determined through experiments, etc.

[0027] Also, in the case of a 0.5 mm drill, Smax (0
．． By changing 5), a threshold with a similar effect can be obtained. The function of S'(1.5) is shown in Figure 5(a).
As shown inside, it becomes low near t1 at the time of breakage,
Breakage can be predicted more reliably than using a fixed threshold value.

The integral value of this variation and the function value of the threshold value for the integral value are determined by the Sma input from the keyboard 20.
It is calculated by the CPU 11 based on the value of x and the constant a.

Further, as a function of the threshold value, as shown in FIG.
．． If you set a function that becomes 0 at Ci1.5 for a 5mm drill, then the accumulation of cutting force fluctuations due to external force applied to the rotating tool will reach a predetermined value, regardless of the magnitude of the fluctuation component. An alarm will be activated to indicate that the rotary tool fatigue has reached a predetermined limit.

Furthermore, the function in FIG. 6 and the function in FIG. 7 may be combined to set the function in FIG. 8 as a threshold function. That is, the threshold function is S' (1.5) = aCi + Smax (1.5) (0
≦Ci≦Ci1.5)S'(1.5)=0
(Ci>Ci1.5).

In this case, as the fatigue of the rotary tool increases, even a slight variation in the cutting force due to external force will exceed the threshold value, and furthermore, as soon as the integral value of the variation component reaches a predetermined value, a warning will be displayed. The Rukoto.

Next, the breakage pattern will be explained. Figure 9
, and Fig. 10 are signal examples when the breakage occurred. Fig. 9 shows an example where the torque and thrust both increased and the breakage occurred, and Fig. 10 shows an example where the cutting force did not increase much after the cutting force fluctuation component increased and the breakage occurred. According to the present invention, it is possible to predict even a breakage pattern example of the type shown in FIG. 9, which could not be predicted conventionally.

As described above, in the rotary tool abnormality prediction apparatus according to the present invention, the CPU 11 detects the cutting force fluctuation component and the cutting force for each cutting, and calculates the integral value of the fluctuation component. The predetermined threshold value is changed depending on the magnitude of this integral value, and compared with the current cutting force.As a result, as cutting progresses and a rotating tool with many variable components is subjected to repeated fatigue, it is determined that it is dangerous. An abnormal signal is output.

[0034] Furthermore, even if a fluctuation component occurs only sporadically, if the fluctuation component is large, an abnormality signal is issued, indicating that the load acting on the drill is large. Furthermore, even if the number of small fluctuation components such as noise is large, the magnitude of the integrated value is small, so no abnormal signal is output, and breakage can be accurately predicted.

Note that in this embodiment, the cutting force component in a predetermined frequency band formed by the low-pass filter 10 and the high-pass filter 18 was used as the fluctuation component.
Focusing only on a predetermined frequency component (it may be multiple),
Only the frequency component may be extracted and used as the fluctuation component.

Furthermore, since the frequency characteristics of the cutting force vary depending on the drill diameter, cutting conditions, material of the workpiece, etc., an appropriate threshold function value is determined in advance through experiments, and is input into the RAM 13 using the keyboard 20. It may also be used as a threshold.

Furthermore, in this embodiment, the cutting force of the rotary tool was used as the abnormal condition detection data, but the vibration of the rotary tool or the drive current of the drive device that drives the rotary tool, which indirectly represents the cutting force, was used as abnormal state detection data. It may also be used as state detection data. In this case, the cutting force will be influenced by the vibration or drive current.

[0038]

[Effects of the Invention] In the rotating tool abnormality prediction device of the present invention, the threshold value is set as a function of the integral value of the fluctuating part of the cutting force that indicates the abnormal state of the rotating tool. It is possible to accurately predict an abnormal state depending on the state.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the configuration of an embodiment of the present invention.

FIG. 2 is an exploded perspective view showing a specific example of the configuration of the cutting force detector of the embodiment shown in FIG. 1;

FIG. 3 is a perspective view showing another example of the cutting force detector of the embodiment shown in FIG. 1;

FIG. 4 is a graph showing an example of cutting force data in a normal state.

FIG. 5 is a graph showing an example of cutting force data in an abnormal state.

FIG. 6 is a graph showing an example of a threshold function.

FIG. 7 is a graph showing another example of a threshold function.

FIG. 8 is a graph showing a third example of a threshold function.

FIG. 9 is a graph showing a pattern of deformation of cutting force when the rotary tool breaks.

FIG. 10 is a graph showing another pattern of deformation of cutting force when the rotary tool breaks.

FIG. 11 is a block diagram showing the configuration of an example of a conventional abnormality prediction device for a rotating tool.

FIG. 12 is an example of a signal waveform representing cutting force detected by the conventional cutting force detector of FIG. 11;

[Explanation of symbols]

1 NC device 2 Cutting force detector 4 Demodulator 10 Low pass filter 11 CPU 17 Alarm device 18 High pass filter 20 Keyboard

Claims (1)

[Claims] 1. A detection means for detecting the cutting force of a rotary tool, an extraction means for extracting only the fluctuation component of the cutting force detected by the detection means, and an integral of the fluctuation component extracted by the extraction means. an integrating means for determining a threshold value as a function of an integral value of the integrating means; and a threshold setting means for setting a threshold value as a function of an integral value of the integrating means; 1. An abnormality prediction device for a rotating tool, comprising a comparison means for outputting a signal.