JPH0460780B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for detecting abnormalities in tools used in multi-edge tools, such as milling machines and hobbing machines.

In order to achieve unmanned machining, various means have been proposed and put into practical use for automatically detecting abnormalities (for example, a chipped blade) in a tool of a machine tool.

However, multi-blade tools such as milling tools and hobs perform intermittent cutting, and the cutting conditions (cutting width, depth of cut, etc.) change significantly.
Conventional abnormality detection methods cannot accurately detect such abnormalities, and therefore, there has been a desire for a means that can accurately detect abnormalities in such multi-edged tools.

In view of this situation, the present invention seeks to provide a method that can accurately detect abnormalities in multi-blade tools such as the above-mentioned milling tools.

For this reason, in the present invention, the cutting force change of the multi-edge tool is detected, and the detected cutting force change is subjected to cepstrum processing, and the cepstrum of the period at which the next blade reaches an arbitrary blade position of the multi-edge tool is calculated. An abnormality in the multi-blade tool is detected based on a comparison result between the magnitude of the cepstrum and the magnitude of the cepstrum for one rotation period of the tool.

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 shows an example of the configuration of an apparatus for carrying out the method according to the present invention, and FIG. 2 shows a flowchart for explaining the operation of the apparatus.

Milling cutter 1 of the vertical milling machine shown in Fig. 1
Since this is a multi-blade tool having a plurality of cutting blades (chips) 1a, when the main shaft 2 of the milling machine is rotated, intermittent cutting is performed by the individual blades 1a. Therefore, the cutting force of the milling cutter 1 changes periodically during cutting, and this change in cutting force is transmitted to the main shaft 2 as a change in the thrust direction.

In this embodiment, a piezoelectric stress sensor 5, for example, is interposed between the housing 4 and the bearing 3 that receives the thrust force of the main shaft 2.
The thrust force corresponding to the cutting force is converted into voltage (step 100 in FIG. 2).

The output voltage of the stress sensor 5 is amplified by a factor of A by the amplifier 6 (step 101), and the contained high frequency component is removed by the low pass filter 7 and the DC component is removed by the capacitor 8. (Steps 102, 103). As a result, the capacitor 8 outputs a signal as illustrated in FIG. 3 or FIG. 6, which corresponds to the change in cutting force of the milling cutter 1.

Figure 3 shows the signal waveform when each blade of the milling cutter 1 is normal, and Figure 6 shows the signal waveform when one of the blades is defective. This is the case when milling was performed under the following machining conditions.

Processing machine: NC vertical milling machine (AC18.5kw) Cutter: Outer diameter 125φ, number of teeth 6, approach angle 150,
Rake angle (axial + 8°, radial
0゜) Chip: Rotation speed 285 rpm, cutting speed 112 m/min,
Feed rate: 0.3 mm/blade (513 mm/min), depth of cut: 3 mm, cutting width: 60 mm, down cut Work material: S45C plate material (400 x 300 x 50 mm) The above signal output from the capacitor 8 is as follows:
After being A/D converted by an analog/digital converter (hereinafter referred to as an A/D converter) 9 (step 104), it is input to a cepstrum processor 10,
Here, so-called cepstrum processing is performed.

That is, the cepstrum processor 10 first performs fast Fourier transform (FFT) on the output signal of the A/D converter 9 to obtain the power spectrum of the signal (step 105). FIGS. 4 and 7 show power spectra when the time waveforms shown in FIGS. 3 and 6 are subjected to FFT, respectively. Next, the cepstrum processor 10 takes the logarithmic value of the amplitude of the obtained power spectrum (step 106), and then fast Fourier transforms this logarithmized spectrum to obtain a cepstrum (step 107). FIGS. 5 and 8 show cepstrums based on the power spectra shown in FIGS. 4 and 7, respectively.

The arithmetic circuit 11 receives a signal indicating the cepstrum and performs peak hold on the signal. and the rotation speed S of the milling cutter 1
and the number of blades n (step 108), and calculate the amplitude C 1 of the cepstrum at the period (in this example, 35 ms, hereinafter referred to as the intermittent period) at which the next blade arrives at an arbitrary blade position of the cutter ₁ . And the above katsuta 1
The amplitude of the cepstrum during one rotation period (210 ms in this example, hereinafter referred to as the rotation period)
Find Cn (step 109). Next, the ratio Cn/C ₁ of the above-mentioned amplitudes C ₁ and Cn is determined (step 110), and a threshold value for this ratio is input (step 111), so that the above-mentioned ratio Cn/C ₁ is greater than the above-mentioned threshold value. Determine whether it is large or not (step 112),
NC that controls the milling machine if Cn/C ₁ >
A feed stop signal is output to the controller 12 (step 113). As a result, the NC controller 12 stops feeding the cutter 1 and notifies the operator of the abnormality of the tool (step 114).

Here, the above threshold value will be explained. Fifth
As is clear from the comparison between the figure and FIG. 8, when the tool is abnormal, the cepstrum for the intermittent period decreases, and conversely, the cepstrum for the rotation period increases. This phenomenon occurs for the following reasons. In other words, in the case of normal cutting without any chipping of the blade, the change pattern of the cutting force for each blade is repeated in a similar manner, so the cepstrum of the intermittent period becomes large, while the Since the cutting force changes are not emphasized so much, the cepstrum of the rotation period becomes small.

However, if something goes wrong with the tool,
Since the repeatability of the cutting force change pattern for each blade is disrupted, the cepstrum of the intermittent period decreases. In this case, the cepstrum of the rotation period increases because the change in cutting force each time the cutter rotates once is emphasized.

FIG. 9 shows the variation in the ratio value Cn/ _C1 between when the tool is normal and when the tool is abnormal when milling is performed under the above machining conditions.
As shown in the figure, when the tool is normal, the above ratio value is
The value of this ratio becomes smaller than 0.5, and the value of this ratio becomes larger than 1 when the tool is abnormal. Therefore, in the above example, by setting the threshold value to 1.0, it is possible to determine an abnormality in the tool with 99.9% accuracy.

The ratio value distribution diagram shown in Fig. 9 is obtained when sampling is performed 20 times each when the tool is normal and when the tool is abnormal, and the average value and standard deviation of the ratio values when the tool is normal are each 0.29. and 0.3, and those during tool abnormality are 1.82 and 0.27.

In the above embodiment, tool abnormalities are detected based on the value of the cepstrum ratio between the intermittent period and the rotation period. Of course, detection is also possible.

Further, in the embodiment, changes in cutting force are detected from the thrust force acting on the main shaft 2, but the cutting force in the feed direction of the cutter 1, the cutting force in the direction perpendicular to the feed,
The spindle torque or the like may be detected by an appropriate sensor. Furthermore, during cutting, vibrations and cutting noises corresponding to the cutting force are generated in various parts of the processing machine, so these may be detected by a vibration sensor or a microphone.

Of course, the present invention can be applied not only to milling machines but also to other machine tools equipped with multi-blade tools.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an example of the configuration of an apparatus for carrying out the method according to the present invention, and FIG.
Flowchart for explaining the operation of the device shown in the figure, Figure 3 is a waveform diagram illustrating changes in cutting force when the tool is normal, Figures 4 and 5 are respectively related to the waveform diagrams shown in Figure 3. Figure 6 is a waveform diagram illustrating the cutting force change during tool abnormality, Figures 7 and 8 are the spectrum and cepstrum of the waveform diagram shown in Figure 6, respectively. The waveform diagram, FIG. 9, is a graph showing the distribution of cepstrum ratios in the intermittent period and rotation period when the tool is normal and when the tool is abnormal. DESCRIPTION OF SYMBOLS 1...Milling cutter, 1a...Chip, 2...Main shaft, 3...Bearing, 5...Stress sensor, 10...Cepstrum processor, 11...Arithmetic circuit, 12...NC
controller.

Claims (1)

[Claims] 1. Detecting changes in cutting force of a multi-blade tool,
Cepstrum processing is performed on the detected cutting force change, and the size of the cepstrum for the period at which the next blade arrives at an arbitrary blade position of the multi-blade tool, and the size of the cepstrum for the period at which the multi-blade tool rotates once. A method for detecting an abnormality in a multi-edged tool, characterized in that the abnormality in the multi-edged tool is detected based on the comparison result. 2. The method for detecting an abnormality in a multi-blade tool according to claim 1, wherein a change in the thrust force acting on the main shaft of the multi-blade tool is detected as a change in the cutting force. 3. The method for detecting an abnormality in a multi-blade tool according to claim 1, wherein the ratio of each of the cepstrums is determined, and abnormality in the tool is detected based on the value of this ratio. 4. The method for detecting an abnormality in a multi-blade tool according to claim 1, wherein the difference between the respective cepstrums is determined, and an abnormality in the tool is detected based on the magnitude of this difference. 5. The method for detecting an abnormality in a multi-blade tool according to claim 1, wherein the multi-blade tool is a milling tool of a milling machine.