JPH11179636A - Replacement timing decision system for tool - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a replacement timing decision system of a tool which can be constitution by reducing a cost further with high reliability. SOLUTION: A replacement timing decision system of a tool measuring the vibration of a rotary tool 31 having a cutting edge by a vibration sensor 51 to decide replacement timing of the rotary tool 31 is provided with a filter 53 separating a prescribed frequency band from a vibration sensor signal S3 output from the vibration sensor 51, taking means synchronizing with a trigger signal S2 of 1 pulse relating to one turn of the rotary tool 31 to take in the vibration sensor signal S3 through the filter 53, and a decision means deciding replacement timing of the rotary tool 31 from the vibration sensor signal S3 taken in by this taking in means. The vibration sensor signal S3 is taken in synchronized with the rotation of the rotary tool 31, so that a vibration characteristic in according with a cutting edge position can be clearly detected.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a tool replacement time judging system for measuring the vibration of a rotary tool provided with one or a plurality of cutting blades with a vibration sensor and judging the time for changing the rotary tool.

[0002]

2. Description of the Related Art Conventionally, in a machine tool equipped with a numerical control device (hereinafter referred to as an NC device), when a rotary tool such as an end mill is determined to be replaced by an NC device, and when the rotary tool reaches a use limit due to wear or the like. A tool change time determination system that automatically changes the rotary tool by an automatic tool changer (ATC) is employed. By providing such a tool replacement time determination system, the machine tool can be fully automated, so that the productivity of products by the NC machine tool can be significantly improved.

The above-described tool replacement time determination system detects, for example, a torque acting on a main shaft of a rotary tool, and when the detected torque exceeds a certain threshold value,
A system that determines that it is time to change the tool, or detects the machining vibration of the rotary tool and converts the vibration level value (rms value) obtained by converting the processing vibration into a value that exceeds a certain threshold value. Is known. As another tool replacement time determination system, there is known a system in which an accumulated use time of a rotary tool is stored in an NC device, and a tool replacement time is determined based on the accumulated use time of the rotary tool.

[0004]

However, such a conventional tool replacement time determination system has the following problems. That is, since the system that detects and determines the torque of the main shaft detects the state of the rotating tool through a relatively large inertia element such as a motor and a power transmission system, it requires high measurement accuracy, and accordingly, There is a problem that an expensive measuring device must be used. In addition, the system for detecting machining vibration and converting it into a vibration level value to determine the vibration level value greatly changes depending on the machining conditions such as the material of the work, so that the NC machine tool system in which various machining conditions can be set. In general, there is a problem that reliability is not so high.

[0005] Further, a system for judging from the accumulated use time of the tool is required for the wear that progresses with the use time.
Although it is an effective system, it has a problem that it cannot cope with chipping, breakage, and the like of the cutting edge of the rotating tool that occurs accidentally. Further, in such a system, for safety, it is necessary to set a set time for determining a tool replacement time to be shorter than an actual usable time. There is a problem that soars.

SUMMARY OF THE INVENTION An object of the present invention is to provide a highly reliable tool replacement time determination system which can be configured at a low cost.

[0007]

According to the present invention, there is provided a tool replacement time determining system for determining the replacement time of a rotary tool by measuring the vibration of a rotary tool provided with one or more cutting edges with a vibration sensor. A tool replacement timing determination system, which separates a predetermined frequency band from a vibration sensor signal output from the vibration sensor, and synchronizes with a trigger signal of one pulse for one rotation of the rotary tool, The apparatus is characterized in that it comprises a capturing means for capturing the vibration sensor signal that has passed through the filter, and a determining means for determining a replacement time of the rotary tool based on the vibration sensor signal captured by the capturing means.

According to the present invention, disturbance vibration other than vibration of the rotary tool is removed by the filter, and the vibration sensor signal is taken in synchronization with the rotation of the rotary tool. The corresponding vibration characteristics can be clearly detected. By measuring the temporal change of the vibration characteristic, it is possible to detect the wear state of each cutting edge of the rotary tool with high accuracy. Therefore, by accurately grasping the state of wear of the rotary tool, it becomes possible to make a highly reliable determination of the replacement time of the rotary tool.

In addition, since the tool replacement time judging system according to the present invention can accurately grasp the wear state of the rotary tool, the rotary tool can be used to its limit, and the machining cost can be reduced without increasing the tool cost. Reduction is achieved.
Furthermore, since a system can be constructed simply by connecting a filter, a vibration sensor, etc., the cost required for system construction is reduced, and since the system configuration is simple, the system can be easily constructed on a machine tool. It becomes possible.

In the above, a low-pass filter for removing high-frequency components of the vibration sensor signal is used as the above-mentioned filter, and the above-mentioned determination means is based on a vibration waveform for one rotation of the vibration sensor signal passed through the low-pass filter. A determination unit including a detection unit that detects a peak value according to the number of cutting edges of the rotary tool and a determination unit that determines a replacement time of the rotary tool based on a temporal change in the peak value detected by the detection unit is considered. Can be

That is, since disturbance vibrations other than vibrations of the rotary tool frequently appear in a high frequency band, it is possible to easily and efficiently remove them by using a low-pass filter.
In addition, since the determination unit includes a detection unit that detects the peak value of the vibration sensor signal according to the number of cutting edges of the rotary tool, the determination unit measures the temporal change in the peak value, and determines the cutting edge of the cutting edge. The wear state can be determined with high accuracy, and a highly reliable tool replacement time determination system can be provided.

[0012] Further, the above-mentioned determination means includes an analysis unit for performing a wavelet analysis on the vibration sensor signal passed through the filter to obtain a wavelet coefficient corresponding to the vibration sensor signal, and a time change of the wavelet coefficient. Therefore, it is conceivable to employ a judging means provided with a judging unit for judging the time to replace the rotary tool. That is,
Since the determination means uses wavelet analysis to analyze the vibration sensor signal, it is possible to analyze the vibration waveform in the frequency range of the high frequency range with high time resolution, and to determine the wear state of the rotating tool with extremely high accuracy. Can be detected. Accordingly, the determination of the tool replacement time by the determination means is also performed with high accuracy, and it is possible to perform extremely reliable determination.

The above-mentioned determination section includes a first step of calculating an average value of wavelet coefficients included within a predetermined time, and a second step of forming a smoothing curve representing a temporal change of the average value of the coefficient. And a third step of determining the peak of the smoothing curve as a wavelet coefficient at a position corresponding to the cutting edge of the rotary tool. Here, the average value of the wavelet coefficients can be obtained by using a moving average method or the like. In other words, if the wavelet coefficient at the position corresponding to the cutting edge is obtained in this way, it is possible to measure the temporal change of the wavelet coefficient with high accuracy in a stable state without being affected by the peak change due to the intrusion of random disturbance. Becomes possible.

When the vibration sensor signal is A / D-converted and taken into the taking-in means, it is preferable that the cutoff frequency of the filter is set to a half or less of the sample frequency in the A / D conversion. That is, by setting the cutoff frequency to be equal to or less than 1/2 of the sample frequency in the AD conversion, it is possible to prevent a so-called aliasing phenomenon in which a frequency component higher than the highest frequency in the FFT analysis appears to be folded below the highest frequency. It becomes possible.

[0015]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One embodiment of the present invention will be described below with reference to the drawings. FIG. 1 shows an NC employing a tool replacement time determination system according to a first embodiment of the present invention.
A machine tool is shown. The NC machine tool 1 includes a machine body 3 for cutting a work 21 placed on a table 2 and an NC device 4 for numerically controlling the machine body 3.

The machine body 3 includes an end mill 31 for cutting the surface of the work 21 and an electro-hydraulic pulse motor 33 for rotating the end mill 31 around a main shaft 32. Electric-hydraulic pulse motor 33
Is provided with a vibration sensor 51 composed of an accelerometer, whereby it is possible to detect a vibration accompanying rotation of the end mill 31. The end mill 31 has two cutting edges 31A, 31A as shown in the front view on the cutting surface side in FIG.
B is a rotary tool that performs cutting work, and has two cutting edges 3
If any of the cutting edges 1A and 31B cannot be cut normally due to wear, chipping, or the like, it is necessary to replace the end mill 31 with a new one.

As shown in FIG. 1, the NC device 4 includes a control device 41 for controlling the above-described electric-hydraulic pulse motor 33, and a controller 42 for sending a control signal S1 to the control device 41. Controller 4
2 is provided with an IO extension box 43 for signal input / output. Incidentally, the IO extension box 43 has an AD converter inside, thereby enabling high-speed sampling of signals.

The IO expansion box 43 has a plurality of input / output ports, and includes a control signal line 61, a trigger signal line 6,
2 and the vibration sensor 51 via a vibration sensor signal line 63. The vibration sensor signal line 63 is provided with a vibration amplifier 52 that amplifies the vibration sensor signal S3 acquired by the vibration sensor 51 and a low-pass filter 53 that separates and removes high frequency components from the acquired vibration sensor signal. I have.

The controller 42, as shown in FIG.
A central processing unit 46 including an arithmetic unit 44 and a recording unit 45; and a display unit 4 such as a CRT display
7 and an input device 48 such as a keyboard. The arithmetic unit 44 has a high-speed arithmetic element including a microprocessor or the like, and includes a control unit 441 that controls various input / output devices, a capturing unit 442 that captures various signals input from the IO extension box 43, A determination means 443 for determining the state of wear of the end mill 31 based on the signal captured by the capture means 442 is provided. Although not shown in FIG.
Reference numeral 43 denotes a detection unit that detects the peak value of the signal, and a determination unit that determines the replacement time of the end mill 31 based on the peak value of the signal detected by the detection unit.

In the machine tool 1 having such a configuration, the system for determining a tool replacement time operates as follows. The control signal S1 is output from the controller 42 to the control device 41 via the control signal line 61. The control device 41 drives the electric-hydraulic pulse motor 33 while controlling the number of revolutions, the feed rate, and the like based on the control signal S1, and together with the drive control of the electric-hydraulic pulse motor 33, the end mill 31 rotates once. On the other hand, a one-pulse trigger signal S2 is output to the controller 42 via the trigger signal line 62.

The controller 42 has a trigger signal S2
Is instructed to output a vibration sensor signal S3 to the vibration sensor 51 via a sensor control line not shown in FIG. Vibration sensor signal S3 output from vibration sensor 51
Is amplified by the vibration amplifier 52 and the high-frequency component is removed by the low-pass filter 53, and is taken into the taking means 442 in synchronization with the trigger signal S2.

The cut-off frequency of the low-pass filter 53 is set to about several times the cutting frequency of the end mill 31, and the cutting frequency f (Hz) is obtained by changing the rotation speed of the main shaft 32 to R (rp).
m), assuming that the number of teeth of the end mill 31 is N (sheets), f = R × N / 60. By repeating the above, the temporal change of the vibration sensor signal S3 can be measured, and the measurement result is displayed on the display device 47 via the control unit 441, and
It is also sent to the judgment means 443.

Next, a procedure for determining the tool replacement time by the determining means 443 will be described based on a specific example. FIG.
In FIG. 2, a two-blade end mill 31 shown in FIG. 2 was used, the number of revolutions was 500 rpm, the feed rate was 250 mm / min, and the cutting depth was 4
The figure shows a time waveform (sample time 0.5 msec, cut-off frequency of the low-pass filter 53 100 Hz (approximately 6 times the cutting frequency)) when the workpiece 21 made of SS400 is machined. ing. 4A is a time waveform of the vibration sensor signal S3 output from the vibration sensor 51, and FIG. 4B is a time waveform of the vibration sensor signal S3 after passing through the low-pass filter 53. FIG. 4C is a time waveform of the trigger signal S2 output from the control device 41.

[0024] From the capturing means 442 to the determining means 443,
As shown in FIG. 4B, the vibration sensor signal S3 at which the peak values PA and PB are recognized at predetermined intervals is sent. The peak values PA and PB are the cutting edges 31A and 31B of the end mill 31.
Is the vibration when it collides with the work 21 and the end mill 3
Since 1 has two cutting edges 31A and 31B, the peaks of the cutting edges 31A and 31B appear alternately. The determination means is
Based on the vibration sensor signal S3 in FIG. 4 (B), the time change of each of the peak values PA and PB is measured to determine the tool replacement time.

For example, the judgment means 443 determines that the peak value P
As shown in FIG. 5A, when A and PB are changing with little variation, the cutting edge 31 of the end mill 31 is used.
It is determined that no wear, chipping, or the like has occurred on A and 31B, that normal processing has been performed, and that no tool replacement is necessary.
On the other hand, as shown in FIG. 5B, when the variation between the temporal change of the peak value PA and the temporal change of the peak value PB becomes large, one of the two cutting edges 31A and 31B is used. Is determined to be damaged due to wear, chipping, etc., and it is determined that the tool should be replaced.

When the temporal change of the peak values PA and PB is to be determined over a longer period, the number of times the end mill 31 is machined and the variance between the peak values PA and PB are plotted as shown in FIG. The determination unit 443 may determine that the tool should be replaced when the variance becomes a constant value. Based on the determination of the determination unit 443,
A control signal indicating that the tool should be replaced is output from the controller 42 to the control device 41. In addition, the threshold values of the variation and the dispersion, which are the criteria for determining whether or not to change the tool, are appropriately determined in consideration of the material of the work 21, the shape of the end mill 31, the cutting feed speed, and the like.

According to the above-described first embodiment, the following effects can be obtained. That is, disturbance vibrations other than the vibrations of the end mill 31 are removed by the low-pass filter 53, and the trigger signal S of one pulse is generated for one rotation of the end mill 31.
Since the vibration sensor signal S3 is taken in synchronism with 2, the vibration characteristics PA and PB corresponding to the positions of the cutting edges 31A and 31B of the end mill 31 can be clearly detected, and their temporal changes are measured. By doing so, each cutting edge 31
The wear state of A and 31B can be detected with high accuracy. Therefore, by accurately grasping the state of wear of the cutting edges 31A and 31B of the end mill 31, it is possible to make a highly reliable determination of the replacement time of the end mill 31.

Further, since the wear state of the end mill 31 can be accurately grasped, the end mill 31 can be used to the limit, and the machining cost can be reduced without increasing the tool cost. Furthermore, since the system can be constructed simply by connecting the low-pass filter 53, the vibration sensor 51, and the like, the cost required for constructing the system is reduced, and the system configuration is simple.
A system can be easily constructed on the NC machine tool 1. Since disturbance vibration such as noise often appears in the high frequency band, as shown in FIG.
The adoption of 3 is effective in increasing the accuracy of the present system.

Next, a second embodiment of the present invention will be described. In the following description, the description of the same or similar parts or the like as the already described parts or members will be simplified or omitted. In the first embodiment described above, the vibration sensor signal line 63 is provided with the low-pass filter 53 that separates and removes high-frequency components of 100 Hz or more from the vibration sensor signal S3 output from the vibration sensor 51.
On the other hand, in the second embodiment, the low-pass filter 5
3 is 1/2 of the sample frequency in the AD conversion of the AD converter inside the IO extension box 43.
The differences are as follows.

In the first embodiment, the determination means 4 set in the arithmetic unit 44 inside the controller 42 is used.
Reference numeral 43 determines the tool replacement time based on the peak values PA and PB of the vibration sensor signal S3. On the other hand, in the second embodiment, as shown in FIG. 7, the determination unit 543 determines the vibration sensor signal S transmitted from the acquisition unit 442.
3 is subjected to wavelet analysis, and the vibration sensor signal S3
544 for finding wavelet coefficient corresponding to
And a determination unit 545 that determines the tool replacement time based on the temporal change of the wavelet coefficient. When the vibration characteristics are analyzed by the above-described wavelet analysis or FFT analysis, the low-pass filter 53 is used.
Is preferably about 70 db / oct.

As shown in FIG. 7, the determining unit 545 includes a first step 545A, a second step 545B, and a third step 545C. In a first step 545A, an average value of wavelet coefficients included within a certain time is calculated. In the second step 545B, a smoothing curve representing a temporal change of the coefficient average value calculated in the first step 545A is formed. Third step 5
At 45C, the peak of the smoothing curve formed at the second step 545B is changed to the cutting edge 31A of the end mill 31,
It is determined that the wavelet coefficient corresponds to 31B.

Then, the determination means 543 determines whether or not to change the tool using the peak of the smoothing curve as a characteristic value. The other parts have substantially the same structure as the tool replacement time determination system according to the first embodiment described above, and the operation from the output of the control signal S1 to the vibration sensor signal S3 is also substantially the same. , The description of which is omitted.

Next, a procedure for determining the tool replacement time by the above-described determination means 543 will be described based on a specific example.
In FIG. 8, the rotation speed is 500 rpm, the feed speed is 250 mm / min,
Work 2 consisting of SS400, with the cutting depth set to 4 mm
1 and the time waveform (sample time 0.4
69 msec, cut-off frequency 1 of low-pass filter 53
kHz). In FIG. 8, FIG.
(A) shows the time waveform of the vibration sensor signal S3 and the time waveform of the trigger signal S2 after the vibration sensor signal S3 output from the vibration sensor 51 has passed through the low-pass filter 53. FIG. 8B shows this vibration sensor signal S3.
Is subjected to a wavelet analysis for a frequency range of a high frequency range having a high time resolution, and a temporal change W1 of a wavelet coefficient corresponding to the vibration sensor signal S3 is shown.

[0034] From the taking-in means 442 to the determining means 543,
The vibration sensor signal S3 is input in a form as shown in FIG. The analysis unit 544 performs a wavelet analysis on the vibration sensor signal S3 to calculate a temporal change W1 of a wavelet coefficient corresponding to the vibration sensor signal S3. Judgment unit 5
In step 45, first, a first step 545A calculates an average value of wavelet coefficients included in the fixed time T by a moving average method, and a second step 545B forms a smoothing curve SM1 representing a temporal change of the average value of the coefficients. And third
Step 545C determines the positions of the peak values QA and QB of the smoothing curve SM1 by using the cutting edge 31 of the end mill 31.
It is determined that the position corresponds to A, 31B.

The judging means 543 calculates the peak values QA, Q
The time change of B is measured, and the tool replacement time is determined. For example, as shown in FIG.
When B is measured, if a large change occurs in the peak value QB, the determination unit 543 determines that the cutting edge 31B has been damaged, and instructs the control device 41 to change the tool. Similarly, as shown in FIG. 9B, the peak value QA,
When a large variation occurs in the variance between the QBs and a large change occurs in the ratio between the peak values QA and QB, as shown in FIG.
Instruct 1

According to the second embodiment described above, the first
The following effects are obtained in addition to the effects described in the embodiment.
That is, since the determination unit 543 uses the wavelet analysis for the analysis of the vibration sensor signal S3, it is possible to analyze the vibration waveform in a frequency range of a high frequency range with a high time resolution, and to determine the wear state of the end mill 31. It can be detected with extremely high accuracy. Therefore, the judgment means 54
The determination of the tool replacement time by the method 3 is also performed with high accuracy, and extremely reliable determination can be performed.

Further, since the peak values QA and QB of the averaged smoothing curve SM1 are set as the wavelet coefficients corresponding to the cutting edges 31A and 31B, the peak values QA and QB are affected by the change of the peak values due to the intrusion of random disturbance or the like. Without this, it is possible to measure the temporal change of the wavelet coefficient with high accuracy in a stable state. Furthermore, since the cutoff frequency of the low-pass filter 53 is set to be equal to or less than 1/2 of the sample frequency in the AD conversion, it is possible to prevent aliasing in which a frequency component higher than the highest frequency in the FFT analysis appears to be folded below the highest frequency. be able to.

It should be noted that the present invention is not limited to the above embodiments, but includes the following modifications. That is, in each of the above embodiments, the end mill 3
1 is composed of two cutting blades 31A and 31B, but is not limited to this, and as shown in FIG.
End mill 131 having 31A, 131B, 131C
The present invention may be applied to an NC machine tool provided with a.

In this case, when the end mill 131 makes one rotation, the cutting blades 131A, 131B, 131C
Collide three times, the three cutting edges 131A, 131
The peak of the wavelet coefficient occurs in response to the collision between B and 131C, and is represented as a graph as shown in FIGS. 11 (A) and 11 (B). In FIG. 11, FIG. 11 (A) shows the change over time of the wavelet coefficient when the end mill 131 is new, and FIG. 11 (B) shows the change in the wavelet coefficient when the cutting edge 131B is worn due to cumulative use. It is a graph showing a temporal change.

Further, in each of the above-described embodiments, the present invention is used to determine the replacement time of the end mill 31, but the present invention may be used for other rotary tools. further,
In the second embodiment described above, the low-pass filter 53 is used as a filter, but the present invention is not limited to this, and a band-pass filter may be used. In short, what kind of filter is adopted may be appropriately determined according to the frequency band to be analyzed. In addition, specific structures, shapes, and the like at the time of carrying out the present invention may be other structures and the like as long as the object of the present invention can be achieved.

[0041]

According to the system for judging tool replacement time of the present invention, disturbance vibration other than vibration of the rotary tool is removed by the filter, and the vibration sensor signal is taken in synchronization with the rotation of the rotary tool. The vibration characteristics according to the cutting edge position of the rotating tool can be detected, and a highly reliable tool replacement time determination system can be provided.
Since the system can be constructed only by connecting the vibration sensor and the like, the tool replacement time determination system can be configured at low cost.

[Brief description of the drawings]

FIG. 1 is a schematic diagram illustrating a configuration of a tool replacement time determination system according to a first embodiment of the present invention.

FIG. 2 is a front view showing a cutting edge of a rotary tool according to the embodiment.

FIG. 3 is a schematic diagram illustrating a configuration of a capturing unit and a determining unit in the embodiment.

FIG. 4 is a graph showing a vibration waveform of a vibration sensor signal and a trigger signal in the embodiment described above.

FIG. 5 is a graph for explaining a judgment method by a judgment means in the embodiment described above.

FIG. 6 is a graph for explaining another judgment method by the judgment means in the embodiment described above.

FIG. 7 is a schematic diagram illustrating a configuration of an intake unit and a determination unit of a tool replacement time determination system according to a second embodiment of the present invention.

FIG. 8 is a graph showing a vibration waveform of a vibration sensor signal and a wavelet coefficient corresponding to the vibration waveform in the embodiment.

FIG. 9 is a graph for explaining a determination method by a determination unit in the embodiment described above.

FIG. 10 is a front view showing a rotary tool used in a tool replacement time determination system which is a modification of the above-described embodiment.

FIG. 11 is a graph showing a vibration waveform of a vibration sensor signal in a modification of the above-described embodiment.

[Explanation of symbols]

 31 rotating tool 51 vibration sensor 52 trigger signal 53 filter (low-pass filter) 442 capture means 443, 543 determination means 544 analysis unit 545 determination unit 545A first step 545B second step 545C third step QA, QB peak S3 vibration sensor signal SM1 smoothing curve T fixed time

Claims (5)

[Claims] 1. A tool replacement time determination system for measuring the vibration of a rotary tool provided with one or a plurality of cutting blades with a vibration sensor and determining a replacement time of the rotary tool, comprising: an output from the vibration sensor. A filter for separating a predetermined frequency band from the vibration sensor signal to be obtained, a capturing means for synchronizing with a trigger signal of one pulse for one rotation of the rotary tool, and capturing the vibration sensor signal passed through the filter; Determining means for determining a replacement time of the rotary tool from a vibration sensor signal captured by the capturing means. 2. The tool replacement timing determination system according to claim 1, wherein the filter is a low-pass filter that separates and removes a high frequency component of the vibration sensor signal, and the determination unit passes the low-pass filter. A detector for detecting a peak value corresponding to the number of cutting edges of the rotary tool from a vibration waveform for one rotation of the vibration sensor signal, and replacing the rotary tool based on a temporal change in the peak value detected by the detector. A tool replacement timing determination system, comprising: a determination unit configured to determine a timing. 3. The tool replacement time determination system according to claim 1, wherein the determination unit performs a wavelet analysis on the vibration sensor signal passed through the filter, and calculates a wavelet coefficient corresponding to the vibration sensor signal. A tool replacement time determination system, comprising: an analysis unit to be obtained; and a determination unit that determines a replacement time of the rotary tool from a temporal change of the wavelet coefficient. 4. The tool change timing determination system according to claim 3, wherein the determination unit calculates an average value of wavelet coefficients included in a predetermined time, and determines a time of the average value of the coefficient. A second step of forming a smoothing curve representing a temporal change, and a third step of determining a peak of the smoothing curve as a wavelet coefficient at a position corresponding to the cutting edge of the rotary tool. Tool replacement time determination system. 5. The tool replacement time determination system according to claim 3, wherein the vibration sensor signal is A / D converted and taken into the taking means, and the cutoff frequency of the filter is set to the following value. A tool replacement time determination system, wherein the tool replacement time is set to be equal to or less than 1/2 of a sample frequency in AD conversion.