JPS60228058A - Device for detecting degree of wear of tool - Google Patents

Abstract

PURPOSE:To automatically detect the degree of wear of a tool without stopping a cutting process, by processing the output of an AE sensor through a band-pass filter, rectifying the output of the filter, and using a data sampling means to send out sampled data as a calculation output indicating the degree of wear of the tool. CONSTITUTION:An AE sensor 7 is secured to the shank 6 of a tool 5. The output of the sensor 7 is amplified by a preamplifier 8 and thereafter removed, by a band-pass filter 9, of a noise produced by a machine vibration or the like. The output of the filter 9 is rectified by a full-wave rectification circuit 10, the output of which is subjected to A/D conversion. A signal generated by the A/D conversion is entered into a microcomputer 12 so that the degree of wear of the tool 5 is determined through a system program and digitally indicated on a wear display unit 13. When the degree of wear has reached a prescribed level, a wear degree limit output is sent out. In order to eliminate the effect of the sporadic AE and take out the level of only the continuous AE correlative to the width of the wear of the flank of the tool, the maximum density AEm as to the probability density function of the full-wave rectification output of the AE is determined. As a result, the degree of wear of the tool is accurately detected even if the form of each chip of a workpiece becomes spiral.

Description

[Detailed Description of the Invention] (Field of the Invention) This development is based on a method for detecting the degree of tool wear in a machine tool by using acoustic emission (hereinafter referred to as ΔF) generated during machining. This invention relates to a tool wear degree detection device.

(Background of the Invention) Conventionally, in a machine tool such as a lathe, in order to detect the degree of wear of the cutting tool, an operator has to observe the tip of the cutting tool under a microscope after the cutting process is finished, or temporarily interrupt the cutting process and check the cutting tool. It is customary to test the tip with a touch sensor.

However, recently flexible Manikov 1 curing
As systems (hereinafter referred to as FMA), factories and automation (hereinafter referred to as FA) become popular,
There is a demand for online detection of the degree of wear of the cutting tool during the cutting process.

Therefore, in Japanese Patent Application No. 59-6747, etc., the present applicant previously proposed an AE sensor installed on or near a tool of a machine tool, and amplified the output of the AE sensor to an appropriate level. an amplifier circuit; a bandpass filter that is provided on the output side of the amplifier circuit and passes frequency components that have a strong correlation with tool wear; and a bandpass filter that rectifies and smoothes the output of the bandpass filter to reduce tool wear. The company has filed an application (unpublished) for a tool wear degree detection device that is equipped with a DC converter circuit that obtains a DC signal at a level corresponding to the tool wear degree, and outputs the output of this DC converter circuit as the tool wear degree.

However, in such a tool wear degree detection device, although the AE amplitude average value signal and the flank wear width of the cutting tool have a strong correlation with the specific gravity under constant cutting regulations, changes in cutting conditions, especially Changes in chip form have a large effect on the average AE amplitude signal, which has the disadvantage of destroying the strong correlation with the wear width of the flank face of the cutting tool.

(Object of the Invention) The object of the present invention is to be able to automatically detect the degree of wear of the cutting tool during the cutting process, that is, without interrupting the cutting process, and to detect the shape of chips. The object of the present invention is to provide a more reliable tool wear degree detection device that eliminates the influence.

(Structure and Effects of the Invention) In order to achieve the above object, the present invention includes an AE sensor attached to a tool or workpiece of a machine tool or the vicinity thereof, and an amplification device that amplifies the output of the AE sensor to an appropriate level. a bandpass filter that is provided on the output side of the amplifier circuit and that avoids passing frequency components that have a strong correlation with wear of the tool; a rectifier circuit that rectifies the output signal of the bandpass filter; and the rectifier circuit. a data sampling means for sampling the output signal at every minute time, and a maximum value of the probability density function of the sampled data;
The present invention is characterized by comprising a screening means for outputting this as a tool wear degree.

According to such a configuration, when applied to a lathe, for example,
In addition to being able to automatically detect the degree of wear of the cutting tool during the cutting process, that is, without interrupting the cutting process, it is also possible to detect the degree of wear of the cutting tool without being affected by changes in the shape of chips. It can be detected with high accuracy.

(Description of Embodiments) FIG. 1 is a block diagram showing a configuration in which the WA device of the present invention is applied to a lathe to detect the degree of wear of a cutting tool.

In the figure, a cylindrical workpiece 1 is a tailstock 2, as is well known.
A cutting tool 5 is fixed to the tool rest 4, and an E sensor 7 is fixed to the shank 6 of the cutting tool 5.

The AE sensor 7 has a relatively wide band of 100 fl-1
It is configured to be able to detect AE of z to 1 MHz.

The preamplifier 8 has a gain of 20 dB in this example.

The band is set to 100 KHz to IMHz, and the output of the AE sensor 7 is amplified to an appropriate level.

The output side of preamplifier 8 has a passband of 100KHz.
A bandpass filter 9 set at ~300KHz, -24dB10ctk is provided, and this bandpass filter 9 cuts noise components associated with mechanical vibrations and the like.

1 extracted through the bandpass filter 9.
The 00KHz to 300KHz component has an averaging time constant of 1
It is rectified by a full-wave rectifier circuit 10 set to m5ec.

The A/D converter 11 converts the output of the full-wave rectifier circuit 10 into A with a sampling period of 0, 1 m5ec, -12 bits.
A/D conversion is performed, and the output of this A/D converter is read into the microcomputer 12.

The microcomputer 12 executes a system program to be described later.

Therefore, the degree of wear of the cutting tool is measured, and this is digitally displayed as a percentage, for example, on the wear trace display device 13, and when the degree of wear of the cutting tool reaches a predetermined limit value, the wear degree limit output is outputted to the outside. do. □Therefore, based on this wear degree limit output, a command can be given to a handling means such as a robot to perform appropriate control such as automatically replacing the cutting tool 6.

Next, the configuration of the system program executed by the microcomputer 12 will be explained with reference to the flowchart shown in FIG.

In the RAM of the microcomputer 12, as shown in FIG. 3, a series of sample data areas x(0),
(1) , X (2> ・-X (N) is set 4F'
E), and these 1 rears can be specified by the rimple data pointer I shown in FIG.

Furthermore, as shown in FIG. 5, the RAM of the microcomputer 12 contains a series of probability density data areas A (,0
), A (1), A (2>...△(K)), and these areas are specified by the value of each 1 number data. The probability density function A(X(1
)) is to be formed as a table.

Further, in FIG. 4, the maximum value register M of the probability density function stores the sample data number I corresponding to the sample data X (I) with the highest probability of occurrence as a result of the performance processing described later.

At J5 in the flowchart in Figure 2, first step (1)
00), the sample data area X (0),
), X(2)...X(N) and probability data area A
(0), A (1), Δ(2)...Δ(K), and set the sample data pointer I to O.
Reset to J.

In the following step (101), the A/D output is sampled at a sampling period T (in this example, 0,1 m5ec), and this is stored in the sample data area x(0), X(1 m5ec).
), X(2>...X(N)).

Note that FIG. 6 shows the contents of the sample data pointer I and each sample data X(1) for an arbitrary full-wave rectified waveform.
It is a graph showing the relationship between

Thereafter, the number data pointer is incremented in step (105), and until the value of the sample data pointer ■ becomes N in step (102>), steps (103) and (
104) is repeatedly executed.

That is, in step (103), the sample data pointer! Lean pull data area specified by x (1)
In the following step (104), +1 is added to the content of the probability data area A (X(I)) corresponding to the value of the read sample data X(1).

In this way, steps (102)-)(103)→
By repeatedly executing (104) → (105) → (102>), each sample data 11 is stored in each probability data area.
The frequency of occurrence of 0.1.2...K is stored and a probability density function Δ(X(I)) is determined.

When the above processing is completed for all sample data, the execution result of step (102) is YI3S,
In the following step (106), the sample data pointer (and the maximum value register M are each reset to O).

In the following steps (107) to (110), probability data areas A (0), A (1), A (2)
...For A (K), the maximum value A (X (M)) is determined among all stored data, and this maximum value A (
The υ sample data number i corresponding to X (M) ) is stored in the maximum value register M.

In this way, when the data number corresponding to the most frequently occurring data is stored in the maximum value register M, the execution result of step (107) is YES, and in the following step (111), the value is It is determined whether or not the threshold value KTH corresponding to the limit wear degree of the cutting tool has been exceeded, and if it is determined that the limit wear degree of the tool bit has been exceeded, a wear limit signal is output in the following step (112). In the following step (113), the contents of the maximum probability density register M corresponding to the wear degree of the cutting tool are sent to the wear degree display device @13 as display data. As a result, the wear degree display device 13 digitally displays the wear degree of the cutting tool during cutting, for example, in %.

Next, the operation of the device of the present invention will be explained with reference to FIGS. 7 to 10. Flank Wear Width of Bit
FIG. 10 shows the relationship between the AE amplitude average value AE (v) and the maximum density AE (In).

Here, the average value AE (V) is calculated from the data obtained in the same row 1 as when 〒(1) was set at the maximum density.

As is clear from this figure, in the case of wear width detection based on the previously applied ΔF amplitude average value ΔI3(v), the A'E average corresponds to the flank wear width until the flank wear width is about 200 μm. The value AE(v) changes linearly with G, lL+, but when the flank wear width exceeds 220 μm, the average AE value AE(V) increases rapidly as shown by the dotted line in the figure, and therefore the flank wear width and If it is based on the correlation with the AE average value AE (V), it is not possible to accurately detect the flank wear width based on the AE average value AE (V).

Here, the present applicant has found that the cause of the correlation being deviated in this way is due to a change in the shape of the chips.

Fig. 7(a), Fig. 8(a) and Fig. 9(a) respectively show the shape of the chips as shown in Fig. 7(C);
9 shows the full-wave rectified waveforms of the F signal when classified into the coil shape shown in FIG. 8(0) and the spiral shape shown in FIG. 9(C).

As shown in Fig. 7(a), in the case of arc-shaped chips, the AE signal amplitude AE (V) reaches a whisker-like peak value every few milliseconds, and this is caused by sudden fluctuations when the chips keep cutting. It is considered to be E.

As shown in Figure 8(a), in the case of coiled chips,
The frequency of occurrence of sudden type F is lower than that of arc type chips,
When the length of the coiled chips becomes long, a considerably high level of AE is generated at the moment of cutting.

In the case of spiral-shaped chips, the level of sudden-type AE is significantly higher than that of arc-shaped and coil-shaped chips. When the chip shape becomes spiral like this, the level of sudden type AE becomes very large, which causes the AE average of 1i11AE (
V) increases rapidly.

In FIG. 10', when the flank wear width is 20 μm, arc-shaped chips are formed, when it is 220 μm or more, spiral-shaped chips are formed, and in other cases, arc-shaped and coil-shaped chips are formed.

On the other hand, for arc-shaped, coil-shaped, and spiral-shaped chips, the maximum density in the probability density function is -(m
) is observed in Figure 7(b).

As shown in FIG. 8(b) and FIG. 9(b), even if the shape of the chips changes from an arc shape or a coil shape to a spiral shape, the maximum density AE (m) is determined by the AE average value AE. (v) shows no extreme fluctuations.

Therefore, in order to remove the influence of sudden type AE and extract only the level of continuous type AE that is correlated with flank wear width, as in the present invention, the full-wave rectified output of AE (AE signal amplitude AE
(v) ) in the probability density function of the maximum density AE (
m), a linear relationship between the flank wear width and the maximum density AE (m) can be maintained even if the chip shape becomes spiral, as shown in Figure 10. Therefore, it is possible to provide a tool wear degree detection device with high accuracy and reliability.

As described above, according to this embodiment, when the wear of the cutting tool progresses in the NC lathe, the degree of progress can be visually displayed, and when a preset limit wear level is reached, Byte exchange output can be emitted.

In addition, since the degree of wear of the cutting tool is detected via AE, the cutting process can be performed without any interruption, and even if the shape of the chips changes from an arc or coil shape to a spiral shape, the cutting tool can be The degree of wear can be detected precisely, and the requirements for FMS, FA, etc. can be fully satisfied.

Although the above description has been made in the case of a lathe, it is of course applicable to various machine tools such as a milling machine, a drilling machine, a grinding machine, and a honing machine.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing an embodiment in which the present invention is applied to detecting the degree of wear of a cutting tool in a lathe, Fig. 2 is a flowchart showing the configuration of a system program executed by the microcomputer of the device of the present invention, and Fig. 3 is a sample. Figure 4 shows the contents of the data storage area; Figure 4 shows the data pointer; the memory block shows the contents of the maximum probability register; Figure 5 shows the contents of the probability data area; Figure 6 shows the contents of the random data area; A graph showing the relationship between the sample data pointer I and each V block data X(1) for wave rectification waveforms. FIG. 7 shows the full wave rectification waveform of AE and the AE average value for arc-shaped chips. An explanatory diagram showing the maximum weight 1 east and the form of chips, Figure 8 is a diagram similar to Figure 7 for coiled chips, and Figure 9 is a diagram similar to Figure 7 for spiral chips. , 10th
The figure is a graph showing a comparison between the previously applied tool wear degree detection method using the AE average value and the tool tool wear degree detection method using the maximum density of the probability density function according to the present invention. 1...Workpiece 2...Shinshindai 3...Chuck 4...Turret 5...Bite 6...Shank 7...E sensor 8...Preamplifier 9...Band pass Filter 10... Full wave rectifier circuit 11... A/D converter 12... Microcomputer 13... Wear mark display device Patent applicant Tateishi Electric Co., Ltd. Figure 3 1 X(1) X(1 )'4 Figure 4 Figure 5 A (X (I)) Figure 7 (.) (b) TIME Fns (C) Figure 8"' (b) (c) TIME ms Figure 9 (0) (b) Figure 10

Claims (1)

[Claims] (1) An AI sensor installed on a tool or workpiece of a machine tool or in the vicinity thereof; An amplifier circuit for amplifying the output of the AE sensor to an appropriate level; An installation number stamped on the output side of the amplifier circuit; , and a bandpass filter that passes a frequency component that has a strong correlation with wear of the tool; a rectifier circuit that rectifies the output signal of the bandpass filter; and a data sampling that samples the output signal of the rectifier circuit every minute time. A tool wear degree detection device comprising: means; and a calculation means for determining the maximum value of the probability density function of the sampled data and outputting this as a tool wear degree.