JPH0659595B2 - Abnormality detection method for drilling - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION (Industrial field of application) The present invention relates to a method for detecting abnormalities in drilling, more specifically, by subdividing abnormal states in drilling and accurately identifying each abnormal state in drilling. It's about what you're trying to detect.

(Prior Art) A machine tool that repeatedly processes a work according to a predetermined program requires an abnormality detection means for detecting an abnormality in the work processing.

Among such abnormality detecting means that can be applied to drilling, as shown in, for example, Japanese Patent Application Laid-Open No. 58-126042, fluctuations in the load applied to the processing tool fall within a set allowable range. The one that detects an abnormality by whether or not it exceeds
Alternatively, as disclosed in Japanese Patent Laid-Open No. 59-156650, there is a method in which a load applied to a processing tool is patterned and an abnormality of the processing tool is detected based on a comparison with a preset reference pattern.

(Problems to be Solved by the Invention) However, since the conventional abnormality detecting means is supposed to detect an abnormality by simply comparing the load detected at the time of processing and the set reference value, it is particularly necessary to make a hole. The processing abnormality detection was inaccurate due to the following factors.

Large variations in load due to variations in work hardness (see Fig. 10). In addition to individual differences, the hardness of this work also includes variations in the systematic hardness of one work.

In addition, variations in the load include bending of the machined hole, runout of the tool, and wear of the cutting edge.

In addition to these factors, if factors such as seizure of the drill, dusting of the chips, etc. are compounded, the conventional means cannot accurately detect the abnormality, resulting in a problem of misjudgment. .

Further, this means that the conventional one has been proposed as a means for roughly classifying the normality and abnormality of the work processing, and it is ineffective for detecting the subdivided abnormal mode. .

By the way, in the drilling process, the countermeasures are different depending on various aspects of the abnormal state such as the bending of the drilled hole, the enlargement of the hole diameter or the life of the drilling tool.

The present invention has been made in view of the above-mentioned circumstances, and an object thereof is to enable appropriate treatment for a larger number of single or multiple abnormal aspects. It is in.

(Means and Actions for Solving Problems) The present invention is based on the premise of abnormality detection of drilling which is repeatedly performed according to a predetermined program, and when the rotation load and the feed load of the processing tool are patterned. , Rotational load pattern,
It is configured by focusing on the fact that various abnormal modes of drilling can be identified by combining the feed load patterns.

Specifically, the step of detecting the rotational load and the feed load applied to the punching tool, the step of patterning the detected values of the rotational load and the feed load, the patterned rotational load pattern and the feed load Forming a composite pattern by combining the patterns, and categorizing an abnormal aspect of the drilling process by comparing the composite pattern with a preset reference classification pattern, and then the rotating load pattern or the feed load And a step of detecting various abnormal aspects of the drilling process by comparing the pattern and a reference identification pattern showing the characteristics of the various abnormal aspects of the drilling process.

(Example) In the following, an example of a case where a long hole such as an oil hole of an engine part (processing length is 1 or 5 times the processing diameter or more) is repeatedly machined according to a predetermined program The detection method will be described.

FIG. 1 shows an outline of an apparatus to which the embodiment is applied.
As in the conventional case, a rotary load sensor 1 for detecting a rotary load applied to a drill (not shown) and a feed load sensor 2 for detecting a feed load associated with the feed movement of the drill are provided.
The load signals S ₁ and S _{2 from 2} are amplified by an amplifier (low-pass filter) 3, converted into digital signals by the A / D converter 4, and then input to the microcomputer 5,
An abnormality detection signal is output from the microcomputer 5 to alarm means (not shown). In this microcomputer 5, various abnormal modes are subdivided and discriminated based on the waveform pattern calculated based on the rotation load signal S ₁ and the feed load signal S ₂ .

Of the abnormal aspects in long hole machining, the most important ones are the enlargement of the diameter near the hole entrance, the bending of the machining hole, the scattering of chips, and the seizure of the drill.The diameter enlargement near the hole entrance mentioned above is This is due to runout of the drill and wear of the cutting edge.
The bending of the drilled holes described above is due to the drill entering the work while bending.

Such a phenomenon is manifested individually or in a mixed manner, and the combination is such that the diameter expansion near the hole inlet (hereinafter referred to as abnormality A) and the bending of the machined hole (hereinafter referred to as abnormality B) are cut. Absorption of powder, seizure of drill (hereinafter referred to as abnormality C) Abnormality A + abnormality B (hereinafter referred to as abnormality A + B) Abnormality A + abnormality C (hereinafter referred to as abnormality A + C) Abnormality B + abnormality C (hereinafter referred to as abnormality B + C) It is possible to classify into seven abnormal modes: abnormal A + abnormal B + abnormal C (hereinafter referred to as abnormal A + B + C).

Of the abnormal modes classified in this way, the rotational load and feed load applied to the drill at the time of machining are actually measured for the basic abnormal modes, which are abnormal A, abnormal B, and abnormal C, and time is taken as the horizontal axis. Each waveform pattern is second when patterned
This is shown in FIGS.

By patterning the value of each load, there is an advantage that the load detection value based on the work hardness can be made uniform, that is, the absolute value of the load detection value in the normal condition can be made uniform. The short parts connected by wavy lines show the normal value pattern.

FIG. 2 corresponds to the abnormality A, and abnormal portions protruding from the normal pattern are shown by a _{1 to} a ₄ , and the basic pattern shown in this figure is determined from the actually measured pattern. It is understood that the characteristic of the abnormality A is that the abnormal portions a _{1 to} a _{4 in} both the rotation load and the feed load are descending to the right (the values become smaller with the passage of time, that is, as the machining progresses). It is considered that this phenomenon shows large values in both the rotational load and the feed load in the initial stage of machining, because the drilling position and the wear of the cutting edge do not determine the drilling position of the drill tip with respect to the work.

FIG. 3 corresponds to the abnormality B, and abnormal portions protruding from the normal pattern are shown by b _{1 to} b ₃ , and the basic pattern shown in this figure is determined from the actually measured pattern. The feature of this abnormality B is that the rotational load is
While abnormal portion b ₁ ~b ₃ is right up (value increases with time), no abnormality is observed in the transportation load. This phenomenon is that the shape of the drill is generally thickened so that only the cutting edge of the drill hits the work piece. . From this, it is considered that the rotational load shows a larger value toward the latter stage of processing.

FIG. 4 corresponds to the abnormality C, and abnormal portions protruding from the normal pattern are shown by C _{1 to} C ₃ , and the basic pattern shown in this figure is determined from the actually measured pattern. It is understood that the characteristic of the abnormal C is that the rotational load and the feed load are both upper right, but the abnormal portions C _{1 to} C ₃ are seen after the midpoint of the machining. It is considered that this phenomenon is caused by baking when the tip of the drill penetrates deeply into the work and the chips are not sufficiently discharged.

The above three basic abnormal modes (abnormal A, abnormal B, abnormal C)
The waveform patterns in FIG. 5 are combined to create a composite pattern in the abnormal mode, which is shown in FIGS. 5 to 8. FIG. 5 corresponds to abnormality A + B,
The figure shows abnormality A + C, FIG. 7 shows abnormality B + C, and FIG. 8 shows abnormality A.
Corresponds to + B + C.

When the rotational load pattern and the feed load pattern in various abnormal modes obtained in this way were combined and examined, each abnormal mode had a characteristic pattern. Therefore, the abnormal mode can be determined.

That is, for example, when considering a pattern in which the feed load pattern is subtracted from the rotational load pattern, the pattern including the abnormality B (FIG. 3, FIG. 5, FIG. 7, FIG. 8) is always in the upper right corner. Since it shows a waveform, this causes abnormal B
It will be possible to classify whether or not it is an abnormality including.
Then, in the abnormal mode including the abnormal condition B, the respective feed load patterns show clearly different waveforms, respectively, whereby the abnormal mode can be individualized.

In addition, those that do not include abnormality B (see FIGS. 2, 4, and 6)
Even in the figure), the respective feed load patterns respectively show clearly different waveforms, which allows individualization of each abnormal state. In other words, a complex abnormality mode (abnormal A + B, abnormality A + C, abnormality B + C, abnormality A) is created by an abnormality pattern created by combining load patterns in basic abnormality modes (abnormal A, abnormality B, abnormality C).
+ B + C) can also be detected.

Such an abnormality detection method will be specifically described based on a flowchart.

First, the storage unit of the microcomputer 5 stores various preset reference patterns as a reference for determining an abnormal state.

That is, in order to classify whether or not the abnormality B is included, in the basic pattern shown in FIG. 3, the reference abnormality B-containing classification pattern (P0) obtained by subtracting the feed load pattern from the rotational load pattern and the various abnormality modes are individually To detect
The feed load patterns in the abnormalities A, C, A + B, A + C, B + C, and A + B + C are the reference abnormal A identification patterns (P
A), reference abnormality C identification pattern (PC), reference abnormality A +
B identification pattern (PA + B), reference abnormality A + C identification pattern (PA + C), reference abnormality B + C identification pattern (PB
+ C), reference abnormality A + B + C identification pattern (PA + B +
C) and are set and stored. With respect to the waveforms of these reference identification patterns, detailed description will be omitted by assigning symbols such as PA and PC to FIG. 2 and the like.

In FIG. 9, first, in step S1, the rotation load signal S
After the data based on ₁ and the feed load signal S ₂ is fetched, the rotation load data and the feed load data are multiplied by a correction constant in step S2 to perform patterning. Here, the correction constant is set according to the hardness of the work. Then, in the next step S3, after the subtraction process of subtracting the feed load detection pattern PS from the rotational load detection pattern PT is performed, this (PT-PS) and the reference abnormal B-containing classification pattern (P0)
Is compared (step S4), and (PT-PS) ≧
If it is determined to be P0, it is determined that the abnormality B is included, and the determination routine is further divided.

That is, the feed load detection pattern PS patterned in step S2 described above, the reference abnormality identification A + B pattern (PA + B), and the reference abnormality B + C identification pattern (PB +).
C), reference abnormality A + B + C identification pattern (PA + B +
C) is made in steps S5, S6, and S7, and the feed load detection pattern PS is one of the reference identification patterns PA + B, PB + C, PA + B + C.
If it is determined that the above is the case, abnormality A + B, abnormality B +
Assuming that C or abnormality A + B + C, abnormality detection signals are output (step S8 to step S1).
0).

On the contrary, if it is determined in any of steps S5 to S7 that the answer is "No", that is, the feed load detection pattern PS is smaller than the reference identification patterns PA + B, PB + C, and PA + B + C, it is determined as an abnormality B, and step S11 is performed.
The abnormal B detection signal is output at.

On the other hand, if it is determined in step S4 that (PT-PS) is smaller than the classification pattern P0, it is determined that the abnormal state does not include the abnormality B, that is, whether there is an abnormality A, an abnormality C, or an abnormality A + C. The determination is made in steps S12 to S14.

That is, in steps S12 to S14, the feed load detection pattern PS is compared with the reference abnormality A identification pattern (PA), the reference abnormality C identification pattern (PC), and the reference abnormality A + C identification pattern (PA + C) to determine the feed load. When it is determined that the detection pattern PS is equal to or larger than any of the reference identification patterns PA, PC, and PA + C, it is determined that the abnormality is A, the abnormality C, and the abnormality A + C.
An abnormality detection signal is output (step S15 to step S17). Then, based on each abnormality detection signal, a display lamp that displays each abnormality mode is turned on, or the operation of the machine is stopped, a drill replacement alarm lamp is turned on, and the like.

On the contrary, if it is determined in any of steps S12 to S14 that the result is “NO”, that is, the feed load detection pattern PS is smaller than the reference identification patterns PA, PC, and PA + C, it is determined that the machining state is normal. Step S1
After the determination of reset is made in 8, the loop is rotated to step S1 and the above-described presence / absence of abnormality and the detection of the abnormality mode are repeated for each machining of the work.

As described above, since abnormalities of various types are detected by finely classifying abnormalities in work machining, it is possible to accurately grasp the state of wear of the drill or the state of machined holes, which causes the abnormalities. Therefore, the drill can be replaced or the processed hole can be appropriately corrected.

Although one embodiment has been described above, the present invention can also be applied to detection of the life of a drill. For example, it can be determined whether or not the rotational load detection pattern PT and the feed load detection pattern PS are larger than the life set value.

(Effects of the Invention) As is apparent from the above description, according to the present invention, it is possible to accurately detect various abnormal modes among the abnormalities of the drilling process. The abnormal condition of the machined hole can be accurately grasped, and appropriate measures can be taken for a larger number of individual or complex abnormalities.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram of an apparatus used in an embodiment of the present invention, FIG. 2 is a diagram showing a pattern of abnormality A, FIG. 3 is a diagram showing a pattern of abnormality B, and FIG. FIG. 5 is a diagram showing patterns of abnormalities A + B, FIG. 6 is a diagram showing patterns of abnormalities A + C, FIG. 7 is a diagram showing patterns of abnormalities B + C, and FIG. 8 is a diagram of patterns of abnormalities A + B + C. FIG. 9, FIG. 9 is a flowchart of abnormality detection, and FIG. 10 is a diagram showing conventional problems. 1: Rotational load sensor 2: Feeding load sensor 5: Microcomputer S ₁ : Rotational load signal S ₂ : Feeding load signal PT: Rotational load detection pattern PS: Feeding load detection pattern P0: Reference abnormality B content classification pattern PA: Reference abnormality A identification pattern PC: reference abnormality C identification pattern PA + B: reference abnormality A + B identification pattern PB + C: reference abnormality B + C identification pattern PA + B + C: reference abnormality A + B + C identification pattern

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Yasuo Nishimori 3-3 Shinchi, Fuchu-cho, Aki-gun, Hiroshima Mazda Co., Ltd. 44255 (JP, A) JP 59-107843 (JP, A)

Claims (1)

[Claims] 1. A method for detecting an abnormality in a drilling process for repeatedly detecting an abnormality in a drilling process according to a predetermined program, wherein a rotary load and a feed load applied to a drilling tool are detected, and the rotary load is detected. And a detection value of the feed load is patterned, and the patterned rotation load pattern and the feed load pattern are combined to form a composite pattern, and a comparison is made between the composite pattern and a preset reference classification pattern. Various abnormal modes of drilling are detected by classifying abnormal modes of machining, and then comparing the rotational load pattern or the feed load pattern with a reference identification pattern showing characteristics of various abnormal modes of drilling. An abnormality detection method for drilling, which is characterized by: