JPS60186353A - Fracture predicting device for tool - Google Patents

Abstract

PURPOSE:To prevent damage of machining material and to improve the economy by deciding the fracture of tool from the number of outbreak component in AE signal to be produced from a working tool thereby enabling replacement of tool immediately before fracture. CONSTITUTION:AE conversion chip provided in AE sensor 6 mounted on the table 4 of machining center 1 will convert AE wave from a drill 3 into AE signal. Here, AE signal as a measuring signal includes single outbreak component and complex outbreak component where plural outbreak components are overlapped. Single outbreak component is processed into a pulse signal of same magnitude with Y1 or the maximum output level of outbreak component through processing circuit 7 while the first outbreak component having the maximum output level of Y2 is processed similarly with the single outbreak component and the second outbreak component having the maximum output level of Y4 is processed into a pulse signal having the output level of Y4-Y3. The number of said AE processed signal is counted and compared with the referential level to decide fracture of tool upon exceeding over the referential level.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a device for predicting damage to a tool used in a machine tool such as a machining center.

Generally, it is often experienced that a sound is produced when an object breaks. It is also known that plastic deformation and microscopic cracking occur repeatedly within the object until it reaches its final destruction, and elastic waves (ultrasonic waves) are generated each time. The phenomenon in which sound, or elastic waves, is generated when a solid deforms or breaks is called the so-called acoustic emission phenomenon (A, E phenomenon), and the elastic waves generated in this way are also known as AE. called a wave.

This AE phenomenon also occurs in machine tools such as machining centers. For example, when using a drill to make a hole in a suitable material, the hole is filled with AE waves due to plastic deformation and destruction that occur inside the drill during operation. arise. However, the AE waves caused by minute cracks and plastic deformation inside the drill cannot be felt by human hearing, and the only thing that can be normally felt is the sound of the final break.

Conventionally, there is a device that uses this AE phenomenon to automatically detect damage to a tool such as a drill without relying on human perception. For example, a transducer such as lead zirconate titanate piezoelectric ceramics (PZT) (hereinafter referred to as an AE transducer)
is placed near the tool, and this converter converts the AE wave from the tool into a so-called AE multiplied signal as an electrical signal.
This is a device that activates an alarm based on a sudden AE double signal that occurs when a tool is damaged.

With this type of device, it is possible to know that a tool has been damaged by an alarm from the alarm, so if the tool is replaced using this as a signal, the work can be continued. However, when replacing a tool after it has been damaged, fragments may remain in the material that was being processed at the time of breakage, and in some cases, the material may be damaged, reducing the value of the product. will be gone. Furthermore, depending on the state of damage, it may take time and effort to replace the tool.

In view of the above points, the present invention maintains economic efficiency by making it possible to use the tool until just before breakage, and prevents damage to the workpiece by replacing the tool before breakage occurs. The purpose is to allow tools to be replaced easily.

This objective can be achieved by predicting tool breakage immediately before it occurs, and the prediction of breakage is determined from the number of sudden components in the AE multiplied signal generated from the tool in operation. The fact that tool damage can be determined from the number of sudden components in the AE multiplied signal was discovered from the results of various experiments conducted by the inventors, and the experiments will be described below.

As shown in FIG. 1, a drill 3 is attached to the head 2 of a machine tool, for example, a machining center 1, and a hole is made with the drill 3 in a material 5 fixed on a table 4.

At that time, an AEE converter (not shown) provided in the AE sensor 6 also mounted on the table 4,
The AE wave from the drill 3 is converted into an AE multiplied signal. In this case, the AE multiplied signal as a measurement signal obtained from the AEE commutator, a single sudden component as shown in Figure 2, and a duplicate sudden component consisting of a state in which multiple sudden components overlap as shown in Figure 3. Contains ingredients. In these figures, the horizontal axis represents time t1, and the vertical axis represents output voltage ■.

-5- The single sudden component is processed by the processing circuit 7 in Fig. 1 into a pulse signal of the same magnitude as the maximum output value of the sudden component, y, (Fig. 2) (see Fig. 4). . On the other hand, the third
Regarding the duplicate sudden components in the figure, the first sudden component whose maximum output is y2 is treated in the same way as the single sudden component described above, and the second sudden component whose maximum output is y is treated as follows. It is processed into a pulse signal whose output value is Y4-Y*, which is obtained by subtracting the residual output y3 of the previous signal from the maximum output y4 (see FIG. 5). Hereinafter, the AE multiplied signal obtained as shown in FIGS. 4 and 5 will be referred to as the AEE multiplied signal.

The AEE signal obtained in this way is the drill 3 (first
This is generated in response to the AE waves from the AE wave shown in Fig.), and the main purpose of this experiment was to count the number of these AEE signals and investigate the relationship between the counted value and drill damage. however,
The AEE signals to be counted are those having an intensity of 0.12 V or more at the processed signal output level in order to remove unnecessary components.

Figure 6 is a graph showing the experimental results, where the horizontal axis shows the number H of holes drilled by the 4-drill 3, and the vertical axis shows AB occurring while drilling the number of holes shown on the horizontal axis. The number of processed signals is AEn. As you can see from this graph,
It can be seen that the number of AEE signals increases rapidly from the point two points before the breakage point a of the drill indicated by the X mark. Therefore, conversely, it can be determined that it is time to break the tool by determining the point in time when the number of AEE processing signals becomes greater than the average value Da of the number of processed signals during normal cutting on the near side of the boundary point S. In this case, how much more than the average value Da should be judged as the failure time can be determined depending on the actual situation, but generally it is 3.8σ (σ: the distribution of the number of processed signals). It is desirable to set the value to about the standard deviation). This is to suppress the probability that the AE treatment band number observed during normal cutting is mistaken for abnormal cutting to 0.01% or less. As a concrete value, when drilling a hole with a depth of 15 mm using a drill with a diameter of 3 holes, - the number of AE treatment bands during the entire stroke, that is, drilling one hole, is 45, therefore, per rotation of the drill. It is preferable to set the number of AE processing signals to about 1.4.

EMBODIMENT OF THE INVENTION Hereinafter, the present invention will be described in detail based on drawings showing embodiments thereof.

FIG. 7 is a circuit diagram of a tool damage prediction device showing an embodiment of the present invention. If damage to the drill 3 shown in FIG. 1 is predicted by the illustrated circuit, A E in FIG.
Sensor 6 serves as an AE sensor, shown with the same reference numerals in FIG. The A and E signals from this AE sensor 6 are
As described above, the signal is converted into a pulse-like AE processing signal (FIGS. 4 and 5) by the processing circuit 7 and then sent to the counter 8. This counter 8 is reset and started by a counter signal S from the microprocessor-9, and sends the counting result to the microprocessor-9. In addition to the count signal C from the counter 8, the microprocessor → No. 9 receives a signal R for confirming one revolution of the spindle of the machine tool 1 (Fig. 1) (i.e., a signal for confirming one rotation of the tool) and a signal from the console IO. P is input.

Further, signals are exchanged with the attached memory 11.

Prior to the tool damage prediction work, the microprocessor 9 stores the cutting take for one drill in the memory 11 according to the flow shown in FIG. That is, the counter 8 reads the number of AB processing signals per rotation of the tool based on the one-rotation confirmation signal 1 (1), and stores the value.Such cutting data storage is performed multiple times, and Based on the stored values, an average value J]a of the number of A-E processed signals is calculated as shown in FIG. 9. This average value Da is equal to the average value Da shown in FIG.

Further, the standard deviation σ of each stored number of signals is taken, and Ds = Da + 3.8σ is set as a reference value. This reference value is the standard value for predicting tool breakage, and when the number of AE processing signals per rotation of the drill for which breakage should be predicted exceeds this value, it is determined that the tool is broken. be.

When the reference value Ds is thus stored in the memory 11, the first
Damage prediction work is performed according to the flow shown in Figure 0. That is, confirm that the drill 3 is rotating, count the number of AE processing signals per rotation, compare the counted value with the reference value Ds, and if the counted value is higher than the reference value Ds. If it is large, a corruption signal Q is generated. Based on this damage signal Q, a warning is issued or the machine tool is stopped and the tool replaced.

As described above, damage to the drill 3 is accurately predicted based on the number of AE processing signals (Figures 4 and 5) obtained based on the AE waves from the drill 3, so damage to the drill 3 can be predicted accurately. It is now possible to replace the drill just beforehand, and as a result, the drill can be used as effectively as possible, and damage to materials due to drill breakage can be prevented.

In the above embodiment, the setting of the reference value Ds, the comparison of the number of AE processing signals and the reference value Ds, etc. are performed based on a predetermined flow programmed in the microprocessor-9. The present invention is not limited to this embodiment, and, for example, the reference value Ds may be set using any method, and a well-known comparison device may be used to compare the number of AE processing signals with the reference value Ds.

As described above, according to the present invention, an elastic wave (AE wave) generated from a tool (drill 3) is converted into an electric signal (AE wave), and a sudden component (AE wave) of a predetermined amplitude of the electric signal is converted.
E processing signal) is counted, and this counted value and the reference value (Ds
= Da + 3.8σ) to predict tool damage, so compared to the conventional case where measures such as tool replacement were taken after the tool was damaged, it was possible to reduce the amount of material to be processed. The advantage is that there is no damage and the tool itself can be replaced easily.

[Brief explanation of the drawing]

FIG. 1 is a front view of an example of a machine tool that uses the failure prediction device according to the present invention, FIG. 2 is a graph showing an example of an AE multiplied signal used in the failure prediction device, and FIG. 3 is a graph showing AP, signals and other signals. A graph showing an example. Figure 4 shows the processed signal club created from the A and E signals in Figure 2. Figure 5 shows the AE in Figure 3.
FIG. 6 is a graph showing the experimental results that form the basis of the present invention. The horizontal axis represents the number of drilled holes, and the vertical axis represents the processing results from the drilling during operation. FIG. 7 is a diagram showing a failure prediction device according to an embodiment of the present invention; FIG. 8 is a flowchart showing an example of programming for data storage in the microprocessor in FIG. 7; FIG. 9 is a flowchart showing an example of programming for creating a reference value in a microprocessor, and FIG. 10 is a flowchart showing an example of programming for predicting damage in a microprocessor. 5...Materials 3...Tools 6...AE
Converter (converter) 8... Counter (counting means) 9... Microprocessor - (comparison means) Procedural amendment 1981 4 j'l 16 F, ] Commissioner of the Japan Patent Office Kazuo Wakasugi 1 Display of case 1981 Patent Application No. 037274 2 Name of the invention Tool damage prediction device 3 Relationship with the person making the amendment f'1 Patent, II provisional (1 person's address is abbreviated name of Takesho) Matsuura Machinery Co., Ltd. Manufacturer 4 Agent Address: 2-1 Nishi-Shinbashi, Minato-ku, Tokyo] 32-4 Kaji Kogyo Building, 1920 Showa Letter of attorney and drawings subject to amendment 6th month/day. 7. Details of the amendments are as shown in the attached sheet. 2-

Claims (1)

[Claims] A transducer is placed near the tool that processes the material and converts the elastic waves generated during processing into electrical signals, and the number of sudden components of the electrical signals from the transducer that exceed a predetermined amplitude are counted. The tool has a counting means for comparing the counted value by the counting means with a predetermined reference value, and detects when the counted value exceeds the reference value by the comparing means. A tool breakage prediction device characterized by predicting breakage of a tool.