JP2533971B2 - Tool abnormality detection device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a detection device for detecting an abnormality in each machining tool in a machine tool that performs machining using a plurality of machining tools.

[Prior art and its problems] Machine tools equipped with multiple machining tools, such as machining centers and NC lathes, sequentially select tools based on a preset machining program, and perform machining in accordance with the machining conditions of that machining program. To execute.

In such machining, between different types of tools,
The machining load naturally varies depending on the machining shape and machining conditions, but even with a single tool, when the tool is machining multiple machining points based on the machining program, the machining shape and machining conditions for each machining stage are different. The processing conditions are different and the processing load changes.

Conventionally, in this type of machine tool, as a method of determining a tool abnormality from a machining load, a threshold value for abnormality determination is set in advance for each machining tool, and the tool is changed every time the tool is replaced. Select the threshold value for the replaced tool by the number signal, compare the threshold value with the machining load,
A method is used to determine the abnormality of the tool.

However, in this conventional method, since only one threshold value is set for a single tool, in order to adapt the threshold value to a large number of machining stages, the maximum machining load in all stages is required. It is necessary to adjust the threshold to. However, this has a drawback that the set threshold value does not function for judgment in the machining stage under light load, and sufficient tool monitoring cannot be performed.

In addition, when the machining conditions such as the change of the machining shape and the change of the tool rotation speed are changed during the machining while monitoring the threshold value and the machining load, the change of the machining load caused by the change can be handled accurately. Unable to set threshold,
There is a drawback that a reliable abnormality determination cannot be performed.

For example, as shown in FIG. 3, when the workpiece 21 is machined at a plurality of locations with a single machining tool 20, if the shape of the workpiece 21 changes due to the inclined surface 22, the machining load changes, Although it is necessary to change the threshold value as well, in the conventional method, since exactly the same threshold value is set at each processing location, there is a problem that proper abnormality determination cannot be performed.

Furthermore, in actual machining, periodic or aperiodic fluctuations in machining load occur, but the magnitude of the load itself does not change significantly. However, in such a case, there is a drawback that the conventional method of comparing the machining load with a fixed threshold value cannot stably detect the tool abnormality.

That is, in order to discriminate the above-mentioned tool abnormality and noise occurrence, it is necessary to have a function that can detect the amount of fluctuation peculiar to the machining load and a function that can accurately detect only the increase in the load due to tool wear. The method of detecting an abnormality by simply turning the load on and off with respect to the threshold value of the load cannot accurately detect the variation amount or the increase amount of the processing load, which causes a problem that a malfunction easily occurs.

The present invention has been made in view of the above-mentioned drawbacks, and an object thereof is to determine a threshold value corresponding to a machining load for each machining stage for each machining tool, and to accurately detect an abnormality of the tool. An object of the present invention is to provide an anomaly detection device that is capable of accurately detecting the amount of change in machining load and the amount of change in load associated with an abnormality in a tool, and that can perform stable tool wear management.

[Means for solving the problem]

In order to achieve the above-mentioned first object, the present invention corresponds to a machining load detecting means for detecting a load during machining by a machining tool, detection start time data and detection time data for each machining stage, and these. And storage means for storing data for determining a tool life determination threshold value. The threshold value determination data is a reference value specified for each machining tool and each stage. The coefficient data is used to determine the threshold value specified in 1., and the coefficient data is a plurality of three or more kinds for detecting at least the upper limit value of the contact detection between the tool and the workpiece, the upper limit value of the tool breakage, and the lower limit value. Corresponding to the threshold value data from the detection time data stored in the storage means, including threshold value determining means for determining the threshold value data by multiplying the reference value and the coefficient data, respectively. Of the tool by comparing the threshold value determined by the threshold value determination means with the output value of the processing load detection means within the monitoring time set by the monitoring time data for the machining tool drawn out The tool abnormality detecting device for a machine tool is provided with a comparing and determining means for selecting a machining tool based on a machining program to perform machining.

Further, according to the present invention, as a second means, the processing load detecting means in the above-mentioned structure is provided with a waveform processing circuit for performing waveform processing of a load signal at the time of processing, and the storage means has a threshold value corresponding to the waveform processing. We adopted a structure that stores data.

[Action]

In the means for achieving the first object, when the comparison / determination means receives an external signal such as a tool number or a workpiece number from a machine tool or the like, the machining stage to be detected is determined from the signal, and the machining stage The corresponding detection data is retrieved from the storage means. When the machine tool starts machining, the threshold value and the machining load are compared only during the detection time after the detection start time has elapsed.

In the above structure, since the detection period for the machining load is specified by time, the detection time can be exactly matched with the period during which the tool is actually machining the workpiece in the machine tool.

Further, since the threshold value can be set for each processing stage, the threshold value can be set appropriately corresponding to the processing load of each stage.

In the above structure, since the threshold value data is composed of the reference value and the coefficient data, if the threshold value reference value is determined for each tool, the reference value and the coefficient data can be multiplied. Thus, it is possible to easily set a plurality of threshold values for each processing stage. Therefore, the work of setting a plurality of different threshold values for each processing stage can be omitted.

Also, since multiple threshold data can be set internally, turning on / off the machining load for multiple thresholds
It is possible to detect OFF by shifting the time by appropriately setting the monitoring time of each threshold value, and by combining the detection order, it is possible to detect the variation amount of the machining load over time. Therefore, it is possible to accurately determine the irregular fluctuation of the processing load.

Also, in the second means, for the load signal at the time of machining,
Waveform processing such as integration processing, maximum value detection, and vibration component extraction is performed, and the processed signal is compared with a threshold value. By performing the waveform processing of the load signal in this way, it is possible to accurately detect only the increment or variation of the machining load due to tool wear without being affected by the instantaneous change of the machining load.

〔Example〕

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

As shown in FIG. 1, the abnormality detection device 1 of the embodiment includes a processing load detection means 2 connected to a drive motor 9 of a machine tool.
And a microcomputer (CPU) 3 functioning as a comparison / determination means, and a storage means 4 connected to the CPU 3.

The drive motor 9 described above is intended for an electric motor involved in machining of a machine tool, and a tool, a spindle rotation motor, a feed shaft drive motor, or the like is used.

The processing load detection device 2 continuously detects power consumption of the drive motor 9 during processing, and for example, a power detector or a current detector is used.

The processing load detecting means 2 is connected to a filter circuit 5 including a circuit for removing noise components included in power consumption and a circuit for removing only power increase during dream load (when the motor is idling). The processing power detected by the detection means 2 is input to the CPU 3 after being actually extracted as the power resulting from the processing.

A numerical control device 10 (hereinafter referred to as an NC device) of a machine tool and a storage means 4 are connected to the CPU3 so that they can be transmitted and received.
Is output to a drive circuit 12 that drives an electric motor 13 for operating a machine tool and an alarm device 14 via an interface 11. The CPU 3 also includes a memory unit 8 that receives and stores the data signal from the storage unit 4.

The storage means 4 is composed of a stroke number data file 6 for determining a stroke number from a workpiece number and a tool number, and a detection information data file 7 for obtaining detection information at each machining stage from the stroke number.

In the stroke number data file 6, as shown in Table 1, the stroke numbers are set in association with the workpiece numbers and the tool numbers. This process number designates all the processing steps of the processing stage executed by each processing tool based on the processing program for an arbitrary workpiece.

On the other hand, in the detection information data file 7, as shown in Table 2, detection data is stored for each process number in association with each processing stage of the process.

In this file, the detection data corresponding to each stage executed by each tool in each stroke is stored. When the CPU3 specifies the stroke number and the stage number, the detection data corresponding to that stage number is stored in the CPU3. It can be pulled out to the memory section 8 of.

The above detection data is composed of a tool life determination threshold value and time data that determines the detection timing and the detection time. Among these, the tool life determination threshold value has a predetermined coefficient for the machining load. It is set by crossing. This setting is performed by actually performing trial machining of the workpiece and determining the machining load from the consumed power.

The threshold value can be set in plural by changing the coefficient value variously. By setting a plurality of threshold values in this way, it is possible to make comparative judgments simultaneously for a plurality of detection targets such as steady wear of tools, small chipping, large-scale damage such as breakage, and there is an advantage that fine judgments can be made. .

The above time data is composed of the detection start time (timer T1) and the detection time (timer T2). As shown in FIG. 2, these both times are set such that only the actual machining power when the tool actually cuts into the workpiece to perform machining is within the detection range. That is, the detection time (timer T2) is set within a range from when the machining power starts to rise due to the start of machining to just before the machining is finished and falls to the no-load state. Also, the detection start time (timer T1) is from the time when the detection time (timer T2) of the previous stage has ended (in the case of the first machining stage, from the time of the machining start in the machining program or the time of the tool change command). , Set the range until the processing of the next stage is started.

By setting the detection timing in this way, the actual processing and the detection time can be accurately matched. This is because in a machine tool with an NC device in which machine operations such as tool change and spindle rotation are controlled based on the machining program, the intervals and execution periods of each stroke and each stage are accurately determined by the machining program. Therefore, the detection time can be matched with the actual machining by adjusting the value of the timer to the interval and the execution period by this program.

The setting of such time data (timer T1 and timer T2) can be calculated from the machining program stored in the NC unit 10, or the trial machining of the workpiece can be performed similarly to the setting of the threshold value, and the machining power can be set. There is a method of sampling over time and calculating from the sampling result.

The detection device of this embodiment has the above-mentioned structure, and its operation will be described below.

When the tool is replaced or the workpiece is changed in the machine tool, the tool number of the tool newly selected from the NC device 10 or the target workpiece number is output to the CPU 3 alone or simultaneously.

Upon receiving this signal, the CPU 3 specifies the tool number or the workpiece number in the stroke number data file 6 of the storage means 4 to select the corresponding stroke number, and the selected stroke number is detected information data file 7 Is specified, and all the detection data corresponding to it is read and stored in the memory unit 8. This series of operations is performed while changing tools or converting a workpiece.

In the machine tool, when the machining of the first stage is started by the newly selected tool, the CPU 3 simultaneously starts monitoring the machining of the first stage. In this monitoring, after the processing power detected by the processing load detection means 2 is input to the CPU 3 for T1 time, the processing power is compared with the threshold value set by the time data for T2 time. It is performed by executing up to.

When the time T2 has elapsed, the CPU 3 determines that the monitoring of the first stage has ended, and continues to monitor the second stage. In this case, the threshold value and the timer value in the memory unit 8 of the CPU 3 are switched to the second stage data in the detection information data file. As a result, similarly to the case of the first stage, the comparison between the threshold value and the processing power is performed during the time T2 after the time T1 has elapsed. When the monitoring of the second stage is completed, the monitoring of each stage is successively performed in order of the third stage and the fourth stage.

When all the machining stages for one tool are completed, the tool number of the next tool selected by the tool change command signal is input from the NC device 10 to the CPU 3. When this signal is input, the CPU 3 extracts the stroke number and the detection data corresponding thereto based on the tool number selected from the storage means 4. As a result, the threshold value and the timer value in the memory unit 8 of the CPU 3 are switched to the detection data of the new stroke number, and the CPU 3 monitors the machining power based on the detection data.

The above operation is similarly performed even when the workpiece is changed. That is, in this case, the workpiece number signal of the changed workpiece from the NC device 10 and the number signal of the tool that first executes the machining are input to the CPU 3, and the CPU 3 is based on the workpiece signal and the tool number. The stroke number and the detection data are selected from the storage means 4 and the monitoring is executed.

In this way, for each machining stage of various workpieces and all machining tools, the machining load and the threshold value are compared only during the actual machining time, and the abnormality of the tool is detected.

The determination of the tool abnormality by the CPU 3 is performed by determining the form of the abnormality at the time when the machining power exceeds the threshold value in the above detection, based on the type and number of times the threshold value is exceeded. For example, when it is determined that an emergency such as breakage of the cutting edge is required, a signal is output to the drive circuit 12 to control the operating electric motor 13 to change the number of revolutions of the machine tool and the cutting amount, or an alarm device. Activate 14.

Further, when it is determined that an emergency such as normal wear is not required, the number of times the processing power exceeds the threshold value is stored, and a signal is output when the accumulated number exceeds a certain amount.

Although the stroke number is selected according to the workpiece number or the tool number output from the NC device 10 in the above embodiment, the stroke number may be directly specified to the CPU 3 by an external signal such as a binary code.

Further, although the same number of stages is set for each stroke number in the detection information data file 7, the number of stages does not necessarily have to be the same, and the number can be increased or decreased according to the number of processing locations of each tool. it can.

 4 to 10 show another embodiment.

As shown in FIG. 4, the abnormality detecting device 31 of this example comprises a processing load detecting means 32, a storing means 33, and a comparing and judging means 34, and its basic configuration is the abnormality detecting device 1 described above. Is the same as. However, the processing load detection means 32 is provided with a circuit for performing waveform processing such as integration and maximum value detection on the processing power signal, and the storage means 33 stores the input waveform other than the stroke number data and the detection information data. There is a difference in that it stores data for determining types, data for special functions, etc., and due to this difference, the control function of the comparison and determination means 34 also
A function different from that of the abnormality detection device 1 described above is provided.

First, the processing load detecting means 32 holds a sampling circuit 37 for holding a non-negative power f (to) from the power consumption P (t) obtained through the power detector 35 and the noise removal filter 36, and the sampling circuit 37. An arithmetic circuit 38 for calculating the machining power f (t) by subtracting the load power f (to) from the power consumption P (t) is provided, and the arithmetic circuit 38 is provided with a signal of the obtained machining power f (t). On the other hand, a plurality of circuits that perform waveform processing are connected to each other. The waveform processing circuit includes an integrating circuit 39 that performs an integrating process on the waveform of the processing power f (t), a maximum value detecting circuit 40 that detects a maximum value, and an oscillating wave that extracts an oscillating wave component. A detection circuit 41 and a detection circuit 42 that detects a stable region of processing power are provided.

The integrator circuit 39 and the maximum value detection circuit 40, in response to the start and end of processing, when one of the signal input from the outside or the signal output by detecting the processing power inside is input. A gate circuit 43 that opens and closes is provided, and waveform processing is performed accurately only during the processing period.

As shown in FIG. 5, the vibration wave detection circuit 41 includes a vibration wave extraction circuit 47 including a high pass circuit 44, a band pass filter 45 and a full wave smoothing circuit 46, an integration circuit 48, a vibration wave extraction circuit 47 and an integration circuit. Timing circuit that opens and closes the gate between circuits 48
It is configured by connecting with 49.

In the above structure, the extraction circuit 47 has the structure shown in FIG.
As shown in, the high-pass circuit 44 removes the DC component from the processing power, and the extracted vibration wave component S ₁ changes the band-pass frequency of the band-pass filter 45 as shown in FIG. 6 (b). Oscillation wave of required frequency (S ₂ , S ₃ )
Is extracted, and its oscillating wave is subjected to full-wave smoothing and output.

Then, the integration circuit 48 performs integration processing ∫S on the vibration wave output from the extraction circuit 47 within the range of the period set by the timing circuit 49 according to the processing time. The output waveform in that case is as shown in FIG. 6 (c).

On the other hand, the processing power stable region detection circuit 42 makes a comparison determination with a predetermined threshold value when the processing power is changed drastically and stable comparison determination cannot be performed with a constant threshold value, such as in the initial stage of drilling. Is to detect the stable point of the processing power that can be performed and to determine an appropriate value for the threshold value.

As shown in FIG. 7, the detection circuit 42 includes a gate circuit 50 that opens and closes according to the processing period, and an integration circuit 51 that integrates the signal waveform of the processing power. An average value calculation circuit 53 for calculating the average value f (t ₂ ) = Σ∫f (t) / N of the integrated value from the average number N obtained as the number of times of processing and the total sum of the integrated values input from the integrating circuit 51. Are connected. In addition, in this average value calculation circuit 53,
The register 54 is connected to the comparison circuit 55, and the comparison circuit 55 is connected to the count circuit 56.

In the above structure, the average value calculation circuit 53 counts the signals input from the counting circuit 52, and the average number (N) is counted every time the count number reaches a preset average number (N) of 2 or more. While counting, the total sum Σf (t) of the input integrated values is divided by the average number of times (N), and the calculated average value is output to the register 54 and the comparison circuit 55.

The register 54 is a signal of the average value from the arithmetic circuit 53 (if this is the signal calculated for the n-th time, f (t ₂ ) _n )
Is input, it is stored once and the next average value (f
When (t ₂ ) _{n + 1} ) is input, the previously stored average value (f
(T ₂ ) _n ) is sent to the comparison circuit 55, and the average value (f (t ₂ ) _{n + 1} ) input by replacing it is stored. Then, this operation is repeatedly executed each time a signal is input from the average value calculation circuit 53.

The comparison circuit 55 sequentially stores the average values f (t ₂ ) _n and f (t ₂ ) _{n + 1} input from the average value calculation circuit 53, and the latest average value f (t ₂ ) _{n + 1} , Absolute value of the value obtained by subtracting the average value f (t ₂ ) _n before input from the register 54 | a | (= |
f (t) _{n + 1} −f (t) _n |) is compared with a preset allowable value b, and when the absolute value | a | falls below the allowable value b, a signal is output to the counting circuit 56. To do.

The count circuit 56 is set and stored with a continuous number (M) of 2 or more in advance, and outputs a signal to the comparison circuit 55 when the signal input from the comparison circuit 55 reaches the continuous number (M).

When the calibration signal is input from the count circuit 56, the comparison circuit 55 receives the average value f stored in the storage unit at that time.
(T) _{n + 1} is output to the CPU 62 to calibrate the reference value of the threshold.

Thus, at the time when the average value for calibration is output from the comparison circuit 55, the fluctuation of the average value of the machining power is within a certain range, and the rapid fluctuation of the machining power is within the stable fluctuation region. Therefore, in the CPU 62, by making the average value the reference value of the threshold value at the end, stable comparison determination can be performed.

In the above example, the integration circuit 51 is unique to the detection circuit 42.
Although the structure provided with is shown, the signal of the integrating circuit 39 in FIG. 4 can be used. Further, the maximum value f (t) max of the processing power can be used as the signal for detecting the stable region.

As shown in FIG. 4, the storage means 33 is composed of five data tables of a stroke table 57, a mode table 58, a stage table 59, a channel table 60, and a special function table 61.

As shown in Table 3, the stroke table 57 is specified in correspondence with each workpiece number (O number) and each tool number (T number), as in the stroke number data file 6 of the above-described embodiment. The stroke conditions are stored, but the stored stroke conditions are not a simple stroke number but a plurality of data are integrally set.

That is, in the stroke table 57, the mode numbers (the numbers of the mode table 58 in which the detection conditions corresponding to the processing conditions are stored) and the thresholds are used on the matrix selected by the arbitrary O number and T number. Value reference value (100% load power value when the workpiece is processed by the tool specified by the T number, which is the reference power for each SET value (threshold value) to be described later), detection start stage No and end stage number (number of stage table 59 to set detection time and threshold value), presence or absence of repeat function (function to repeat defined stage number), and data indicating presence or absence of special function are stored. By selecting a specific OT number, all the above-mentioned data can be read by the CPU 62.

On the other hand, as shown in Table 4, in the mode table 58,
Various conditions necessary for abnormality detection are stored for each mode number, and as described above, when the mode number is selected according to the stroke condition of the stroke table, the various conditions are read simultaneously.

The conditions stored in this mode table 58 include data for selecting the threshold value and the type of machining power (power consumption, machining power, integral value, maximum value) to be compared with the threshold value, and obstacles when detecting tool wear. There is data such as a rush cut time for cutting the power rise at the time of starting the motor, and a hold delay time and a correction value when sampling the no-load power.

In addition, data for determining whether to start the processing stage for monitoring the processing period from the outside, internally detect the processing power level to instruct, or constantly monitor during start-up, and four data described later. Threshold (SET1 to SET4)
Regarding the over detection (upper limit value) or under detection (lower limit value) command, or data for setting the monitoring time from OFF to ON, and other input gain of the current sensor of the power detector 35. Switching data and the like are stored in the mode table 58.

Also, stage table 59 and channel table 60
Corresponds to the detection information data file 7 of the above-mentioned embodiment, but its internal structure is adapted to accommodate a wider range of data amount that can be stored. That is, the stage table 59 stores the detection start time (timer T ₁ ), the detection time (timer T ₂ ), and the channel number for each stage number, and the channel table 60 stores the channel number. Each time, with four levels (SET1 to SET4) as one set, coefficient data 40, 120, 130, ... for setting the threshold value, and monitoring time data (timer for defining each threshold monitoring period) ST) is stored.

The above coefficient data indicates a percentage value with respect to the reference value of the stroke table 57, and by multiplying the coefficient of each channel number by the reference value of the threshold value set in the stroke table 57, each machining stage 4 different thresholds can be set. Therefore,
In this embodiment, if the load power for actually performing trial machining of the workpiece and determining the index is obtained for each tool, the coefficient data of the channel table 60 can be set appropriately for all machining stages. On the other hand, a plurality of thresholds can be set. Therefore, it is not necessary to set a plurality of different thresholds for each processing stage, and the data setting work can be greatly shortened. Also, there is an advantage that the size of the threshold value can be easily changed by changing the coefficient. The four threshold values (SET1 ~ SET
4) is usually the detection of the contact between the tool and the workpiece (upper limit),
It is set as detection of tool wear (lower limit value), detection of tool breakage (upper limit value), and detection of tool breakage (lower limit value).

In the channel table 60, the monitoring time (timer ST) set corresponding to the coefficient data is used in the special function described later, and therefore the description thereof will be made together with the corresponding special function.

The repeat function set in the stroke table 57 is a function for repeatedly monitoring the machining power under the condition of the same stage, and has an effect of reducing the number of stages to be set.

That is, in the apparatus of the embodiment, the machining power is sequentially monitored by the number of stages designated by the start stage No. and the end stage No. of the stroke table 57, but when the machining power to be monitored does not fluctuate much due to continuous machining, , It is useless to set the stage No. by the number of machining. In this case, by specifying consecutive machining stage No. and setting the repeat function in the stroke table,
The same monitoring operation is repeated for the number of stages.

On the other hand, as special functions similarly set in the stroke table 57, the apparatus of this embodiment is provided with three functions of an automatic calculation function, a vibration wave detection function, and a sequential determination function.

The automatic calculation function is based on the processing power stable area detection circuit described above.
This is performed by using the 42, and when the same type of product is repeatedly machined with one tool, and the initial processing power and the processing power during the process of changing the tool are different, the processing power is stabilized from several times of processing. Is detected to obtain a reference value.

As described above, the function of this function is the predetermined average number N
When the average value of the processing electric power in step S3 is less than the allowable value by the number of consecutive times M, it is determined that the processing power is stable, the last average value is stored as a reference value, and the value of the average value is stored in the stroke table 57. The reference value set in is automatically calibrated.

In this case, by appropriately setting conditions such as the average number of times and the number of consecutive times, it is possible to set a certain point of the machining power in the trial machining as 100% load power and set it as a reference value of the threshold value.

This automatic calculation function sets whether or not to use for each O number and T number defined in the stroke table 57, and when it is used, the electric power at any time during machining for determining the reference value is set to 100. Set the time from the start of machining that determines whether to feed back as% load power for each workpiece and tool.

The vibration wave detection function is to determine the change of the vibration wave in the processing power, using the vibration wave detection circuit 41 described above,
This is a function to extract a vibration wave component that fluctuates due to tool wear, chipping, breakage, etc., and to integrate the component to make the tool abnormal when the magnitude of the fluctuation exceeds a threshold value.

This vibration wave detection function is set for each O number and T number in the stroke table 57, and when it is used, a threshold value for abnormality determination is set.

Further, the sequential determination function is used to determine the power change when the change of the machining power causes a specific change when the tool is abnormal.

For example, in drilling, etc., when an initial abnormality occurs in the cutting edge of the tool, the machining power once increases as shown in FIG. 8 and then decreases from the normal machining power, and then the abnormal state of the tool continues. Then, when the breakage occurs, the reduced processing power tends to increase, and the increased state tends to be maintained until the breakage occurs.

To detect such an abnormal state, as shown in FIG. 8, three threshold values (SET1 (upper limit), SET2 (lower limit),
SET3 (upper limit) is set, and the thresholds for SET1 and SET3 are set to the monitoring time (timers ST1, ST3) from when the threshold level is exceeded to when an abnormality is detected. As a result, when the machining power turns on SET1 and then SET2, the increase and decrease of the power can be detected.Next, when SET3 is turned on during the duration of the timer ST1, the movement that has turned to the increase of the machining power is detected. it can. Then, if the time during which SET3 is on continues for the timer ST3, a continuous increase in machining power can be detected, and therefore an abnormal signal is output at that time.

Turning on the machining power for such multiple thresholds,
By combining OFF and the order thereof, it is possible to determine a peculiar variation form of the machining power, and it is possible to detect a tool abnormality.

When using the function of this sequential determination, data of over detection or under detection for a plurality of threshold values is selected from the mode table 58, and the monitoring time (corresponding to each SET value from the channel table 60 ( Read timer ST).

On the other hand, the CPU 62, which constitutes the comparison / determination means 34, controls the power consumption P of the drive motor 9 from the processing load detection means 32.
(T), processing power f (t), integrated value ∫f (t), maximum value f (t) max, integrated value of vibration wave, and stable region detection circuit 4
The feedback signals from 2 are respectively inputted, and the CPU 62 carries out the comparison judgment based on the input signal and the data signal from the storage means 33, and outputs the control signal to the drive circuit 12 according to the judgment result. It has become.

The abnormality detection device 31 of this embodiment has the above-mentioned configuration, and its operation will be described below in sequence based on the processing flow of control by the CPU 62 shown in FIG.

When the T number of the newly selected tool and the O number of the workpiece are fetched from the numerical controller 10 in step 1, the O number and T number are specified in the stroke table 57 of the storage means 33 in step 2, Determine the monitoring conditions.

In the determination of the monitoring conditions, for example, as shown in Table 4, if the No. 1 is specified for both the O number and the T number, the matrix of (O, T) = (1, 1) is displayed. Based on the stored data, the mode table No. = 100 Start stage No. = 4 End stage No. = 6 Reference value (100% reference power) = 10 Repeat function Yes 6 special conditions (automatic calculation) Yes Is set.

When the above conditions are determined, next, the driving electric motor 9 is started in step 3, and in step 4, only during the rush cut time (5 sec) specified by the mode number No. 100 of the mode table 57, The starting power of the motor at the start-up is cut off from the monitoring period.

Then, in step 5, the no-load power f (t ₀ ) held in the sampling circuit 37 is stored, and the machining power f obtained by calculating the no-load power f (t ₀ ) and the power consumption P (t). Store (t).

Next, in step 6, a trigger signal for starting the stage is detected by an external signal or an internal level. The type of this trigger signal is determined by the data stored in the mode table 58.

When the above process is executed, in step 7, machining is executed from the start stage set in step 2,
Monitoring of processing power is started according to the monitoring conditions.

That is, first, in step 8, the stage table 59
To detection start time (timer T1) and detection time (timer T2)
The channel number set in the stage table 59 and set 1 from the channel table 60.
~ Coefficient data is fetched according to SET4, and this coefficient data is multiplied by the reference value of the monitoring condition to obtain four threshold values (SE
T1 to SET4) is determined.

Then, in step 9, the machining power and the threshold value are compared within the range of the designated time T2 to monitor abnormality. In this case, the type of power (processing load, integrated value, etc.) that is the target in this case and the detection method of whether each threshold value (SET1 to SET4) is the upper limit detection or the lower limit detection is described in the mode table 58. Determined by the stored data.

In this case, if the monitoring condition has a special function, that function is executed after step 9 above.

For example, if there is an automatic calculation function, the special function table
Calculation processing is performed based on the conditions set in the automatic calculation table 61 (step 10). When the reference value is determined, the value is fed back to the travel table 57 (steps 11 and 12). In this feedback, each power (power consumption, processing power, integrated value, maximum value) when the reference value was determined by automatic calculation is fed back to the same type of reference value (100% load power) in the stroke table. And calibrate its current reference value to the new value.

Also, if there is a function of vibration wave detection and sequential judgment,
The function is sequentially executed (steps 13 and 14). Both of these functions are also executed based on the data in the condition table attached to the special function table 61.

When an abnormality is identified in the above special function,
The abnormal output operates and the control signal is output.

On the other hand, if an abnormality is determined in addition to the special function, the SET output and the abnormality output are operated in the next step 15.

When step 15 ends, step 16 determines the end of the monitoring period, but if the motor continues to start in that state, the stage executed in step 17 is the final stage set in the stroke table 57. Determine if it is a stage. And if it's not the final stage,
Change the monitoring condition to the condition of the next stage in step 18,
Returning to step 8, the above process is repeated. This repetition is performed by sequentially switching the conditions until the stage reaches the final stage, that is, until the end stage defined in the stroke table 57 is reached.

When the set stage ends, if the repeat function is used, the process returns to step 7 and the stage is restarted from the start stage by the number of repeats (step 19).

On the other hand, when the repeat function is not used or when the repeat function is finished, it is judged that the monitoring is finished (step
20), the drive motor is stopped.

〔The invention's effect〕

As described above, since the present invention sets the threshold value for each stage of each machining tool, it is possible to compare the machining load of each machining stage with a threshold value appropriately corresponding thereto. Therefore, the tool abnormality can be detected accurately and finely.

Further, since the detection of the processing load is controlled by time, there is an advantage that the actual processing time and the detection time can be accurately matched, and the processing powers can be reliably compared.

Further, in the present invention, the processing load can be detected by combining a plurality of threshold values and the monitoring time, and the load signal at the time of processing can be subjected to waveform processing so as to be compared with the threshold value. It is possible to accurately detect the variation that is peculiar to the above and the increase amount of the load due to tool wear, and it is possible to make a reliable determination of the tool abnormality in a wide range.

[Brief description of drawings]

FIG. 1 is a block diagram showing the entire structure of an embodiment of the present invention, FIG. 2 is a graph showing the setting range of time data of the same, FIG. 3 is a perspective view showing a workpiece and its processing stage,
FIG. 4 is a block diagram showing the overall structure of another embodiment, FIG.
FIG. 6 is a block diagram showing the above-mentioned vibration wave detection circuit, and FIGS. 6 (a), (b) and (c) are waveform diagrams showing the process of vibration wave detection,
FIG. 7 is a block diagram showing a processing power stable region detection circuit,
FIG. 8 is a waveform diagram showing a detection example of the sequential determination function, and FIG. 9 is
FIG. 10 is a block diagram showing a processing flow of the CPU, and FIG. 10 is a diagram showing a power waveform obtained in the control of the CPU. 1, 31 ... Tool abnormality detection device, 2, 32 ... Machining load detection means, 3, 62 ... Microcomputer (CPU), 4, 33 ... Storage means, 5 ... Filter circuit, 6 ... Stroke number Data file, 7 ... Detection information data file, 8 ... Memory section, 9 ... Driving electric motor, 10 ... Numerical control device (NC device), 34 ... Comparison judgment means, 39 ... Integrating circuit, 40 ... … Maximum value detection circuit, 41 …… Vibration wave detection circuit, 42 …… Machining power stable region detection circuit, 57 …… Stroke table, 58 …… Mode table, 59 …… Stage table, 60 …… Channel table, 61… … Special function table.

Claims (2)

(57) [Claims] 1. A machining load detecting means for detecting a load during machining by a machining tool, detection start time data and detection time data for each machining stage, and a tool life judgment threshold value corresponding thereto. Data for storing the threshold value, the data for determining the threshold value is a reference value specified for each machining tool and a threshold value specified for each stage. The coefficient data is composed of at least the upper limit value of the contact detection between the tool and the workpiece, the upper limit value of the tool breakage, and three or more kinds of data for detecting the lower limit value, and the reference value and the coefficient. Threshold value determining means for determining the threshold value data by multiplying the respective data, and a machining tool extracted corresponding to the threshold value data from the detection time data stored in the storage means. Comparing and judging means for judging an abnormality of the tool by comparing the threshold value determined by the threshold value determining means with the output value of the processing load detecting means within the monitoring time set by the visual time data. In addition, a tool abnormality detection device for machine tools that selects a machining tool based on a machining program for machining. 2. The processing load detecting means is provided with a waveform processing circuit for performing waveform processing of a load signal at the time of processing, and the storage means stores threshold data corresponding to the waveform processing. The tool abnormality detection device according to item (1).