JPH0455820B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a device that uses AE signals to detect abnormalities in a tool during cutting.

Conventional devices that detect tool abnormalities using AE signals generally output an abnormality signal when the peak value of the sudden AE signal exceeds a certain threshold. However, there are problems such as the precision is not sufficient because it may operate due to noise caused by machine vibration, collision of chips, splashing of cutting oil, etc., and the equipment is complicated and expensive. It has not been possible to reliably and economically detect damage to small diameter tools. In other words, Figure 1 shows the results of analysis of the number of lap breaks in the AE signal during normal cutting and when the tool is abnormal.
Pass the signal through a filter to remove high frequency components (e.g.
150KHz or higher), and if the level exceeded a set threshold, it was determined to be abnormal. In Fig. 1, the level is clearly higher in abnormal conditions than in normal conditions, but high-level noise was also observed in signals generated by mechanical vibrations and chips, which could lead to incorrect judgments. There were also devices that detected frequency ratios, but they sometimes made erroneous decisions due to noise.

The present invention aims to solve these conventional problems.As is clear from Fig. 1, during normal cutting, the signal of approximately 100KHz or less is large and the overall level is low, but during abnormal cutting, the signal is low. about
100K to 300KHz is large and the overall level is large. The present invention focuses on this feature to detect tool abnormalities, and includes a sensor that detects an AE signal,
The present invention relates to a tool abnormality detection device that includes a preamplifier, a full-wave rectifier circuit, an averaging circuit, a bandpass filter, a full-wave rectifier circuit, a divider, a comparator, an AND circuit, and an output circuit. In other words, the signal after the preamplifier and full-wave rectifier circuit is used as the denominator of the divider, and the preamplifier and bandpass filter (approximately
100K to approx. 300KHz or approx. 100K to approx. 500KHz), the signal after the full-wave rectification circuit is input to the divider as the numerator of the divider, and the output of the divider is compared with a set threshold using a comparator to detect frequency abnormalities. In addition, the signals after the preamplifier, full-wave rectification circuit, and averaging processing circuit are compared with a comparator to determine the level abnormal output, and the frequency abnormal output and level abnormal output are input to the AND circuit, and both abnormal outputs occur simultaneously. It is now determined that the tool is abnormal only when the problem occurs.

Below, an embodiment of the present invention will be explained with reference to FIG. 2.The AE signal is detected by the sensor 1 and the preamplifier 2
is amplified to a moderate level. As mentioned above, during abnormal cutting, the AE level is high, and in order to detect this, we focused on the fact that the frequency component of approximately 100K to 300KHz or approximately 100K to 500KHz has a large ratio to the whole frequency, and in order to detect this, we detected a level abnormality in the preamplifier signal. and the frequency abnormality detection unit.

The level abnormality detection section consists of an overall rectification circuit 3, an averaging processing circuit 4, and a comparator 9. The signal from the preamplifier 2 is extracted by the full-wave rectification circuit 3 and the averaging processing circuit 4 to extract the average level of the AE signal. 9, and when the AE average level exceeds the set threshold, the comparator 9 outputs a level abnormal signal.

The frequency abnormality detection section is first divided into two systems, a denominator system and a numerator system of the divider 8. For the denominator system, the entire frequency component is passed through the full-wave rectifier circuit 5 and inputted as the denominator X of the divider 8. The molecular system consists of a bandpass filter 6 of about 100K to 300KHz or about 100K to 500KHz and a full-wave rectifier circuit 7. The bandpass filter 6 extracts the necessary frequency components from the signal from the preamplifier 2, and the full-wave rectifier circuit 7 extracts the necessary frequency components from the signal from the preamplifier 2. input as the numerator Y of the divider 8. The divider 8 is an analog divider and outputs the value of Y/X. Y/X is a frequency component and is a value proportional to the ratio of 100K to 300KHz or 100K to 500KHz to the overall level. This frequency ratio Y/X is input to the comparator 10, and when the frequency ratio Y/X exceeds a set threshold, the comparator 10 outputs a frequency abnormality signal.

The level abnormality signal of the comparator 9 and the frequency abnormality signal of the comparator 10 are input to the AND circuit 11. Only when the level abnormality signal and the frequency abnormality signal occur simultaneously, the AND circuit 11 determines that there is an abnormality and outputs the signal to the output circuit 12. An abnormal signal is output, and the output circuit 12 outputs a control signal such as a relay output.

As described above, the present invention includes a level abnormality detection section consisting of a full-wave rectifier circuit 3, an averaging circuit 4, and a comparator 9, a denominator system consisting of a full-wave rectifier circuit 5, a bandpass filter 6, and a full-wave rectifier circuit 7. It is equipped with a frequency abnormality detection section consisting of a divider 8 and a comparator 10, which receive the AE signal from the preamplifier 2 through a molecular system consisting of a molecular system, and receive the level abnormality signal and frequency abnormality signal from both detection sections. Circuit 1 outputs a tool abnormality signal when these occur at the same time.
Since the AND circuit 11 that outputs the output to the output signal 2 is provided, there is no judgment error due to noise caused by mechanical vibration, collision of chips, splashing of cutting oil, etc., and the circuit configuration is simple. Furthermore, since it operates even when an abnormal sound is generated due to abnormal wear immediately before a breakage, it can also be used as a breakage prediction/detection device.

[Brief explanation of the drawing]

FIG. 1 is a comparison diagram of AE signal levels during normal cutting and during tool abnormality, and FIG. 2 is a block diagram of an embodiment of the present invention. 1...Sensor, 2...Preamplifier, 3...Full wave rectifier circuit, 4...Averaging processing circuit, 5...Full wave rectifier circuit, 6...Band pass filter, 7...Full wave rectifier circuit, 8 ...Divider, 9...Comparator,
10... Comparator, 11... AND circuit, 1
2...Output circuit.

Claims (1)

[Claims] 1 A sensor that detects an acoustic emission (hereinafter abbreviated as AE) signal, a preamplifier that amplifies the AE signal from the sensor to an appropriate level, a full-wave rectifier circuit that rectifies the signal from the preamplifier, and a full-wave rectifier circuit that rectifies the signal from the preamplifier. Input the signal from the wave rectifier circuit
An averaging processing circuit that extracts the average level of the AE signal, and a level abnormality that inputs the signal from the averaging processing circuit, compares it with a set threshold, and outputs a level abnormal signal when the AE average level exceeds the set threshold. A signal comparator, a denominator full-wave rectifier circuit that detects the entire frequency component from the signal from the preamplifier and outputs it to the divider as the denominator A band-pass filter that extracts frequency components, a molecule-based full-wave rectifier circuit that detects the signal from the band-pass filter and outputs it to the divider as the numerator Y, and a numerator-based full-wave rectifier that combines the signal of the denominator-based full-wave rectifier circuit with the molecule-based full-wave rectifier Input the circuit signal and perform division Y/X to obtain the frequency component ratio Y/
A divider that outputs X, and a frequency abnormality signal comparator that inputs the divider signal, compares it with a set threshold, and outputs a frequency abnormal signal when the frequency component ratio Y/X exceeds the set threshold. , an AND circuit that inputs the output of the comparator for the level abnormal signal and the output of the comparator for the frequency abnormal signal, and outputs the tool abnormal signal when the level abnormal signal and frequency abnormal signal occur simultaneously, and the signal from the AND circuit. A tool abnormality detection device comprising an input circuit and an output circuit for outputting a control signal.