JPH03178727A - Machining condition detecting device for electric discharge machine - Google Patents

Abstract

PURPOSE:To quantitatively express machining conditions so as to enable segmentation of the machining conditions by providing a processing device for reading per block the lines of digital values stored in a memory, performing fast Fourier transform of the values and identifying the machining conditions according to frequency distribution obtained thereby. CONSTITUTION:Lines of digital values output from an A/D converter 5 for the duration of generation of dielectric breakdown start signals by an dielectric breakdown start signal generator 6 until generation of-off start signals by an off-start signal generator 7 are grouped into blocks and stored at each block in a memory 9. Next, a processing device 10 reads per block the lines of digital values stored in the memory 9 and performs fast Fourier transform thereof and identifies machining conditions according to frequency distribution obtained thereby.

Description

DETAILED DESCRIPTION OF THE INVENTION [Industrial Field of Application] The present invention relates to a machining state detection device for an electrical discharge machining machine that detects the machining state of the electrical discharge machining machine.

[Conventional technology]

Conventional devices of this type are capable of measuring the average or maximum frequency of the voltage across the machining gap (voltage between machining electrodes) when the machining current is flowing.
Extract one special @ quantity and check whether the machining condition is good or not from that value.
It is determined based on two values: bad or bad.

[Problem to be solved by the invention]

In this way, conventional equipment can only determine whether the machining condition is good or bad, and it is not possible to subdivide the machining condition.
There was a problem that fine processing control was not possible.

SUMMARY OF THE INVENTION An object of the present invention is to provide a machining state detection device for an electrical discharge machining apparatus that can subdivide machining states and perform fine machining control.

[Means to solve the problem]

The above object is to provide an electric discharge machining device that performs machining while supplying machining pulses to a machining gap formed by facing a machining electrode and a workpiece by on/off switching control of a switch element. a detection circuit that detects current; a sampling clock generator that generates a sampling clock; an A/D converter that digitizes the analog signal from the detection circuit every time the sampling clock is generated; , a dielectric breakdown start signal generator that detects a point in time when the machining gap is dielectrically broken down and a machining current starts flowing, and generates a dielectric breakdown start signal; and a dielectric breakdown start signal generator that detects a point in time when the machining pulse becomes an off period and starts turning off. An off start signal generator that generates a signal, and a digital value string for each block, where one block is a digital value string output from the A/D converter during a period from generation of the dielectric breakdown start signal to generation of the off start signal. This is achieved by having a memory that stores the sequence, and a processing device that reads out the digital value sequence stored in this memory block by block, performs fast Fourier transform, and determines the machining state from the resulting frequency distribution. be done.

[Effect]

The detection circuit detects the voltage or current in the machining gap, and the signal (analog signal) is digitized by an A/D converter in synchronization with a sampling clock from a sampling clock generator.

The memory stores a digital value string outputted from the A/D converter during the period from generation of the dielectric breakdown start signal of the dielectric breakdown start signal generator to generation of the off start signal of the off start signal generator as one block, and stores the digital value string for each block. Stores a digital value sequence.

The processing device reads out the digital value string stored in the memory block by block, performs fast Fourier transform, and determines the machining state from the frequency distribution obtained thereby. For example, from the above frequency distribution, the machining state can be subdivided using multiple types of feature values, such as: ■ the frequency with a peak, ■ its peak value, and ■ the degree of spread of the frequency distribution around the peak. This allows fine-grained machining control. It becomes possible.

〔Example〕

Hereinafter, the present invention will be described in detail with reference to the drawings. FIG. 1 is a block diagram showing an embodiment of a machining state detection device for an electric discharge machining apparatus according to the present invention, in which 1 is a machining electrode, 2 is a workpiece, and g is a machining electrode 1. This is a machining gap (also referred to as a machining gap) formed by the workpiece 2. During machining, machining pulses are supplied to this machining gap g by on/off switching control of a switch element (not shown) of a machining pulse generator, which will be described later. 3 is a detection/voltage division/filter circuit, which includes a detection circuit section that detects the voltage or current in the machining gap g, here the voltage (voltage between electrodes), and a voltage division circuit section that divides the voltage detected by this detection circuit section. and a filter circuit section. 4 is a sampling clock generator that generates a sampling clock of a predetermined frequency, in this case 20 MH2; and 5 is A that digitizes the analog signal output from the detection/divider/filter circuit 3 every time the sampling clock is generated. /D converter 6 is a dielectric breakdown start signal generator that detects the point in time when the machining gap g undergoes dielectric breakdown and the machining current starts to flow after the machining pulse is applied, and generates a dielectric breakdown start signal.

Reference numeral 7 denotes an off-start signal generator to which a signal synchronized with the machining pulse from the machining pulse generator 8 is applied, and which detects the start point of the off-period of the machining pulse and generates an off-start signal. Reference numeral 9 denotes a memory, which stores a digital value string for each block, with the digital value string outputted from the A/D converter during the period from generation of the dielectric breakdown start signal to generation of the OFF start signal as one block. . In this case, each digital value is stored in the memory 9 in synchronization with the sampling clock.

10 is a processing device, for example, a CPU (Central-
Process-ssing-Unit) or DSP
(Digital-Signal ・Processo
r), here, the CPU reads and transforms the digital value string stored in the memory 9 in units of blocks, and performs Fast-Fourier transform (Fast-Fourier transform).
-Tra-nsform) The machining state is determined from the frequency distribution obtained thereby.

Next, the operation will be explained.

When machining pulses are supplied from the machining pulse generator 8 to the machining gap g and machining is performed, the detection/partial pressure/filter circuit 3 connected to the machining gap g detects the voltage (machining current) of the machining gap g at this time. Detects the inter-electrode voltage during the period when the voltage is flowing.

In addition, this detection/voltage dividing/filter circuit 3 detects the detected voltage by
The voltage is divided and stepped down to suit the input voltage of the next-stage A/D converter 5, and based on the sampling theorem, the frequency (fs) of the sampling clock generated by the sampling clock generator 4 is 1/2 or more. Cut the frequency component of.

On the other hand, the dielectric breakdown start signal generator 6 generates a dielectric breakdown start signal when dielectric breakdown occurs in the machining gap g and machining current begins to flow.

The memory 9 starts storing the digital value outputted from the A/D converter 5 every sampling from the time when the dielectric breakdown detection signal from the dielectric breakdown start signal generator 6 is received.

Further, the off-start signal generator 7 detects the off-start point of the machining pulse based on the signal from the machining pulse generator 8,
An off start signal is given to the memory 9. The memory 9 finishes storing the digital signal of the A/D converter 5 in response to this off start signal.

As described above, one block of sampled digital value (=electrode voltage) strings are stored in the memory 9 for each applied pulse.

The CPU 10 reads out the digital value string sampled as described above and stored in the memory 9 block by block, performs fast Fourier transform on each block, and obtains a frequency distribution. As a result, a frequency distribution for each block (every -1 processing pulse) is obtained, and for each block frequency distribution, for example, ■ the frequency with a peak, ■ its peak value, ■
The machining state can be subdivided based on multiple features such as the spread of the frequency distribution near the peak. This makes it possible to predict the machining state at each point in time, the transition of the machining state up to that point, and the future machining state, thereby enabling fine-grained machining control.

Figures 2 and 3 are diagrams showing examples of frequency distributions obtained by fast Fourier transform for one block of digital value strings stored in the memory 9, respectively; Figure 2 is for normal discharge; indicates the case of short-circuit discharge. As is clear from these FIGS. 2 and 3, the more the short circuit discharge (=abnormal discharge), the more the peak frequency shifts to the lower frequency side, and the spread of the frequency distribution around the peak frequency becomes smaller.

In this way, the peak frequency 1 is used as the characteristic amount of the discharge state.
In any case, it is clear that the machining state is subdivided because it becomes a quantitative value through fast Fourier transformation.

Moreover, since such characteristics and quantities change for each machining pulse, it is easy to predict how the machining state (frequency distribution) has changed up to the present time and what will happen in the future. Note that it goes without saying that the machining state is subdivided even when the feature amount is a feature amount other than the peak frequency or a combination of multiple types of features.

[Effects of the Invention] According to the present invention, the machining state can be expressed quantitatively, and it is possible to subdivide the machining state. This has the effect of allowing processing control.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing one embodiment of the device of the present invention, and FIG.
Figures 3 and 3 are diagrams showing examples of frequency distributions obtained by fast Fourier transform by the CPU in Figure 1, respectively. Figure 2 is a frequency distribution diagram for normal discharge, and Figure 3 is a frequency distribution diagram for short-circuit discharge. It is. DESCRIPTION OF SYMBOLS 1... Electrode for processing, 2... Workpiece, 3... Detection/divider/filter circuit, 4... Sampling/clock generator, 5... A/D converter, 6... - Dielectric breakdown start signal generator, 7... Off start signal generator, 8... Applied pulse generator, 9... Memory, 10... CPU. Go. Machining gap.

Claims (1)

[Claims] 1. In the machining gap formed by facing the machining electrode and workpiece,
In electrical discharge machining equipment that performs machining while supplying machining pulses through on/off switching control of switch elements,
A detection circuit that detects the voltage or current in the machining gap, a sampling clock generator that generates a sampling clock, and an A/D that digitizes the analog signal from the detection circuit every time the sampling clock is generated.
a converter; a dielectric breakdown start signal generator that detects a point in time after application of the machining pulse when the machining gap undergoes dielectric breakdown and a machining current begins to flow and generates a dielectric breakdown start signal; and a dielectric breakdown start signal generator that generates a dielectric breakdown start signal; an off-start signal generator that detects a start point and generates an off-start signal; and one block of digital value strings output from the A/D converter during a period from generation of the dielectric breakdown start signal to generation of the off-start signal. , a memory that stores a digital value string for each block, and a process that reads out the digital value string stored in this memory block by block, performs fast Fourier transform, and determines the machining state from the frequency distribution obtained thereby. 1. A machining state detection device for an electric discharge machining device, comprising: