JP2962583B2 - Strain-free precision numerical control processing method and apparatus by radical reaction - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for precise numerical control of strain using a radical reaction, and more particularly, to a semiconductor such as silicon single crystal or a conductor or an insulator such as glass or ceramic. TECHNICAL FIELD The present invention relates to a distortion-free precision numerical control processing method and apparatus capable of processing with high accuracy without introducing a thermal deterioration layer.

[0002]

2. Description of the Related Art Conventionally, a neutral radical based on a reaction gas generated by an electrode to which a high voltage is applied is supplied to a processing surface of a workpiece, and a radical reaction between the neutral radical and atoms or molecules of the processing surface is performed. Volatile substances generated by vaporization can be removed by vaporization, and processing can be performed with high precision without introducing defects or thermally altered layers in semiconductors such as silicon single crystals or conductors or insulators such as glass or ceramics. The distortion-free precision machining method has been disclosed by the present applicant in Japanese Patent Application Laid-Open No. 1-125.
No. 829 has already been disclosed.

On the other hand, in lathe processing or the like based on the conventional processing principle using fine cracks, numerical control is performed in which coordinates of a target processing surface are input numerically and a feed amount of a cutting tool in a direction approaching a workpiece is controlled based on the input. Machine tools are already known.

[0004] However, in the method and the apparatus for distortion-free precision numerical control by radical reaction, the processing principle is fundamentally different from that of the numerically controlled machine, so that the control mechanism used in the conventional numerically controlled machine is used. It cannot be applied as is. For example, in a conventional numerically controlled machine tool, after a workpiece is processed into a target shape, even if the cutting tool is moved along the target shape, the workpiece is not further processed. In the method of the present invention using
When the processing electrode is similarly moved, neutral radicals generated based on the reaction gas are supplied to the processing surface, whereby the processing proceeds further than the target shape.

[0005]

SUMMARY OF THE INVENTION In view of the above situation, the present invention aims at solving the problem by generating a radical reaction between a neutral radical based on a reactive gas and an atom or a molecule on a processing surface of a workpiece. In removing the volatile substances that have been removed and proceeding with the machining, the fact that there is a correlation between the machining time and the machining amount by the machining electrode at a certain point on the machining surface is used, and the pre-machining surface input to the control circuit and the target machining Set the processing time according to the processing amount corresponding to the deviation of the coordinates of the surface, and specifically, numerically control the scanning speed according to the processing amount or set the stop time according to the processing amount when scanning stepwise. It is an object of the present invention to provide a distortion-free precision numerical control processing method by a radical reaction and an apparatus therefor, which can perform numerical control and perform numerical control processing with a shape accuracy of 0.01 μm or more.

[0006]

According to the present invention, a neutral radical based on a reactive gas generated by a processing electrode to which a high voltage is applied is supplied to a processing surface of a workpiece to solve the above-mentioned problem. Volatile substances generated by the radical reaction between the neutral radicals and atoms or molecules on the processing surface are removed by vaporization, and the processing is performed by changing the processing electrode relative to the processing surface. It is determined according to the material type and the workpiece gas, processing time and the processing amount and the correlation data and the machining time according to the coordinates difference based on the front working surface and the coordinate data of the object processed surface between the A numerical control method for distortion-free precision numerical control by radical reaction was established.

In controlling the processing time, the shape of the pre-processed surface is measured before starting the process, and the coordinate data of the pre-processed surface and the target processed surface are input to a control circuit, and the coordinate data and the correlation data are input. based on the NC data relating to scanning speed and the coordinates created by the controls the relative scanning speed of the machining electrode and the workpiece at numerically controlled drive mechanism by the control circuit, the scanning speed is continuously The acceleration was controlled to change.

In controlling the machining time, the shape of the pre-machined surface is measured before machining is started, and coordinate data of the pre-machined surface and the target machined surface are input to a control circuit, and the coordinate data and the correlation data are input. gradual movement of the scanned and processed electrode based on the NC data relating the stop time and coordinates created, the machining electrode and the workpiece at numerically controlled drive mechanism by the control circuit to a relatively stepwise by a It is also possible to control the speed so that

Furthermore, the pre-machined surface shape of the measured during machining progress, enter the coordinate data of the previous working surface and purpose processing surface to the control circuit, the coordinates created by said correlation data and the coordinate data Based on the NC data relating the scanning speed, the driving mechanism numerically controlled by the control circuit controls the relative scanning speed between the processing electrode and the workpiece,
When controlling the acceleration so that the scanning speed changes continuously or when scanning the processing electrode and the workpiece relatively stepwise, control the stop time according to the coordinate difference between both processing surfaces, It is also possible to control the speed so that the movement of the processing electrode becomes gentle.

In order to carry out the method, a reaction vessel capable of sealing or flowing an atmospheric gas containing a reaction gas at a high pressure, a processing electrode disposed in the reaction vessel, and a direct current or alternating current A power supply for applying a high voltage to generate neutral radicals based on the reaction gas, a drive mechanism having a servomotor capable of displacing one of the processing electrode or the workpiece relative to the other, and Or a measuring means for measuring the shape of the pre-processed surface of the workpiece during the progress of processing, and determined by the type of reaction gas and the material of the workpiece, the processing time by the processing electrode, neutral radicals seat front working surface and purpose processing surface of the correlation data are inputted, further workpiece between the processing amount can be removed by vaporizing the volatile material produced by the radical reaction with atoms or molecules of the working surface Data is input, it constituted their coordinate difference workpiece with no strain precision numerical control machining apparatus a control signal related to the relative scanning by more becomes a radical reaction and a control circuit for outputting to the drive mechanism of the machining electrode in accordance with the.

[0011]

According to the present invention, there is provided a method and an apparatus for precise numerical control of distortion-free processing by radical reaction according to the present invention, the processing time by a processing electrode, the radical reaction between neutral radicals and atoms or molecules constituting a processing surface. Between the type of reaction gas and
Utilizing the fact that there is a correlation according to the material of the workpiece, when the coordinates of the pre-processed surface and the target processed surface of the workpiece are input to the control circuit to which this correlation data has been input in advance, it corresponds to the coordinate difference. The optimum machining time at a certain point on the machining surface is calculated from the correlation data according to the machining amount to be machined, and the machining electrode and the workpiece are relatively moved by a drive mechanism having a servomotor to which a control signal is input based on the calculated machining time. Moving and processing,
This is to obtain a target processed surface without introducing a defect or a thermally altered layer.

Further, in order to machine a certain point on the machined surface in an optimal machining time, the relative scanning speed between the machining electrode and the workpiece according to the machining amount corresponding to the coordinate difference between the previous machined surface and the target machined surface. If the machining electrode is controlled numerically , the machining electrode
Optimally set the scanning speed so that
Acceleration control so that the speed changes continuously, and machining
When scanning the electrode and workpiece relatively stepwise
Controls the stop time according to the coordinate difference between both processing surfaces
At the same time, control the speed so that the machining electrode moves slowly.
Vibrations and stresses caused by rapid movement of the machining electrode.
Abnormality of workmanship is suppressed as much as possible, and shape accuracy of 0.01μm or more
Is issued.

Further, when measuring the shape of the pre-processed surface before starting the process and inputting the coordinate data to the control circuit,
Since the relative scanning speed or stop time between the processing electrode and the workpiece is calculated in advance, the number of coordinate calculations during processing can be reduced.

When the shape of the pre-processed surface is measured during the processing and the coordinate data is input to the control circuit, the optimum scanning speed or stop time is calculated as the processing progresses. It is possible to proceed while correcting the inside machining error.

[0015]

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a first embodiment of the present invention;

FIG. 1 is a conceptual diagram of a precision distortion-free numerical control apparatus by a radical reaction according to the present invention. In the figure, 1 is a reaction vessel, 2 is a working electrode, 3 is a power source, 4 is a driving mechanism, and 5 is a control mechanism. A computer as a circuit is shown, and S indicates a workpiece.

According to the present invention, a DC or AC high voltage is applied from a power source 3 to a processing electrode 2 disposed in a reaction vessel 1 capable of sealing or flowing an atmosphere gas containing a reaction gas, thereby forming a processing electrode 2 and a processing target. A high electric field is generated in the processing gap between the workpieces S, and a neutral radical based on the reaction gas is generated by the electric field, and a radical reaction between the neutral radicals and atoms or molecules constituting the processing surface of the workpiece S is performed. The generated volatile substances are vaporized and removed to advance the processing. The processing electrode 2 attached to the drive mechanism 4 receives a control signal for scanning calculated based on the correlation data of the processing time and the processing amount previously input from the computer 5, and inputs the control signal to the drive mechanism 4. The rotation angle or the number of rotations of the servo motor constituting the drive mechanism 4 is numerically controlled, scanned along a predetermined scanning line, and processed into a target shape. Also, although not shown, a measuring means for measuring the shape of the front processing surface is provided, and the shape data of the workpiece S is measured before starting the processing, or is measured during the processing by using the measuring means, and the coordinate data is transmitted to the computer 5. It is possible to enter.

Here, the processing time can be controlled by controlling the scanning speed, and when scanning stepwise, by controlling the stop time. When the processing electrode 2 and the workpiece S are relatively scanned, there is a case where one of them is fixed and the other is moved, and a case where both are moved. In the present embodiment, an embodiment will be described in which the shape of the pre-processed surface is measured before starting the process, the processing electrode 2 is moved based on the coordinate data of the pre-processed surface and the target processed surface, and the scanning speed is controlled.

The shape and structure of the processing electrode 2 are to be appropriately adopted according to the shape of the target processing surface, and are roughly classified into an electric field concentration type and an electric field dispersion type. Is shown, the tip is narrowed, the plasma generated between the tip and the processing surface of the workpiece S is localized, and the region where neutral radicals are generated is limited. This makes it possible to perform point machining on a complicated machining surface. Examples of the electric field dispersion type electrode include a planar electrode, and a smoothing process can be performed using this electrode instead of lapping.

The high voltage applied from the power source 3 to the processing electrode 2 is DC or AC depending on whether the workpiece S is a conductor, a semiconductor or an insulator.
The F region is used, and the voltage is set to the minimum necessary for the generation of neutral radicals based on the reaction gas from the ambient gas containing the reaction gas and the inert gas, and excessively high energy charged particles are generated. Thus, the workpiece S is not damaged. Here, when the workpiece S is a conductor or a semiconductor, a direct current high voltage is applied to the processing electrode 2, the workpiece S is grounded, and the workpiece S is a semiconductor or an insulator. , A high-frequency voltage is applied to the processing electrode 2 and the reaction vessel 1 is grounded.

The processing speed depends on the type of reaction gas and the workpiece S
Therefore, it is necessary to select an optimum reaction gas according to the workpiece S. Where the processing speed and
Is the machining amount divided by the machining time, and therefore
The processing speed is related to the processing time and the processing amount.
The correlation data between the processing time and the processing amount is
Related data. For example, when the workpiece S is silicon single crystal or quartz glass, the reaction gas is S
F ₆ is suitable. Other reactive gases include CF _{4 in the case of} fluorine and Cl ₂ , CCl ₄ , and the like in the case of chlorine.
PCl _{4 and the} like. The reaction gas is not directly excited by the high electric field, but generates a plasma of an inert gas which is easily ionized or excited by the high electric field, and generates neutral radicals by collision with the reaction gas. is there. Here, even if the inert gas plasma and the neutral radicals due to the reaction gas are generated in the same region, the inert gas plasma is generated first, and then transported to collide with the reaction gas in different regions. It is also possible to generate neutral radicals.

Next, the driving mechanism 4 includes the machining electrode 2
And a stage (X,
Y, Z), and each stage is driven by a servomotor via a ball screw. The minimum moving distance (moving distance per one pulse of the motor control signal) and the maximum moving speed of each stage are 0.16 μm and 8 mm / X-axis (direction along the reference plane), respectively.
sec, and 0.5 μm, 0.1 mm / sec in the Y axis (direction along the reference plane) and the Z axis (direction perpendicular to the reference plane).
c. Each servo motor is connected to a computer 5
FIG. 2 shows a system configuration diagram of the driving. Each stage has a similar drive system, and a normal drive state is always guaranteed by a feedback pulse from a rotary encoder as shown in the figure.

As shown in FIG. 3, the scanning at the time of numerical control (NC) processing is carried out by feeding a predetermined length by an X stage capable of high-speed driving, then by one step by a Y stage, and again by an X stage. The operation of sending in the reverse direction is repeated. At this time, the processing surface is obtained as an accumulation of spot-like processing traces indicated by circles in the drawing.

In this type of machining method, when performing numerical control machining, stability of machining speed is very important for machining accuracy, and at the same time, what is selected as a control variable is an important problem. Volatile substances generated by a radical reaction between the neutral radicals based on the reaction gas of the present invention and atoms or molecules constituting the processing surface of the workpiece S are vaporized and removed, and the processing amount when the processing is advanced by removing and Since the scanning speed has a correlation, the scanning speed is used as a control variable. The amount of processing can be controlled by the scanning speed of each point. The correlation data between the processing amount and the scanning speed is input to the computer 5 in advance.

The coordinates in the numerically controlled machining system according to the present invention are set as shown in FIG. 4, and the conceptual procedure of machining is shown in FIG.
Shown in The reference plane assumed for the workpiece S is set to the XY plane,
The Z axis is taken in a direction perpendicular to this plane. When a target shape is obtained by the present invention, a procedure is adopted in which only the deviation from the shape of the pre-processed surface is processed. As shown in the figure, the pre-processed surface is measured with a predetermined accuracy in a constant temperature chamber, and the shape data (x, y, Z ₁ ) at each coordinate point is input to the computer 5. Also,
Numerical data (x, y, Z ₂ ) of the shape of the target machined surface is also input to the computer at the same time, and the required machining amount Z ₁ − at each point.
Z ₂ is calculated. Correlation data between the processing amount and the scanning speed for the workpiece S at that time has already been input to the computer. Based on this value, the scanning speed at each point as a control variable is calculated, and numerical control CVM is performed. (Chemi
cal VaporizationMachinin
g) Processing is performed.

Next, software for performing NC processing using NC data calculated from the coordinate difference between the measurement result of the pre-processed surface and the shape to be processed will be briefly described.

FIG. 6 shows a flowchart of this software (NC software), and its subroutine is shown in FIGS. 7 and 8 (FIGS. 7 and 8 are continuous). N
The C software reads the NC data, and then controls each axis according to the data therein. The above-described NC data describes a scanning method in which the processing electrode is reciprocated in the X-axis direction and is gradually moved in the Y-axis direction at each turn while being reciprocated. The control of the processing time at each point required for NC processing is performed by controlling the scanning speed of the processing electrode during X-axis scanning.

On the NC data, the scanning speed corresponding to the average processing amount in each divided control section is given as NC data, and the change of the scanning speed during the control section is made as gentle as possible by the NC software. The scanning speed is controlled by such acceleration. With this control method, the amount of processing during each control section can be linearly interpolated. Also, if the scanning speed is suddenly changed without such control, a large acceleration acts at that moment, and at this moment, the vibration generated on the stage changes the distance between the processing electrode and the workpiece, and the The machining amount shows an abnormal value at the point. For this reason, it is important to control the scanning speed to be continuous. Another point to be noted in performing the NC processing is the minimum value of the scanning speed of the processing electrode. If the total processing amount is large, the processing amount can be increased by reducing the scanning speed of the processing electrode, but at this time, the step between the processed portion and the non-processed portion caused by one scan in the X-axis direction increases, The distribution of neutral radicals may change. Therefore, the minimum value of the scanning speed is fixed, and when a large processing amount is required, the processing area is repeatedly scanned to cope with the processing amount.

Next, the specific operation of the NC software will be described. First, NC data stored in an external storage medium such as a floppy disk is stored in a memory of a computer. The NC data includes a header portion and a command data portion. The header portion has information such as a title, a processing amount, a creation date and time, and a version number, and a command data portion which is data for actually executing the NC processing. Contains a loop command and each axis control command.

Based on the above data, the computer reads and executes the command data sequence from the front as shown in the flowcharts of FIGS. (1) If the command data is a termination instruction, the NC processing is terminated there. (2) If the command data is Y-axis or Z-axis movement instruction data, confirm that the YZ-axis stage has stopped by reading the status of the control board of the motor mounted on the expansion I / O slot of the computer. Then, all the pulses and the moving speed are set on the control board of the motor, and the movement is instructed. (3) If it is X-axis movement instruction data, first calculate acceleration (if there is speed control data for NC machining) and speed change position, and confirm that the X-axis stage is stopped. Thereafter, parameters such as initial speed, acceleration, arrival speed, deceleration start point, and deceleration are given to the control board to start the motor. When the motor starts moving, the computer monitors the position of the X-axis stage by the pulse counter on the control board of the X-axis motor, and when it comes to the previously calculated speed change point, the computer changes the speed and acceleration to be changed. And instruct speed change. This speed change is repeated until there is no speed control data. (4) If it is the loop start instruction data, the data from the next instruction data to the instruction data immediately before the loop end instruction data is executed a specified number of times.

Next, the details of the X-axis speed control for NC machining will be described. Motor and controller functions include a function that automatically performs a series of operations such as acceleration start, constant speed rotation, and deceleration stop. When a speed change instruction is given during constant speed rotation, the speed changes at the given acceleration. There is also a function to perform. The speed control for NC machining is performed using this function. As shown in FIG. 9, an acceleration period w (number of rotation pulses of the motor) is provided between the control sections by the NC software, and the acceleration control is performed so that the scanning speed changes continuously. In order to match the speed change point given to the NC data at the point where w / 2 rotation is started after acceleration, the speed change point given to the NC data is changed by w / 2 from the speed change point given to the NC data.
The point just before is the actual speed change point. In order to make the acceleration as small as possible, it is necessary to increase w as much as possible. However, since acceleration is performed in the range of w / 2 before and after the speed change point given in the NC data, the next speed In order not to overlap the acceleration section for the change point and to try to maximize w, the length of the control section before and after the speed change point must be the same as the shorter of the lengths. The speed change position and the acceleration are calculated each time the X axis is scanned in the NC software as described above, from the length of each control section and the speed data there.

As described above, an example has been described in which the processing electrode 2 of the electric field concentration type is scanned, spot-shaped processing traces are integrated and processed into an arbitrary target shape. As shown in FIG. 10, the process can be performed using a flat plate-shaped electric field dispersion type processing electrode 2. The scanning control of the machining electrode 2 in this case is the same as described above, but the feed step in the Y-axis direction is set large.

In addition, as shown in FIG. 11, when the processing electrode 2 which is of the electric field concentration type but extends in the Y-axis direction is used, scanning in the Y-axis direction becomes unnecessary, and only the X-axis direction is used. By scanning, a processed surface having a waveform as shown in the figure is obtained.

Each of the processing electrodes 2 described above generates plasma between the processing gap between the tip of the processing electrode 2 and the processing surface of the workpiece S, thereby generating neutral radicals in the region. It is externally generated. However, when the processing gap changes, the state of plasma generation is different, so that plasma is always generated under the same conditions inside the electrode, neutral radicals are generated, and the neutral radicals are supplied to the processing surface to process. Endogenous types may be preferred. When a high frequency is used for generating plasma, the high frequency may be absorbed and heated depending on the material of the workpiece S, so that the internally generated type is suitable. Two examples of the internally generated type are shown below.

FIG. 12 shows the internally generated type and the electric field concentrated type processing electrode 2. The processing electrode 2 is composed of an electrode 6 for applying a high voltage at the center and a grounded counter electrode 7 around the electrode 6.
The electrode 6 is inserted into a hollow counter electrode 7 via an insulator 8 such as a insulator. The counter electrode 7 is provided with a nozzle portion 9 formed in a tapered shape at the tip and an ejection port 10 at the center thereof, and the tip of the electrode 6 is arranged close to the nozzle portion 9 facing the ejection port 10. ing. Further, in order to supply an atmosphere gas containing a reaction gas into the inside of the counter electrode 7, a supply pipe 1 which can be extended and contracted to a gas supply hole 11 opened in a side surface of the counter electrode 7 or an insulator 8.
2 are connected. When the atmosphere gas is supplied from the supply pipe 12 into the counter electrode 7, plasma is generated between the tip of the electrode 6 and the inner surface of the nozzle portion 9 of the counter electrode 7, and neutral radicals are generated. By the continuous supply of the atmospheric gas, the atmospheric gas containing neutral radicals is supplied from the jet port 10 to the processing surface of the workpiece S, and the processing proceeds as described above. In order to quickly remove the atmosphere gas after the processing and the volatile substances generated by the processing from the processing surface, a gas recovery unit 13 having a tip end side opened around the nozzle 9 of the counter electrode 7 is provided. A recovery pipe 15 similar to the supply pipe 12 is connected to a gas recovery hole 14 formed in the gas recovery section 13. In addition, this processing electrode 2
In the case of processing a spot, it is formed in a columnar shape, and in the case of processing a large area simultaneously, it is formed in a long shape with the cross-sectional shape shown in the figure, and the shape is not particularly limited.

FIG. 13 shows an internally generated and electric field dispersion type processing electrode 2. This processed electrode 2 has a flat electrode 6 disposed inside a flat hollow counter electrode 7,
A support rod 16 extending from the electrode 6 is fixed to one side surface via an insulator 8, and a large number of ejection ports 10 are provided on a flat surface 17 on the other side facing the processing surface of the workpiece S. Has formed. The plate-shaped electrode 6 is disposed close to the inner surface of the flat surface 17, and supplies an atmosphere gas containing a reaction gas through a supply pipe 12 and a gas supply hole 11 connected to the counter electrode 7. Plasma is generated between the flat surface 17 and neutral radicals are generated, and supplied to the processing surface of the workpiece S from the ejection port 10. Although not shown, a recovery unit similar to the gas recovery unit 13 described above is appropriately provided.

FIG. 14 shows an example of an externally generated and electric field concentrated type processing electrode 2. An injection port 10 for an atmospheric gas is formed at the center of the tip of a single tapered electrode and communicates with the injection port 10. A gas supply hole 11 is formed, a supply pipe 12 is connected to the gas supply hole 11, and a gas recovery unit 13 is formed in the same manner as described above.
, A gas recovery hole 14 is formed therein, and a recovery pipe 15 is connected to the gas recovery hole 14.

The adjustment of the processing gap between the processing electrode 2 and the processing surface of the workpiece S can be rigidly performed by moving the Z-axis stage. As shown in FIGS. Can be performed by floating the processing electrode 2 by the dynamic pressure of the ejected atmospheric gas. That is, the gas pressure supplied by the supply pipe 12 is adjusted, and the processing gap can be adjusted as a gas bearing by the ejection pressure of the atmospheric gas from the ejection port 10.

FIGS. 15 and 16 mainly show the work S
Blade-type machining electrode 2 used for cutting a workpiece
Is shown. This is because the processing groove becomes deeper as the cutting progresses, and the normal gas supply means cannot supply the atmosphere gas including the reaction gas sufficiently. Therefore, the gas supply path 18 and the gas recovery path 19 are formed inside the flat processing electrode 2. Is provided.

More specifically, the processing electrode 2 shown in FIG. 15 has a flat electrode 6 disposed at the center, and spacers 2 from both sides thereof.
The flat surface plates 21 and 21 are fixed with 0,.
The gap between the electrode 6 and one surface plate 21 is defined as a gas supply path 1
8 and the gap with the other surface plate 21 is
It is of the following structure. In this case, the tip of the electrode 6 is made to protrude from both surface plates 21 and 21.

The machining electrode 2 shown in FIG. 16 has a large number of grooves 2 on both sides of the flat intermediate plate 22 from the base end toward the tip end.
Are formed, the surface plates 21 and 21 are joined from both sides thereof, and the gas supply path 18 partitioned by the concave groove 23 and the surface plate 21 is formed.
And a gas recovery path 19, and a rod-shaped electrode 6 coated with a material having corrosion resistance to reactive gas and neutral radicals is fixed along the tip of the intermediate plate 22, and is provided on both sides of the electrode. Has a structure in which the gas supply path 18 and the gas recovery path 19 are open. Here, the intermediate plate 22 and the surface plate 21 are made of ceramics.

The structure of the processing electrode described above is to simultaneously supply a reactive gas and an inert gas and to generate plasma by the inert gas and neutral radicals based on the reactive gas in the same region. As shown in the figure, first, a plasma is generated by an inert gas, which is transferred toward the processing surface of the workpiece S, and collides with a reaction gas in a region different from the plasma generation region to generate neutral radicals. It is.
More specifically, in addition to the structure substantially similar to the processing electrode 2 shown in FIG. 12, the inert gas supplied from the inert gas supply pipe 12 is supplied to the inner surface of the nozzle portion 9 at a predetermined distance from the jet port 10. And between the tip of the electrode 6 to be ionized and excited, transferred to the jet port 10 side by the gas pressure, and supplied to a reaction space 24 provided immediately before the jet port 10 with a flexible reaction gas supply pipe 25 and Gas inlet pipe 26
In this structure, a neutral radical is generated by colliding with a reaction gas supplied through the workpiece S and supplied to the processing surface of the workpiece S. With this configuration, there is no possibility that the generated plasma comes into contact with the processing surface, and the temperature of the processing surface hardly increases, so that processing at a lower temperature becomes possible. It is also effective when a reaction gas that captures electrons in the plasma and adversely affects the plasma state is used, for example, when a halogen gas is used.

[0043]

As described above, according to the method and apparatus for precise numerical control of distortion-free by radical reaction of the present invention,
Between the processing time by the processing electrode at a certain point on the processing surface of the workpiece and the processing amount progressing by removing volatile substances due to radical reactions between neutral radicals and atoms or molecules constituting the processing surface Depending on the type of reaction gas and the material of the workpiece
By taking advantage of the fact that there is a correlation, the processing time can be numerically controlled without introducing defects or thermally altered layers on the processing surface .
The target machined surface can be automatically obtained with a shape accuracy of 0.01 μm or more.

Further, when scanning is performed at a scanning speed or in a step-like manner capable of performing accurate and highly repetitive control as a control variable, the stop time is used, so that the optimal processing time at a certain point on the processing surface is reduced. It can be set accurately and easily, and numerical control becomes easy. In addition, the scanning speed
When controlling the degree, the machining electrode
Optimally set the scanning speed so that
The acceleration is controlled so that
And when scanning the workpiece relatively stepwise,
The stop time is controlled according to the coordinate difference between the two machining surfaces.
In addition, speed control is performed so that the machining electrode moves slowly.
Generated by the sudden movement of the machining electrode
Vibration and processing abnormalities are suppressed as much as possible, and 0.01μm or more
Shape accuracy can be achieved.

Further, when measuring the shape of the pre-processed surface before starting the process and inputting the coordinate data to the control circuit,
Since the relative scanning speed or the stop time between the processing electrode and the workpiece is calculated in advance, the number of coordinate calculations during processing can be reduced, so that high-speed processing can be performed.

When the shape of the pre-processed surface is measured during the processing and the coordinate data is input to the control circuit, the optimum scanning speed or stop time is calculated with the progress of the processing. Since it is possible to proceed while correcting the middle processing error, the accuracy can be further improved.

[Brief description of the drawings]

FIG. 1 is a conceptual diagram of the apparatus of the present invention.

FIG. 2 is a control block diagram of a driving servomotor.

FIG. 3 is a conceptual diagram of wide area processing by spot scanning.

FIG. 4 is a conceptual diagram showing a relationship between a front processing surface and a target processing surface.

FIG. 5 is a block diagram showing a procedure of numerical control machining.

FIG. 6 is a flowchart of numerical control software.

FIG. 7 is a flowchart showing the first half of a subroutine of the numerical control software;

FIG. 8 is a flowchart showing the latter half of the subroutine of the numerical control software.

FIG. 9 is a conceptual diagram of acceleration / deceleration control in scanning.

FIG. 10 is a simplified perspective view showing an example of an electric field dispersion type processing electrode.

FIG. 11 is a simplified perspective view showing another example of an electric field concentration type processing electrode.

FIG. 12 is a simplified cross-sectional view illustrating an example of an internally generated processing electrode.

FIG. 13 is a simplified cross-sectional view showing another example of an internally generated processing electrode.

FIG. 14 is a simplified cross-sectional view showing another example of an externally generated processing electrode.

FIG. 15 is a partially simplified perspective view showing an example of a blade-type machining electrode.

FIG. 16 is a partially simplified perspective view showing another example of a blade-type processing electrode.

FIG. 17 is a simplified cross-sectional view showing an example of an internally generated processing electrode in which a plasma generation region and a neutral radical generation region are different.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Reaction container 2 Processing electrode 3 Power supply 4 Drive mechanism 5 Control circuit (computer) 6 Electrode 7 Counter electrode 8 Insulator 9 Nozzle part 10 Spout port 11 Gas supply hole 12 Supply pipe 13 Gas recovery part 14 Gas recovery hole 15 Recovery pipe 16 Support rod 17 Flat surface 18 Gas supply path 19 Gas recovery path 20 Spacer 21 Surface plate 22 Intermediate plate 23 Groove 24 Reaction space 25 Supply pipe 26 Gas introduction pipe

Continuation of front page (56) References JP-A-1-125829 (JP, A) U.S. Pat. No. 4,668,366 (US, A) "SPIE" Vol. 966 Adsrances in Fabrication and Metrology for Optics and Large Optics (1988) PP. 82-90 "SPIE" Vol. 1333 Advanced Optical Manufacturing and Testing (1990) pp. 44-57

Claims (6)

(57) [Claims] 1. A neutral radical based on a reaction gas generated by a processing electrode to which a high voltage is applied is supplied to a processing surface of a workpiece, and a radical reaction between the neutral radical and atoms or molecules of the processing surface is performed. The generated volatile substances are vaporized and removed, and the processing electrode is processed by changing relative to the processing surface, and is determined according to the type of the reaction gas and the material of the workpiece, correlation data and, before processing surface and unstrained precision numerical control machining method by a radical reaction in which a numerically controlled machining time according to the coordinates difference based on the coordinate data of the object processed surface between the processing time and the processing amount . 2. The shape of a pre-processed surface is measured before processing is started,
Enter the coordinate data of the unprocessed surface and the objective working surface to the control circuit, based on the NC data relating to scanning speed and the coordinates created by said correlation data and the coordinate data,
2. A radical reaction according to claim 1, wherein the relative scanning speed between the processing electrode and the workpiece is controlled by a drive mechanism numerically controlled by a control circuit, and the acceleration is controlled so that the scanning speed changes continuously. No distortion precision numerical control processing method. 3. The shape of the pre-processed surface is measured before starting the process,
Enter the coordinate data of the unprocessed surface and the objective working surface to the control circuit, based on the NC data relating the stop time and the coordinates created by the with the coordinate data and the correlation data,
2. A method according to claim 1, wherein the driving mechanism numerically controlled by the control circuit scans the processing electrode and the workpiece relatively stepwise and controls the speed so that the movement of the processing electrode becomes gentle.
A distortion-free precision numerical control processing method by the radical reaction as described. 4. The shape of a pre-processed surface is measured while the process is in progress,
Enter the coordinate data of the unprocessed surface and the objective working surface to the control circuit, based on the NC data relating to scanning speed and the coordinates created by said correlation data and the coordinate data,
2. A radical reaction according to claim 1, wherein the relative scanning speed between the processing electrode and the workpiece is controlled by a drive mechanism numerically controlled by a control circuit, and the acceleration is controlled so that the scanning speed changes continuously. No distortion precision numerical control processing method. 5. The shape of a pre-processed surface is measured while the process is in progress,
Enter the coordinate data of the unprocessed surface and the objective working surface to the control circuit, based on the NC data relating the stop time and the coordinates created by the with the coordinate data and the correlation data,
2. A method according to claim 1, wherein the driving mechanism numerically controlled by the control circuit scans the processing electrode and the workpiece relatively stepwise and controls the speed so that the movement of the processing electrode becomes gentle.
A distortion-free precision numerical control processing method by the radical reaction as described. 6. A reaction vessel capable of sealing or flowing an atmospheric gas containing a reaction gas at a high pressure, a processing electrode disposed in the reaction vessel, and a reaction gas applied by applying a high DC or AC voltage to the processing electrode. A driving mechanism having a servomotor capable of displacing one of the processing electrode or the workpiece relative to the other, and a workpiece before starting the processing or during the processing. Measuring means for measuring the shape of the pre-processed surface, determined according to the type of reaction gas and the material of the workpiece,
Correlation data between the processing time by the processing electrode and the processing amount capable of vaporizing and removing a volatile substance generated by a radical reaction between a neutral radical and atoms or molecules on the processing surface is input, and further processed. A control circuit for inputting coordinate data of a pre-processed surface of a workpiece and a target processed surface, and outputting a control signal relating to relative scanning between the workpiece and the processing electrode to a driving mechanism in accordance with a difference in the coordinates. A unique numerical processing machine for distortion-free precision control by radical reaction.