JPH10128566A - Laser beam plotting device having straightness correcting function of pixel arrangement - Google Patents

Abstract

PROBLEM TO BE SOLVED: To correct the error in straightness of the pixel arrangement by the unit smaller than one pixel of the raster data at a low cost by controlling the modulation of the laser beam to a body to be plotted following the clock pulse of the prescribed frequency based on the raster data to perform the plotting. SOLUTION: The plot starting by the laser beam along the main scanning direction is regulated by the unit of 1/10 of the pixel size by successively shifting the phase of the clock pulse to be outputted from a multiplexer of a straightness correction circuit 84 to a synchronous circuit by π/5 on the positive or negative side when the plotting by the laser beam along the main scanning direction is started. In addition, a Y scale sensor 94 and a signal processing circuit 96 are provided in a main scanning control circuit 86. The output signal from the Y scale sensor 94 is appropriately processed by the signal processing circuit 96, and inputted in the straightness correction circuit 84 as the basic block pulse.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a laser beam, which is deflected in the main scanning direction with respect to an object to be drawn, is moved in the sub-scanning direction, and modulation of the laser beam with respect to the object is performed by raster data. And a laser writing apparatus that performs writing by controlling in accordance with a clock pulse of a predetermined frequency on the basis of the above.

[0002]

2. Description of the Related Art A laser writing apparatus is generally used for writing a fine pattern on a surface of an appropriate object. A typical example of the use is to draw a circuit pattern when manufacturing a printed circuit board using a photolithographic technique, and the object to be drawn is, for example, a photosensitive film for a photomask or a photoresist layer on the substrate. And so on.

In recent years, a series of processes from the process of designing a circuit pattern to the process of drawing the circuit pattern have been integrated, and the laser writing apparatus plays a part in such an integrated system. In addition, such an integrated system includes a CAD (Compu
ter Aided Design) station, a CAM (Computer Aided Manufacturing) that edits the circuit pattern data obtained by the CAD station, that is, vector data.
Stations and the like are provided.

[0004] Vector data created by a CAD station or vector data edited by a CAM station is transferred to a laser drawing apparatus, where the vector data is converted into raster data. In a laser writing apparatus, an object to be drawn such as a photosensitive film for a photomask or a photoresist layer on a substrate is scanned by a laser beam and sequentially moved in a sub-scanning direction. , A predetermined pattern is drawn on the object to be drawn by controlling according to a clock pulse of a predetermined frequency.

Incidentally, one of the main conditions for improving the drawing accuracy of a drawing pattern in such a laser drawing apparatus is to arrange pixels of the drawing pattern straight along the sub-scanning direction. However, in practice, proper straightness cannot be obtained in a pixel array along a sub-scanning direction of a drawing pattern drawn by each laser drawing device,
Moreover, the straightness error differs depending on the individual laser writing apparatus. This is mainly due to an assembly error or the like of a drawing table drive system (sub-scanning direction drive system) included in each laser drawing apparatus.

[0006]

Conventionally, a drawing table driving system has been proposed as one of measures for correcting a unique straightness error existing in a pixel array along a sub-scanning direction of a drawing pattern in an individual laser drawing apparatus. In this case, it is possible to secure the precision of the parts and to suppress the straightness error as much as possible by adjusting the assembly of the parts. However, such straightness error correction processing measures are not efficient, and straightness error correction is not always performed with high accuracy.

Further, a unique straightness error existing in a pixel array along a sub-scanning direction of a drawing pattern in each laser drawing device is measured in advance, and the straightness error correction processing is performed based on the measured error in the above-described CAD. It is also known to perform this at a station or a CAM station. Briefly, such straightness error correction processing adjusts the drawing start position by the laser beam along the main scanning direction based on the measurement error, thereby adjusting the straightness of the pixel array along the sub-scanning direction of the drawing pattern. It is intended to correct the error. However, in the straightness error correction processing performed in the CAD station or the CAM station, the adjustment of the drawing start position by the laser drawing along the main scanning direction is performed only in units of one pixel of the raster data. High-precision straightness error correction in pixel units or less cannot be expected.

Accordingly, an object of the present invention is to move the object to be drawn in the sub-scanning direction while deflecting the laser beam in the main scanning direction with respect to the object to be drawn, and to modulate the laser beam with respect to the object to be drawn by raster data. A laser writing apparatus that performs writing by controlling in accordance with a clock pulse having a predetermined frequency based on a straight line error correction process of a pixel array along a sub-scanning direction of the writing pattern in a unit smaller than one pixel of raster data. Another object of the present invention is to provide a laser writing apparatus which can be performed at low cost.

[0009]

A laser writing apparatus according to the present invention moves a workpiece in a sub-scanning direction while deflecting a laser beam in the main scanning direction with respect to the workpiece, and performs laser irradiation on the workpiece. The modulation of the beam is controlled in accordance with a clock pulse of a predetermined frequency based on the raster data, and a predetermined pattern is drawn. The straightness error of the pixel array along the sub-scanning direction is measured by a laser beam based on the measured data. A drawing start position shift data generating means for generating drawing start position shift data along the main scanning direction, wherein the drawing start position shift data includes a drawing start position along the main scanning direction by a laser beam as one pixel of raster data; A first series of drawing start position shift data shifted in units of size, and a drawing start position along a main scanning direction by a laser beam. A second series of drawing start position shift data that is shifted in units of one pixel size or less of the star data, and a measuring unit that measures the moving distance of the object to be drawn in the sub-scanning direction; Each time the measured moving distance becomes equal to each of the first series of drawing start position shift data, the drawing start position along the main scanning direction by the laser beam is shifted in a predetermined direction in units of one pixel size. And a writing start position along the main scanning direction by the laser beam every time the moving distance measured by the measurement means becomes equal to each of the second series of writing start position shift data. The second to shift in a predetermined direction in units smaller than the size
And a drawing start position shifting means.

In the above-described laser writing apparatus, when the writing start position of the laser beam along the main scanning direction is shifted toward the deflection direction of the laser beam, the measured straightness error data of the pixel array along the sub-scanning direction can be obtained. At this time, the shift of the drawing start position by the first and second drawing start position shift means is performed on the negative side, and the drawing start position of the laser beam along the main scanning direction is determined by the deflection direction of the laser beam. Is shifted to the opposite side, the measured straightness error data of the pixel array along the sub-scanning direction is negative, and at this time, the shift of the drawing start position by the first and second drawing start position shift means is positive. Done in

In the above-described laser drawing apparatus, the first drawing start position shift means can be configured to advance or delay the read timing by one pixel when raster data is read from the memory.

In the above-described laser writing apparatus, the second writing start position shift means generates clock pulses of a predetermined number of clock pulses whose phases are shifted at predetermined intervals as clock pulses, and the clock pulse generation means. Of the predetermined number of clock pulses obtained in
And a clock pulse output from the clock pulse output means for shifting the drawing start position along the main scanning direction of the laser beam by a unit of one pixel size or less in a predetermined direction. Clock pulse output control means for controlling the output of the clock pulse so as to sequentially shift the phase of the clock pulse by the predetermined number.

[0013]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, an embodiment of a laser drawing apparatus according to the present invention will be described with reference to the accompanying drawings.

FIG. 1 schematically shows a laser writing apparatus according to the present invention as a perspective view, which directly draws a circuit pattern on a photoresist layer on a substrate for manufacturing a printed circuit board. It is configured as follows.

The laser drawing apparatus has a base 10 mounted on the floor, and a pair of rails 12 are installed on the upper surface of the base 10 in parallel. X on a pair of rails 12
The X table 14 is mounted on a pair of rails 12 by a suitable drive motor not shown in FIG. 1, such as a servo motor or a stepping motor.
In the longitudinal direction, that is, the X direction.

A drawing table 18 is installed on the X table 14 via a θ table 16, and two fine adjustment drivers 20 are provided on each of the sides facing each other between the X tables 14. The fine adjustment of the angular position in the horizontal plane is performed. Note that FIG. 1 shows only two fine adjustment drivers 20 provided on one side to avoid complication of the drawing.

On the drawing table 18, a substrate having a photoresist layer is conveyed as an object to be drawn by a suitable conveying means such as a belt conveyor.
It is secured above by suitable clamping means. In FIG. 1, a clamp member 2 forming a part of the clamp means is shown.
2 is shown.

On one side of the base 10, an argon laser generator 24 is installed as a laser light source. The laser beam LB emitted from the argon laser generator 24 is deflected upward by a beam bender 26. On the other hand, a fixed table plate 28 supported by a suitable support structure (not shown) is disposed above the drawing table 18.
Various optical elements for processing the laser beam LB deflected by the beam bender 26 are provided on the fixed table 28. In the present embodiment, the argon laser generator 24 is of a water-cooled type, the output is 1.8 W, and the wavelength of the laser is 488 nm.

The fixed table plate 28 has a beam bender 30
The beam bender 30 is provided with the beam bender 26.
And is directed to the beam splitter 32. The beam splitter 32 splits the laser beam LB into two laser beams LB1 and LB2. Laser beam LB1 is directed to beam separator 38 via beam benders 34 and 36, and laser beam LB2 is directed to beam separator 46 via beam benders 40, 42 and 44.

The beam separator 38 is a laser beam LB
1 is divided into four parallel laser beams, and similarly, the beam separator 46 divides the laser beam LB2 into four parallel laser beams. The parallel laser beam from the beam separator 38 is guided to the electronic shutter 52 by the beam benders 48 and 50, and the parallel laser beam from the beam separator 46 is guided to the electronic shutter 58 by the beam benders 54 and 56.

Each of the electronic shutters 52 and 58 is formed of four acousto-optic elements, and each acousto-optic element is assigned a corresponding one of the four laser beams. The four lasers that have passed through the electronic shutter 52 are
0, while passing through the electronic shutter 58, the four laser beams are passed through the beam bender 62 to the photosynthesizer 60.
To be incident. The light combiner 60 is configured as, for example, a polarization beam splitter. The four laser beams that have passed through the electronic shutters 52 and 58 are combined into eight laser beams by the light combiner (polarization beam splitter) 60. The eight laser beams are supplied by beam benders 64 and 6
The light is incident on the polygon mirror 70 via 6 and 68, and is deflected by the respective rotating reflection surfaces over a predetermined scanning range.

The eight laser beams deflected by the respective rotating reflection surfaces of the polygon mirror 70 over a predetermined scanning range are first passed through the fθ lens 72, and then deflected by the turning mirror 74 to the drawing table 18 side. After that, the light is made to reach the drawing table 18 via the condenser lens 76. In short, the eight laser beams are converted into Y light by the respective rotating reflection surfaces of the polygon mirror 70.
It is deflected in the axial direction, that is, in the main scanning direction.

Accordingly, if the substrate having the photoresist layer is placed on the drawing table 18 as described above, the surface of the photoresist layer on the substrate is deflected by the respective rotating reflection surfaces of the polygon mirror 70. Scanning (main scanning direction) is performed at once by eight laser beams. Thus, the four acousto-optic elements of each of the electronic shutters 52 and 58 are operated according to the clock pulse of the predetermined frequency based on the raster data, and when the eight laser beams are modulated, the surface of the photoresist layer is formed. A predetermined circuit pattern based on such raster data is drawn.

During scanning in the main scanning direction by eight laser beams, the drawing table 18
4 sequentially moves along the X-axis direction, that is, the sub-scanning direction, and when the scanning in the main scanning direction by the eight laser beams is completed, the moving distance of the drawing table 18 is equal to 8
The distance corresponds to the width of the book laser beam. Thus, by repeating the scanning in the main scanning direction by the eight laser beams, a predetermined circuit pattern is sequentially drawn on the surface of the photoresist layer on the substrate by the scanning in the main scanning direction by the eight laser beams. .

Assuming that the main scanning direction and the sub-scanning direction are perpendicular to each other, a scanning line in the main scanning direction by eight laser beams is inclined with respect to the X-axis direction (sub-scanning direction). That is, as described above, while the scanning in the main scanning direction is performed by the eight laser beams, the drawing table 18 is sequentially moved by the X table 14 at a predetermined speed along the X axis direction, that is, the sub scanning direction. Because it can be moved. However, in actuality, the main scanning direction is tilted by a predetermined angle with respect to the X-axis direction in advance, so that a scanning line in the main scanning direction by eight laser beams is perpendicular to the sub-scanning direction.

Referring to FIG. 2, there is shown a block diagram of a laser writing apparatus according to the present invention.
Reference numeral 78 indicates a system control circuit. The system control circuit 78 includes, for example, a microprocessor such as a central processing unit (CPU) and a memory (RO).
M, RAM) and the like.

[0027] Two CCD cameras 80 are connected to the system control circuit 78 via an image processing circuit 82. Each of the CCD cameras 80 has a C
As shown in FIG. 1, the camera is a small camera composed of a CD (charge-coupled device), and is disposed at a fixed position above the drawing table 18 and close to the condenser lens 76.

The CCD camera 80 reads the positioning marks provided at each of the four corners of the substrate having the photoresist layer as an image, and outputs the image signal to an image processing circuit 82.
Is taken into the system control circuit 78.
The system control circuit 78 obtains the coordinates of the positioning mark based on the video signal obtained from the positioning mark, whereby the position of the substrate on the drawing table 18 is confirmed, and the circuit is placed at an appropriate position on the substrate. A pattern could be drawn.

The straightness correction processing circuit 84 is connected to the system control circuit 78, and the straightness correction processing circuit 84 is shown in detail in FIG. As can be seen from the figure, the straightness correction processing circuit 84 is provided with two delay lines 84A and 84B. Delay line 84A
, The basic clock pulse YCK-IN is input via the buffer 84C. Similarly, the basic clock pulse YCK-IN is also input to the delay line 84B via the inverter 84D. In short, the phase of the basic clock pulse input to the delay line 84B is the basic clock pulse YCK-IN
Is shifted by π.

In the present embodiment, the delay line 84A has five clock pulses YCK-SFT1, YCK-SFT2, YCK-SFT3, YCK-SFT4 and YCK-SFT1 based on the basic clock pulse YCK-IN.
It is designed to output SFT5. As shown in the time chart of FIG. 4, the clock pulse YCK-SFT1 has the same phase as the basic clock pulse YCK-IN,
The phase of SFT2 is shifted to the positive side by π / 5 with respect to the clock pulse YCK-SFT1. Similarly, each of the clock pulses YCK-SFT3 to YCK-SFT5 is sequentially shifted to the positive side by π / 5 with respect to the immediately preceding clock pulse. Thus, the phase of clock pulse YCK-SFT5 is shifted by 4π / 5 with respect to clock pulse YCK-SFT1 (basic clock pulse YCK-IN).

On the other hand, the delay line 84B has five clock pulses YCK-SFT6, YCK-SFT7, YC based on a clock pulse shifted by π with respect to the basic clock pulse YCK-IN.
It outputs K-SFT8, YCK-SFT9 and YCK-SFT10. As shown in the time chart of FIG. 4, the clock pulse YCK-SFT6 has the same phase as the clock pulse input to the delay line 84B, and the clock pulse YCK-SF
The phase of T7 is shifted to the positive side by π / 5 with respect to the clock pulse YCK-SFT6. Similarly, the clock pulse YCK-SFT7 or
Each of the YCK-SFTs 10 also sequentially shifts to the positive side by π / 5 with respect to the immediately preceding clock pulse. Thus, the phase of clock pulse YCK-SFT10 is 9π / 5 (18π / 10) with respect to clock pulse YCK-SFT1 (basic clock pulse YCK-IN).
The phase is shifted only to the positive side.

In short, the delay lines 84A and 84A
10 types of clock pulses YCK-SF output from both B
T1, YCK-SFT2, YCK-SFT3, YCK-SFT4, YCK-SFT5, YCK-SF
T6, YCK-SFT7, YCK-SFT8, YCK-SFT9 and YCK-SFT10 are sequentially shifted in phase by π / 5 to the positive side. When the phase of the clock pulse YCK-SFT10 is further shifted by π / 5, the phase becomes the same as the phase of the basic clock pulse YCK-IN (that is, the clock pulse YCK-SFT1).

As shown in FIG. 3, the straightness correction processing circuit 8
4 is further provided with a pulse output control circuit 84E and a multiplexer 84F, from which four types of selection signals Y-SEL1, Y-SEL2, Y-SEL3 and Y-SEL4 are output to the multiplexer 84F. It is supposed to be. The output levels of the four types of selection signals Y-SEL1, Y-SEL2, Y-SEL3, and Y-SEL4 from the pulse output control circuit 84E are output from the system control circuit 78 to the pulse output control circuit 8
Control is performed based on a command signal output to 4E.

In short, the system control circuit 78
4 types of selection signals Y-SEL1, Y-SEL2, Y-SEL3 and Y-SE from the pulse output control circuit 84E based on the command signal from
By changing the combination of the output levels of L4, 10 types of clock pulses YCK-SFT1, YCK-
SFT2, YCK-SFT3, YCK-SFT4, YCK-SFT5, YCK-SFT6, YCK-
One of SFT7, YCK-SFT8, YCK-SFT9 and YCK-SFT10 is output. For example, in this embodiment, four types of selection signals Y-SEL1 and Y-SE from the pulse output control circuit 84E are provided.
The relationship between the combination of the output levels of L2, Y-SEL3 and Y-SEL4 and the clock pulse output from the multiplexer 84F is as shown in Table 1 below.

[0035]

[Table 1]

As is clear from Table 1, for example, in order to output the clock pulse YCK-SFT3 from the multiplexer 84F, the selection signals Y-SEL3 and Y-SEL4 are set to low level (L) and the other selection signals Y -SEL1 and Y-SEL2 may be set to high level (H).
In order to output the clock pulse YCK-SFT1, only the four types of selection signals Y-SEL1 need to be set to the high level (H).

As shown in FIG. 2, the system control circuit 78 and the straightness correction processing circuit 84 control the operation of the main scanning control circuit 86, so that the drawing pattern follows the sub-scanning direction during the drawing operation of the laser drawing apparatus. The straightness correction processing of the pixel array is performed as described later in detail.

The main scanning control circuit 86 is provided with a beam position control circuit 90. The beam position control circuit 90 has a buffer memory 90A and a synchronization circuit 90 as shown in FIG.
B is included. At the time of the drawing operation, the raster data output from the raster conversion circuit 92 (FIG. 2) is sequentially written and temporarily stored in the buffer memory 90A, and the raster data is sequentially read from the buffer memory 90A and synchronized. Output to circuit 90B. Writing of raster data to the buffer memory 90A is performed based on a write clock pulse output from the system control circuit 78 to the buffer memory 90A. Reading of raster data from the buffer memory 90A is performed by the system control circuit 78. This is performed based on the read clock pulse output to the memory 90A.

The control unit of the laser drawing apparatus is provided with, for example, a hard disk device (not shown) as data storage means. The hard disk device has circuit pattern data (vector data) created and processed by a CAD station or a CAM station. During the drawing operation, the corresponding circuit pattern data is read from the hard disk device and converted into raster data by the raster conversion circuit 92. In the present embodiment, since the drawing is performed by eight laser beams during the drawing operation, the buffer memory 90A has a capacity capable of holding raster data for at least eight lines in the main scanning direction. You.

As shown in FIG. 5, electronic shutters 52 and 58 are connected to the synchronizing circuit 90B, respectively, and the synchronizing circuit 90B receives raster data of eight lines in the main scanning direction read from the buffer memory 90A. The multiplexer 8 of the straightness correction processing circuit 84 (FIG. 3) that has been input
4 types of clock pulses from 4F YCK-SFT1, YCK-SFT2,
YCK-SFT3, YCK-SFT4, YCK-SFT5, YCK-SFT6, YCK-SFT7,
One of YCK-SFT8, YCK-SFT9 and YCK-SFT10 is input. Thus, a control voltage signal is output from the synchronization circuit 90B to each of the four acousto-optic elements included in each of the electronic shutters 52 and 58 based on the raster data.

More specifically, when each of the control voltage signals output from the synchronizing circuit 90B is input to each of the four acousto-optical element driving circuits built in the electronic shutter 52, each of the acousto-optical element driving circuits outputs The high-frequency drive voltage is output and applied to the corresponding acousto-optic element. The voltage level of the control voltage signal output to each acousto-optic element driving circuit is changed based on the corresponding raster data, whereby the diffraction direction of the laser beam passing through each acousto-optic element is changed. For example, when the pixel of the raster data is a color pixel (ie, “1” as digital pixel data), the laser beam is diffracted toward the light combiner 60 and the pixel of the raster data is a non-color pixel (ie, When the digital pixel data is “0”), the laser beam is diffracted so as to be out of the light combiner 60. On the other hand, in the case of the electronic shutter 58, when the pixel of the raster data is a color pixel (that is, “1” as digital pixel data), the laser beam is diffracted toward the beam bender 62, and When the pixel is a colorless pixel (ie, “0” as digital pixel data), the laser beam is diffracted off the beam bender 62.

In short, the laser beam is directed to the polygon mirror 70 only when the pixel of the raster data is a coloring pixel, and the laser beam is used to draw the drawing table 1.
On the object to be drawn on the dot 8, dots are recorded as coloring pixels. Therefore, by modulating the laser beam by the electronic shutters 52 and 58 based on the raster data as described above, a circuit pattern based on the raster data is drawn on the drawing object on the drawing table 18.

It should be noted that the synchronization circuit 90B
The output timing of the control voltage signal output to the respective acousto-optical element drive circuits of the electronic shutters 52 and 58 from the multiplexer 84 of the straightness correction processing circuit 84
F to the synchronization circuit 90B.
K-SFT1, YCK-SFT2, YCK-SFT3, YCK-SFT4, YCK-SFT5, YC
K-SFT6, YCK-SFT7, YCK-SFT8, YCK-SFT9 and YCK-SFT10
).

More specifically, when the output of the clock pulse from the multiplexer 84F is switched from, for example, the clock pulse YCK-SFT1 to the clock pulse YCK-SFT2 at the start of the drawing by the laser beam along the main scanning direction,
The output timing of the control voltage signal from the synchronization circuit 90B is delayed by a time corresponding to the phase difference π / 5 minutes between the clock pulse YCK-SFT1 and the clock pulse YCK-SFT2, and the laser beam along the main scanning direction by the delay time Is shifted to the positive side. On the other hand, the output of the clock pulse from the multiplexer 84F is, for example, the clock pulse YCK-
Assuming that the clock is switched from SFT1 to clock pulse YCK-SFT10, the output timing of the control voltage signal from the synchronization circuit 90B becomes clock pulse YCK-SFT1 and clock pulse YCK-SFT10.
The time is advanced by the time corresponding to the phase difference π / 5 minutes from SFT2,
The drawing start position by the laser beam along the main scanning direction is shifted to the negative side by the advance time. Here, for convenience, the case where the drawing start position by the laser beam along the main scanning direction is shifted in the deflection direction of the laser beam is defined as the positive side.

In short, in the present embodiment, at the start of drawing with a laser beam along the main scanning direction, the clock pulses (YCK-SFT1, YCK-SFT2, YCK-SFT2, YCK-SFT2) output from the multiplexer 84F of the straightness correction processing circuit 84 to the synchronization circuit 90B YCK-
SFT3, YCK-SFT4, YCK-SFT5, YCK-SFT6, YCK-SFT7, YCK-
By sequentially shifting the phase of SFT8, YCK-SFT9, and YCK-SFT10) to the positive side or the negative side by π / 5, the drawing start by the laser beam along the main scanning direction is reduced to 1/10 of the pixel (dot) size. Can be adjusted in units of

As shown in FIG. 2, the main scanning control circuit 86 is further provided with a Y scale sensor 94 and a signal processing circuit 96. The Y scale sensor 94 detects an optical signal from a Y linear scale (not shown) to measure the deflection distance of the laser beam along the main scanning direction, and is well known in itself. Y scale sensor 94
After being appropriately processed by the signal processing circuit 96, the output signal is input to the straightness correction processing circuit 84 (FIG. 3) as the basic clock pulse YCK-IN.

As is apparent from FIG. 2, the sub-scanning control circuit 88 is provided with a servo motor driving circuit 98,
The servo motor driving circuit 98 controls the servo motor 10
0 is controlled. The servo motor 100 is for moving the X table 14 at a predetermined speed in the X-axis direction, that is, in the sub-scanning direction, whereby the object to be drawn on the drawing table 18 is moved in the sub-scanning direction. During the drawing operation, a clock pulse of a predetermined frequency is output from the system control circuit 78 to the servo motor driving circuit 98, and a driving pulse is output from the servo motor driving circuit 98 to the servo motor 100 based on the clock pulse. You.

As shown in FIG. 2, the sub-scanning control circuit 88 further includes an X scale sensor 102 and a signal processing circuit 104.
Is provided. The X scale sensor 102 detects an optical signal from an X linear scale (not shown) and measures the moving distance of the drawing table 18 (that is, the object to be drawn thereon) along the main scanning direction. Itself is well known. The output signal from the X-scale sensor 102 is appropriately processed by the signal processing circuit 104 and then taken into the system control circuit 78, and a clock pulse to the servo motor driving circuit 98 is created based on the output. In the system control circuit 78,
Based on the output signal from the X scale sensor 102, the moving distance of the drawing table 18 (drawing target) is also detected.

Next, the principle of the straightness error correction processing executed according to the present invention at the time of drawing operation will be described with reference to FIGS. 6 to 14 of the accompanying drawings.

In the graph of FIG. 6, the Y axis indicates the main scanning direction, and the X axis indicates the sub scanning direction. Further, in the graph of the figure, the width of the recording area along the X-axis direction, that is the sub-scanning direction by the laser beam is indicated by X _P. Here one pixel size (i.e., the size of the recording dots) if is assumed to be D _P, the sub-scanning direction line of such recording area X
_P / D _P pixels will be arranged. For example, the recording area width X _P described above is 480 mm, yet when the pixel size is the D _P is 5 [mu] m, so that the 96,000 pixels are arranged in the sub-scanning direction line of 480 mm. In other words, the recording area width X _P described above along the Y-axis direction 96,000
The main scanning direction line of the book will be included.

In this case, as described above, the drawing start position of each line in the main scanning direction is not aligned on a straight line along the X-axis direction, that is, the sub-scanning direction. That is, X in the graph of FIG.
Even if the axis is the drawing start position along the main scanning direction by the laser beam, the drawing start position of each main scanning direction line is actually shifted in the main scanning direction, that is, the Y-axis direction as shown in the graph of FIG. Such a deviation, that is, a straightness error is unique to each laser writing apparatus. In the example of the graph shown in FIG. 6, the drawing start position of each line in the main scanning direction is generally shifted to the positive side with respect to the Y axis, but it goes without saying that the drawing start position may actually be shifted to the negative side.

According to the present invention, first, the recording area width _XP is equally divided by _a , and straightness error correction processing is performed in each section Xa in units of one pixel size (dot). For example, when 8 equally divided recording area width 480mm as described in the case described above, each section X _a is 60mm, and the straightness by one pixel size (5 [mu] m) units for each of its 60mm interval X _a An error correction process is performed. In other words, the straightness error correction processing by a unit of one pixel size in the main scanning direction line of 12,000 included in each 60mm interval X _a is performed. The straightness error correction processing in such a unit of one pixel size will be specifically described below.

[0053] In Figure 6, the deviation amount of the drawing start position of the last main scanning direction line of the m-th interval X _a is represented by E _m,
The shift amount of the drawing start position of the last main scanning direction line in the (m-1) th section is indicated by _Em-1 . Difference in displacement amount E _m-1 for the time shift amount E _m, i.e. straight trueness error Delta] E _m of the total of the m-th interval X _a can be expressed by the following equation. ΔE _m = E _m −E _{m -1}

[0054] Based on the straight trueness error Delta] E _m at m th interval X _a, the ratio of the straightness error for that segment Delta] Tx _m
Is obtained as follows. ΔTX _m = ΔE _m / X _a

As described above, in the present invention, each section X _a
First, the straightness error correction processing is performed with one pixel size D _p
Therefore, the position where the straightness error correction processing is performed is a position where there is a shift amount of one pixel size _Dp or more. To be specific, for example, as shown in FIG. 7, the entire amount of deviation of the m-th interval X _a segment at the point where the deviation amount from the X-axis is an integral multiple of a pixel size D _p Is done.

[0056] Further, as shown in FIG. 7, the interval X _a may be classified in accordance with the shift amount of a pixel size D _p min, they segmented region is denoted by X _c1, X _c2 and X _c3, these classification The regions X _C1 , X _C2 and X _C3 can be represented by the following equations. X _C1 = - In (D _P mod [E _m-1 / D _{p]) / ΔTX m X C2 = D} P / ΔTX m X C3 = mod [E _m / D _p] / Delta] Tx _m in the above formula, mod [ ...] indicates a "remainder" when the division in parentheses is performed by an integer operation.

In the case described above, 1 is _added to the section Xa.
2,000 lines in the main scanning direction will be included,
Of the 12,000 main scanning direction lines, 2,000 are allocated to the divided area X _C1, and each of the two divided areas X _C2 is allocated to 4,000 main scanning direction lines. The area X _C3 will be filled with 2,000 lines in the main scanning direction.

In the example shown in FIG. 7, the divided areas X _C1 and X _C2
Since the deviation amount from the X axis at the end point of an integral multiple of a pixel size D _p, straightness error correction processing by a unit of one pixel size (D _p) for those segmented region X _C1 and X _C2 Is applied. That is, linearity errors processing is performed at the point of drawing start position on the segment grid when divided by integer multiples of the pixel size D _p the shift amount is riding. However, since the drawing start position displacement amount in the segmented region X _C3 is less than one pixel size, not straightness error correction processing by a unit of one pixel size (D _p) is subjected for the said section area X _C3 . The straightness error correction processing in units of one pixel size is performed by advancing the drawing start of the main scanning direction line included in the corresponding segmented area (X _C1 , X _C2 ) by one pixel size, which is performed by beam position control. Circuit 90
This is achieved by accelerating the reading of raster data from the buffer memory 90A by one pixel.

In the example shown in FIG. 7, the straightness error is Y
Since it appears on the positive side with respect to the axis, the straightness error correction processing is performed by advancing the drawing start of the main scanning direction line included in the corresponding segmented area (X _C1 , X _C2 ) by one pixel size. When the straightness error appears on the negative side with respect to the Y axis, the straightness error correction process is performed by delaying the start of drawing the main scanning direction line included in the corresponding segmented area by one pixel size, This is achieved by delaying the reading of the raster data from the buffer memory 90A of the beam position control circuit 90 by one pixel.

[0060] m th interval X _a number of straightness error correcting i.e. one pixel of the rendering start early treatment with a pixel size units in g (m) is expressed by the following equation. g (m) = int [E _m / D _P ] ─int [E _m-1
/ D _P ] In the above formula, int [...] indicates the "quotient" of the division in parentheses.

[0061] Moreover, the start position of one pixel of the drawing start early treatment with m th interval X _a (i.e., the segmented region X _c1, X
_c2, when the X end position of _c3) the distance X _s from the coordinate origin, the distance is given by the following equation.

[0062]

(Equation 1)

[0063] Referring to FIG. 8, there is shown a m th interval X _m after the straightness error correction processing by a unit (D _p) of a pixel size, as described above. As can be seen from the figure, as a result of the straightness error correction processing in units of one pixel size, the shift amount of the drawing start position in the X-axis direction, that is, the sub-scanning direction, is suppressed to one pixel size _Dp or less. .

[0064] Then, according to the present invention, in each of the partitioned regions of each section _{X a (X c1, X c2} , X c3), straightness correction in one pixel size following units (D _p / n) process Is performed. In this embodiment, since n = 10, the straightness correction processing for one pixel size or less is 0.5 μm
Will be performed in units of This will be specifically described below.

First, looking at the segmented area X _c2 ,
As shown in FIG. 9, with respect to the shift amount (D _p ) for one pixel in the _divided area X _c2 , the position where the shift amount from the X axis is an integral multiple of a unit (D _p / n) equal to or smaller than one pixel size. Is subdivided. That is, the shift amount (D _p ) for one pixel in the _divided area X _c2.
Is subdivided into n equal parts (D _p / n). On the other hand, the segmented area _{Xc2 is} also subdivided according to the shift amount of a unit ( _Dp / n) equal to or smaller than one pixel size, and these subdivided area can be represented by _Xc2 / n. In short, as is apparent from FIG. 9, the deviation amount from the X-axis is an integer multiple of D _p / n at the end of each subdivision segmented region X _c2 / n.

The straightness error correction processing using a unit (D _p / n) equal to or smaller than one pixel size is performed for each subdivided area X _c2 / n. In the case of the above case, the subdivision area X
each of _c2 / n is included 400 in the main scanning direction line, straightness error correction processing by the size of one pixel or less of the unit (D _p / n) for these 400 in the main scanning direction line is subjected You. In short, the shift amount is set to one pixel size D _p / n
The straightness error processing is executed at a position where the drawing start position is on the D _p / n division grid when the division is performed by an integer multiple of. In such straightness error correction processing using a unit of one pixel size or less, the start of drawing the main scanning direction line included in each subdivided divided area _Xc2 / n is performed by a unit of one pixel size or less ( _Dp / n). This is achieved by shifting the phase of the clock pulse output from the multiplexer 84F of the straightness correction processing circuit 84 to the negative side by π / 5.

Referring to FIG. 10, there is shown a sectioned area _Xc2 after the straightness error correction processing is performed using the unit (D _p / n) equal to or smaller than one pixel size as described above. In addition, as a result of the straightness error correction processing in units of one pixel size or less, the deviation amount of the drawing start position in the X-axis direction, that is, the sub-scanning direction, is D _p / n (in the present embodiment,
0.5 μm) or less. In short, the segmented area X _c2
Are approximated in a zigzag manner by a unit ( _Dp / n) of one pixel size or less with respect to the X-axis in the main scanning direction line.

Similarly, as shown in FIG.
The shift amount for one pixel at _c1 is also subdivided at a position where the shift amount from the X axis is an integral multiple of a unit ( _Dp / n) equal to or smaller than one pixel size. Further, the segmented area X _{c1 is} also subdivided according to the shift amount of a unit (D _p / n) smaller than one pixel size,
These subdivision areas can be represented by _Xc2 / n in this example. Because, even because the slope of the displacement is the same as the segmented region X _c2 in segmented region X _c1, subdivided divisional area corresponding to the shift amount of a pixel size less units (D _p / n) is X _c2 / This is because n. In short, as it is clear from FIG. 11, subdivided in the segmented region X _c1 segmented region X _c2 /
Each of n is a section in which a relative subdivision deviation amount (D _p / n) is generated _.

[0069] Incidentally, the left-most subdivision segmented region X _d1 of segmented region X _c1 shown in FIG. 11 is expressed by the following equation. X _d1 = mod [X _c1 / (X _c2 / n)]

[0070] partitioned area subdivision division area in the X _c1 (X _d1,
Also for each of the X _c2 / n), straightness error correction processing by the size of one pixel or less of the unit (D _p / n) is performed in a manner similar to the above case. FIG. 12 shows the segmented area _Xc1 after the straightness error correction processing is performed. As is apparent from FIG. 12, the shift amount of the drawing start position in the X-axis direction, that is, the sub-scanning direction is D _p / n (in the present embodiment, 0.
5 μm) or less. In short, the drawing start position of the line in the main scanning direction included in the _divided area _Xc1 is approximated in a zigzag manner with a unit ( _Dp / n) smaller than one pixel size with respect to the X axis.

[0071] Further, as shown in FIG. 13, segmented region X _c3
Is also subdivided at a position where the shift amount from the X axis is an integral multiple of a unit (D _p / n) equal to or smaller than one pixel size. Further, the segmented area _{Xc3 is} also subdivided according to the shift amount of a unit ( _Dp / n) equal to or smaller than one pixel size, and the subdivided area is also _{Xc2 //} for the same reason as described above.
n. In short, as is clear from FIG. 13, each subdivision area _Xc2 / n is a section in which a relative subdivision deviation amount ( _Dp / n) is generated _.

It should be noted that the subdivision area where the relative subdivision deviation amount is equal to or less than D _p / n, that is, the subarea X shown in FIG.
The rightmost subdivision region of _c3 is indicated by X _d3 , and the shift amount from the X axis at the end point of the subdivision region X _d3 is an integer of a unit (D _p / n) equal to or smaller than one pixel size. since multiplication and not, straightness error correction processing by the size of one pixel or less of the unit (D _p / n) in such a subdivision segmented region X _d3 is not executed.

[0073] partitioned area subdivision division area in the X _c3 (X _d3,
Also for each of the X _c2 / n), straightness error correction processing by the size of one pixel or less of the unit (D _p / n) is performed in a manner similar to the above case. FIG. 14 shows the segmented area _Xc3 after the straightness error correction processing is performed. As is apparent from FIG. 14, the shift amount of the drawing start position in the X-axis direction, that is, the sub-scanning direction, is D _p / n (in the present embodiment, 0.
5 μm) or less. In short, the drawing start position of the line in the main scanning direction included in the _divided area _Xc3 is approximated in a zigzag manner with a unit ( _Dp / n) smaller than one pixel size with respect to the X axis.

If the k-th subdivision area located from the coordinate origin is X _k (X _c2 / n), a straight line in the sub-division area X _k in units of one pixel size or less (D _p / n). degree correction process (i.e., a clock pulse phase to speed up the drawing start position in the main scanning direction line by D _p / n π /
Number of times h (k) of drawing start advance processing shifted by 5)
Can be expressed by the following equation. h (k) = int [E _k × n / D _P ] ─int [E
_k-1 × n / D _P ] In the above equation, int [...] indicates the “quotient” of the operation in parentheses.

Further, D in the k-th subdivision area X _k
The start position of the drawing start advance processing of _p / n (that is, the start point position of each subdivision area X _k ) is determined by the distance X _t from the coordinate origin.
Then, the distance is given by the following equation.

[0076]

(Equation 2)

As described above, the straightness error differs for each laser writing apparatus. Therefore, the straightness error in each laser writing apparatus can be known by actually obtaining a writing pattern and measuring the straightness. For example, a photographic emulsion is coated on a plate glass, and a lattice pattern is actually drawn as a drawing pattern on the surface of the photographic emulsion using a corresponding laser drawing apparatus, and after the grid pattern is developed, the straightness of the grid pattern is measured. Then, straightness error data can be obtained based on the actual measurement data.

Such a series of straightness error data is stored in advance in a hard disk device provided in a control unit of a laser drawing device, for example, and is taken into the system control circuit 78 from the hard disk device during drawing operation.

Next, a drawing operation routine of the laser drawing apparatus according to the present invention will be described with reference to flowcharts shown in FIGS.

First, in step 1501, the drawing table 18 is moved from the origin position.
At 502, the positioning mark of the object to be drawn (that is, the substrate having a photoresist layer) on the drawing table 18 is a CCD.
It is determined whether or not detection has been performed by the camera 80. C
When the detection of the positioning mark by the CD camera 80 is confirmed, the process proceeds from step 1502 to step 1503,
Then, the coordinate data of the positioning mark is taken into the system control circuit 78 through the CCD camera 80, and the position of the object to be drawn on the drawing table 18 is calculated.

At step 1504, it is determined whether or not the calculation of the position of the object to be drawn has been completed. When the completion of the calculation is confirmed, the process proceeds to step 1505, where the drawing table 18 is moved to the drawing start position. Can be

At step 1506, desired straightness error data is read out from the hard disk device provided in the control unit of the laser drawing apparatus and is taken into the system control circuit 78. based on the position data of the body, the shift start position data X _s and X _t is calculated. Then, step 15
At 07, a series of drawing start position shift data X _s, X _t
It is determined whether or not the calculation has been completed, and if it is confirmed that the calculation has been completed, the process proceeds to step 1508.

In step 1508, it is determined whether or not a drawing start command has been input to the system control circuit 78 via a keyboard (not shown). When the input of the drawing start command has been confirmed, the flow advances to step 1509, where it is determined. The drawing operation of one control unit along the main scanning direction is started.

At step 1510, the X scale read data C _x from the X scale sensor 102 is sampled at predetermined time intervals and sequentially updated. Then, in step 1511, X scale reading data C _x is whether equal to the arithmetic data X _s for respective sampling is determined. X Unless scale reading data C _x has reached the shift start position data X _s, step 1512
Proceeds to where reading data C _x from X scale sensor 102 whether equal to X _t is determined. If not X scale reading data C _x reaches the shift start position data X _t, the process proceeds to step 1513, where rendering operation is whether the end is determined. If the drawing operation has not been completed, the process returns to step 1509.

[0085] In summary, read data C _x is the shift start position data from the X scale sensor 102 X _s and X _t
If none of the above has been reached, the drawing operation is performed in a normal manner. That is, the raster data is read from the buffer memory 90A in a normal manner, and the modulation of the laser beam is performed based on the raster data.
This is performed in accordance with, for example, the clock pulse YCK-SFT1 output from the fourth multiplexer 84F.

[0086] In step 1511, if the read data C _x from X scale sensor 102 has reached the X _s is confirmed, step from step 1511 15
Then, the output of the clock pulse from the multiplexer 84F is switched, and the phase thereof is shifted by π / 5, whereby the drawing start position along the main scanning direction is shifted by D _p / n. Depends on the sign of the straightness error. In short, when the straightness error is on the positive side, the clock pulse changes from YCK-SFT1 to YC
The K-SFT 10 is switched (that is, the phase of the clock pulse is shifted to the negative side by π / 5), whereby the drawing start position along the main scanning direction is shifted to the negative side by D _p / n, while the straightness is changed. When the error is on the negative side, the clock pulse is switched from YCK-SFT1 to YCK-SFT2 (that is, the phase of the clock pulse is shifted by π / 5 to the positive side), thereby changing the clock pulse in the main scanning direction. The drawing start position along is D _p
/ N is shifted to the positive side. After such a shift control of the drawing start position along the main scanning direction, i.e. straightness error correction processing by the units D _p / n below one pixel size is performed, the process proceeds to step 1515, where the count value of the counter s is The count is incremented by “1”, and thereafter, the process proceeds to step 1512.

[0087] Further, in step 1512, when the read data C _x from X scale sensor 102 has reached the X _t, the process proceeds from step 1512 to step 1516, where the read start timing for reading raster data from the buffer memory 90A It is advanced or delayed by one pixel. In short, when the straightness error is on the positive side, the read start timing when raster data is read from the buffer memory 90A is advanced by one pixel, and thereby the drawing start position along the main scanning direction is D _p. In the case where the straightness error is on the negative side, the read start timing when raster data is read from the buffer memory 90A is delayed by one pixel. drawing start position along is shifted to the positive side by D _p. After such a shift control of the drawing start position along the main scanning direction, that is, the straightness error correction processing by a pixel size D _p has been performed, the process proceeds to step 1517, where the count value of the counter t is only "1" count Up, then step 15
Return to 09.

In short, in the routine consisting of steps 1509 to 1517, straightness error correction processing as described with reference to FIGS. 7 to 14 is sequentially executed. In step 1513, when the end of the drawing operation is confirmed, steps 1513 to 151
Proceed to 8 where the counter s is reset, then proceed to step 1519 where the counter t is reset. Subsequently, in step 1520, the drawing table 1
8 is returned to the origin position, and this drawing operation routine ends.

[0089]

As is apparent from the above description, in the laser writing apparatus according to the present invention, the inherent straightness error correction processing accompanying the laser writing apparatus itself can be performed in units of one pixel size or less and at low cost. Thus, the image quality of the drawing pattern can be greatly improved.

[Brief description of the drawings]

FIG. 1 is a schematic perspective view showing one embodiment of a laser drawing apparatus according to the present invention.

FIG. 2 is a block diagram of the laser drawing apparatus shown in FIG.

FIG. 3 is a detailed block diagram of a straightness correction processing circuit shown in FIG. 5;

4 is a time chart of a straightness correction processing clock pulse created when the straightness correction processing circuit of FIG. 3 operates.

FIG. 5 is a detailed block diagram of a beam position control circuit shown in FIG. 2;

FIG. 6 is a graph conceptually illustrating a straightness error of the laser writing apparatus according to the present invention.

FIG. 7 is a graph for explaining the principle of straightness error correction processing in units of one pixel size executed by the laser writing apparatus according to the present invention.

FIG. 8 is a graph conceptually showing a result of straightness error correction processing in units of one pixel size executed by the laser drawing apparatus according to the present invention.

FIG. 9 is a graph for explaining the principle of straightness error correction processing in units of one pixel size or less executed by the laser writing apparatus according to the present invention.

10 is a graph conceptually showing a result of the straightness error correction processing performed in units of one pixel size or less executed in FIG. 9;

FIG. 11 is another graph for explaining the principle of straightness error correction processing in units of one pixel size or less executed by the laser writing apparatus according to the present invention.

FIG. 12 is a graph conceptually showing a result of the straightness error correction processing performed in units of one pixel size or less executed in FIG. 11;

FIG. 13 is still another graph for explaining the principle of the straightness error correction processing performed in the unit of one pixel or less in the laser drawing apparatus according to the present invention.

FIG. 14 is a graph conceptually showing a result of the straightness error correction processing performed in units of one pixel size or less executed in FIG.

FIG. 15 is a part of a flowchart for describing a straightness correction processing routine of the laser drawing apparatus according to the present invention.

FIG. 16 is the remaining part of the flowchart for explaining the straightness correction processing routine of the laser writing apparatus according to the present invention.

[Explanation of symbols]

 Reference Signs List 10 base 12 rail 14 X table 18 drawing table 24 argon laser generator 52/58 electronic shutter 70 polygon mirror 78 system control circuit 80 CCD camera 82 image processing circuit 84 straightness correction processing circuit 86 main scanning control circuit 88 sub-scanning control Circuit 90 Beam position control circuit 92 Raster conversion circuit 94 Y scale sensor 96 Signal processing circuit 98 Servo motor drive circuit 100 Servo motor 102 X scale sensor 104 Signal processing circuit

Claims (4)

[Claims] An object to be drawn is moved in a sub-scanning direction while deflecting the laser beam in the main scanning direction with respect to the object to be drawn, and modulation of the laser beam with respect to the object to be drawn is determined based on raster data. What is claimed is: 1. A laser writing apparatus for writing a predetermined pattern by controlling according to a clock pulse having a frequency, comprising: starting writing along a main scanning direction by the laser beam based on measured straightness error data of a pixel array along the sub-scanning direction. A drawing start position shift data generating means for generating position shift data, wherein the drawing start position shift data includes a drawing start position along the main scanning direction by the laser beam in a unit of one pixel size of the raster data (D a first series of drawing start position shift data shifted by _p) (X _s), the start of drawing along the main scanning direction by the laser beam A pixel size following units of the raster data location (D _p / n) is shifted in the second series of drawing start position shift data (X _t)
Measuring means for measuring the moving distance of the object to be drawn in the sub-scanning direction; and the moving distance measured by the measuring means is used as the first series of drawing start position shift data (X _s ). A first drawing start position shift means for shifting a drawing start position in the main scanning direction by the laser beam in a predetermined direction in units of one pixel size (D _p ) each time the distance becomes equal to Each time the moving distance measured by the above becomes equal to the second series of drawing start position shift data (X _t ), the drawing start position along the main scanning direction by the laser beam is defined as a unit of one pixel size or less ( D _p /
a laser writing apparatus comprising: a second writing start position shift means for shifting in a predetermined direction in n). 2. The laser writing apparatus according to claim 1, wherein the drawing start position of the laser beam along the main scanning direction is shifted toward the deflection direction of the laser beam, and is along the sub-scanning direction. The measured straightness error data of the pixel array is positive, and at this time, the shift of the drawing start position by the first and second drawing start position shift means is performed on the negative side, so that the laser beam along the main scanning direction is shifted. When the drawing start position is shifted to the side opposite to the direction of deflection of the laser beam, the straightness error measurement data of the pixel array along the sub-scanning direction is negative, and at this time, the first and second straightness errors are measured.
Wherein the writing start position is shifted to the positive side by the writing start position shifting means. 3. The laser drawing apparatus according to claim 1, wherein when the first drawing start position shift means reads the raster data from the memory, the read timing is advanced or delayed by one pixel. A laser drawing apparatus characterized by comprising: 4. The laser writing apparatus according to claim 1, wherein said second writing start position shift means sets said clock pulse at a predetermined interval (2π /
clock pulse generating means for generating a predetermined number (n) of clock pulses whose phase is shifted by n), and selectively one of the predetermined number (n) of clock pulses obtained by the clock pulse generating means. And a clock pulse output means for shifting the drawing start position along the main scanning direction of the laser beam by a unit (D _p / n) smaller than the one pixel size in a predetermined direction. Clock pulse output control means for controlling the output of the clock pulse so as to sequentially shift the phase of the output clock pulse by an amount (2π / n) corresponding to the predetermined number (n) is included. Laser drawing equipment.