JPH05199374A - Light beam scanning method - Google Patents

Abstract

PURPOSE:To prevent occurrence of position deviation at each picture drawing point by correcting the nonlinearity of a deflection angle of a light deflector. CONSTITUTION:Prescribed sweep signals VT, VM are applied to an AOD 270 to scan a laser beam LB 5. The sweep signals VT, VM are synchronized with a system clock in a picture drawing controller 100. On the other hand, digital modulation control signals VDa, VDb applied to digital AOMs 240, 250 are decided by measuring in advance a position deviation at each picture drawing point caused when a signal synchronous with a sweep signal VT is fed to the digital AOMs 240, 250 and correcting an output timing of a sweep signal VT based on the result of the measurement. As a result, the incident time to the AOD 270 of the laser beam LB 4 is controlled by the digital modulation control signals VDa, VDb.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a light beam scanning method using a light deflecting element, and more particularly to a light beam scanning device for drawing an image by scanning a light beam on a photosensitive material. ..

[0002]

2. Description of the Related Art Optical deflection element (hereinafter referred to as AOD).
Is widely used as a key device for scanning a laser beam. As is well known, this AOD is one of the elements that utilize the acousto-optic effect, and the diffraction direction of the light diffracted by the ultrasonic wave excited in the element changes depending on the frequency of the ultrasonic wave. Based on this, the diffraction direction of light is controlled by frequency modulation. AO like this
FIG. 29 shows a conceptual diagram of a conventional example of a scanning optical system using D.

As shown in the figure, the scanning optical system 200B is
AOD270B and analog AOM (light modulator) 220
B and a digital AOM 240B. It should be noted that, in this figure, for convenience of description, the description of other optical systems such as lenses is omitted.

Here, the analog AOM 220B is an element that generally attempts to control the intensity of light by analog modulation based on the fact that the intensity of diffracted light depends on the intensity of ultrasonic waves. AOM220B is
It is used as an element for setting the light amount of the laser beam LB03 to an optimum value according to the sensitivity of the dry plate 1B. That is, the analog AOM 220B plays a role similar to that of a fixed camera diaphragm. Therefore,
The control voltage applied to the analog AOM 220B, once set, is not changed during the scanning of the laser beam LB03. That is, the control voltage is constant.
As a result, the laser beam LB0 incident on the analog AOM 220B is always attenuated to a predetermined light amount and is emitted as a laser beam LB01.

The digital AOM 240B is an element for controlling the presence or absence of light by digital modulation based on the fact that the intensity of diffracted light depends on the intensity of ultrasonic waves. Therefore, the digital AOM 240B has the control signal V
The ON / OFF operation is repeated according to DO. That is, the digital AOM 240B plays the same role as the shutter of the camera. This allows the digital AOM240
The laser beam LB01 that is incident upon the ON operation of B is emitted as a laser beam LB02 and is incident on the AOD 270B.

On the other hand, the operation of the AOD 270B is controlled by the two control signals VTO and VMO. One of the control signals, VTO, is a signal for controlling the deflection angle of the laser beam LB03, and the control signal VTO causes the dry plate 1B to move.
The position of each drawing point above is controlled. This control signal VTO
For example, a signal such as a sawtooth wave whose frequency changes continuously and periodically is used. The other control signal VMO is a signal for controlling the light quantity of the laser beam LB03, and the light quantity at each drawing point is controlled by this control signal VMO.

However, these control signals VTO and VMO
Therefore, in order to control the deflection angle of the laser beam LB03 with high accuracy, there is a problem that the non-linear characteristic of the AOD 270B must be actually taken into consideration.

That is, the relationship between the input frequency and the deflection angle of the diffracted light in the AOD is generally non-linear. For this reason,
Even if the control signal VTO is changed linearly, the positions of the drawing points do not change linearly accordingly, and the actual positions of the respective drawing points deviate from the positions (ideal positions) to be originally drawn. Hereinafter, this phenomenon will be referred to as position shift.

Therefore, it is necessary to determine the control signal VTO capable of correcting the non-linear characteristic of the AOD in order to perform the drawing with good image quality without causing the positional deviation.

Therefore, as a method of obtaining such an optimum control signal VTO, the following technique has been conventionally used.

That is, in the prior art, first, an appropriate value is set as the value of the control signal VTO for each drawing point. next,
These control signals VTO and VMO are actually transferred to the AOD270.
While applying the laser beam to B to scan the laser beam LB03,
A CCD camera (not shown) installed at a position equivalent to the position of the dry plate 1B measures the actual beam position of the laser beam LB03.

Further, from these measurement results, correction data for correcting the actual beam position of the laser beam LB03 to the ideal position is created for each drawing point. Then, the correction data is used as a new control signal VMO in the AOD 27.
0B to the laser beam LB0 again.
3 is scanned, and the actual beam position of the laser beam LB03 is similarly measured by the CCD camera.

Then, at each drawing point, the optimum control signal VTO is determined by repeating the above series of operations until the deviation between the beam position of the laser beam LB03 and the ideal position is within the allowable range. The control signal VM
O also changes in synchronization with the control signal VTO.

As described above, in the prior art, the correction data obtained by repeating the series of processes is used as the control signal VTO, and the AOD 270B is controlled by the corrected control signal VTO together with the control signal VMO. It was possible to prevent the displacement.

[0015]

However, there have been cases where the conventional technique cannot sufficiently correct the nonlinear characteristic of the AOD.

That is, when the time during which the control signal VTO of the AOD changes is longer than the time during which the ultrasonic wave propagates in the AOD, the non-linear characteristic of the AOD can be effectively corrected by the conventional technique. It was possible to prevent the occurrence of positional deviation.

However, conversely, when performing high-speed scanning such that the change time of the AOD control signal VTO is shorter than the propagation time of ultrasonic waves, even if the conventional technique is used, a positional deviation still occurs, which is an effect. There has been a new problem that it is impossible to prevent the occurrence of positional deviation. That is, at the time of high-speed scanning, several μrad. Deviation of the deflection angle, therefore, several μm
The drawing point (scanning point) was misaligned. The reason why such a problem occurs during high-speed scanning can be qualitatively understood as follows.

First, the laser beam itself incident on the AOD has a width of about several millimeters. On the other hand, the propagation velocity of ultrasonic waves is several hundred meters / second,
The propagation time required to propagate from one end of the OD to the other end is several tens of microseconds. Therefore, when the control signal VMO changes faster than the propagation time of ultrasonic waves, the AOD
The inner laser beam will be interacted with not only by the corresponding ultrasonic waves, but also by ultrasonic waves of other frequencies existing around the laser beam. It can be said that the frequency range of ultrasonic waves that can exert this interaction becomes wider as the change time of the control signal VTO becomes shorter. It has been confirmed that in the current system, it is necessary to consider the influence of ultrasonic waves having about 150 different frequencies / intensities.

Here, FIG. 28 is an explanatory diagram schematically drawn to make the understanding of the above description clearer.
The figure shows exactly the control signal VTO of the frequency fi and the amplitude Vi.
The control signal VMO of is applied to the AOD 270E to
The case where ultrasonic waves of frequency fi are excited in 270E is shown. At this time, the laser beam LB03 diffracted by the AOD 270E irradiates the drawing point Pi. or,
This drawing shows a state where the laser beam is sequentially scanned from the drawing point Pi-2 to the drawing point Pi-1 and the drawing point Pi. Moreover, this figure shows a case where the control signals VTO and VMO change before the ultrasonic waves of each frequency corresponding to the drawing points Pi-1 to Pi spread to the entire inside of the AOD 270E.

As shown in the figure, the control signal V of the frequency fi
When the control signal VMO of TO and the amplitude Vi is applied to the AOD 270E, in addition to the ultrasonic wave of the frequency fi, the frequency fi− corresponding to the previous drawing points Pi-2 and Pi−1, respectively.
2, fi-1 ultrasonic waves will also be present in the AOD 270E. Therefore, the above three ultrasonic waves exist within the range of the beam diameter of the laser beam LB02 incident on the AOD 270E, and the laser beam LB02 is affected by each of these ultrasonic waves. Therefore, the deflection angle of the laser beam LB03 is determined by the acousto-optic effect of each of the three ultrasonic waves and the laser beam LB02.

On the contrary, when the conventional technique is effective, that is, when the change time of the control signal VTO is slower than the propagation time of the ultrasonic wave, the laser beam LB02 is AOD27.
Since only one kind of ultrasonic wave (frequency fi) exists when it enters 0E, the laser beam LB03
The deflection angle is determined by the interaction between the laser beam LB02 and the ultrasonic wave of frequency fi, and the control signal VTO can be accurately and easily corrected from the measured beam position.

As described above, during high speed scanning, the AOD
In order to properly correct the deflection angle of, the influence from many ultrasonic waves must be taken into consideration. However, it is considered extremely difficult to actually perform such correction of the deflection angle (correction of the control signal VTO) for all the drawing points as follows.

That is, when the correction data of the control signal VTO for one of all the drawing points is changed from the measurement result, the correction data of the control signal VTO for the other drawing points is also affected by the change, and it is necessary to change the correction data. Occurs. In the example of FIG. 28, when the correction data for the drawing point Pi is corrected, it is necessary to subsequently correct the correction data for the drawing point Pi-1 and further the correction data for the drawing point Pi-2. When the frequency of the ultrasonic wave excited at each drawing point is changed in this way, the interaction between light and each ultrasonic wave also changes,
This again leads to a vicious circle in which the correction data must be corrected sequentially from the drawing point Pi.

Moreover, even the linearity of the frequency characteristic of the voltage controlled oscillator used as the generation source of the control signal VTO is lost by repeating the correction of the control signal VT, which causes a vicious circle.

In order to change the deflection angle of one of all the drawing points, in the memory for storing the correction data of the control signal VTO, the data stored at any one of the addresses is corrected. There is also the drawback that it is not clear what to do.

Therefore, it can be said that the conventional method cannot obtain accurate correction data of the control signal VTO at all.

Further, the position where the laser beam is incident on the AOD and the beam diameter at the time of incidence are subtly different depending on the adjustment state of the optical system. This means that the correction data of the control signal VTO also depends on the adjustment state of the optical system.
Further, even when the beam shape of the laser beam is slightly different for each drawing point, the correction data of the control signal VTO will be affected thereby. Therefore, in the conventional technique, the control signal VTO must be corrected in consideration of these influences. It can be said that such a correction is not realistic at all.

As described above, the method of directly correcting the AOD control signal as in the prior art cannot prevent the positional deviation during high speed scanning.

The present invention has been made to solve the above problems, and an object of the present invention is to prevent the occurrence of positional deviation regardless of whether the scanning is slow scanning or high scanning. It is an object of the present invention to provide a light beam scanning method capable of realizing high quality and high precision writing.

[0030]

According to the present invention, a light beam is scanned by entering the light beam into the light deflection element via the light modulation element. (A) The light deflection element and the light modulation element A predetermined deflection control signal and a first digital modulation control signal synchronized with the predetermined deflection control signal are applied, and the beam position of the light beam emitted from the optical deflection element on the photosensitive material is set within each scanning range. While measuring the scanning point, (b) the positional deviation determined by the difference between the measured beam position and the ideal position corresponding to the beam position,
Calculate for all scan points. Further, (c) the output timing of the first digital modulation control signal is corrected based on the calculation result, and the corrected first digital modulation control signal is determined as the second digital modulation control signal. Then, (d) a second digital modulation control signal and a predetermined deflection control signal are applied to the light modulation element and the light deflection element, respectively, so that the light beam is scanned on the photosensitive material.

[0031]

The optical modulator according to the present invention digitally modulates the presence or absence of a light beam in accordance with the applied second digital modulation control signal. Moreover, the output timing of the second digital modulation control signal is corrected with respect to the output timing of the first digital modulation control signal synchronized with the predetermined deflection control signal. Therefore, the light modulation element exerts a function of adjusting the incident timing of the light beam incident on the light deflection element with respect to the output timing of a predetermined deflection control signal.

On the other hand, the light deflection element emits the light beam emitted from the light modulation element according to the output timing of the second digital modulation control signal at a deflection angle according to a predetermined deflection control signal.

[0033]

EXAMPLES A. Overall Structure of Drawing System and Its Schematic Operation (A-1) Mechanical Structure FIG. 2 is a perspective view showing the mechanical structure of a drawing system 10 according to an embodiment of the present invention. It should be noted that, for the sake of convenience, a drawing control device, a data processing unit, and the like, which will be described later, are omitted from the drawing.

As shown in FIG. 1, the drawing system 10 includes a photosensitive material feeding mechanism 20 and a drawing mechanism 30 on a base 15.

The sensitive material feeding mechanism 20 has a suction table 21 and a pair of guides 22 extending in the horizontal Y direction. The suction table 21 is slidably mounted on the guides 22. Further, the photosensitive material 1 such as a glass dry plate is adsorbed on the suction table 21. or,
The suction table 21 is reciprocated in the (± Y) direction by a ball screw (not shown) rotated by a motor 23. As a result, the photosensitive material 1 also reciprocates in the (± Y) direction.

On the other hand, the drawing mechanism 30 has a pair of guides 31 extending in the horizontal X direction. However, the X direction is Y
The direction perpendicular to the direction. A housing 32 is slidably mounted on the guide 31, and the scanning optical system 200 is housed in the housing 32.
The drawing head 33 shown in the notch in the figure is one component of the scanning optical system 200. Further motor 3
When the ball screw 35 is rotated by 4, the housing 32, and hence the scanning optical system 200, moves in the X direction or (-
Move in the X direction. As a result, the drawing head 33 also moves in the X direction or the (−X) direction.

A laser-oscillator 40A (He-Ne laser, etc.) is provided on the upper surface of the base 15. The laser beam 41 from this laser oscillator 40A is
The beam splitters 42 to 45 split the laser beams into two laser beams 41X and 41Y. However, the beam splitters 44 and 45 are fixed to the drawing head 33. Furthermore, the X-direction end of the suction table 21 and (-
The plane mirrors 46X and 46Y are provided at the end portions in the Y direction.
Is erected. As a result, the laser beam 41X,
41Y is reflected by these mirrors 46X and 46Y, respectively, and returns to the positions of the beam splitters 44 and 45. Then, the mirror reflection optical path length of each of the laser beams 41X and 41Y is detected by an optical interference detector (not shown). As a result, the relative position of the photosensitive material 1 in the horizontal plane with respect to the drawing head 33 is measured. Hereinafter, these optical systems including the laser-oscillator 40A and an optical interference detector (not shown) will be collectively referred to as a laser-length measuring device. Although not shown, the entire sensitive material feeding mechanism 20 is housed in a light-shielding hood that can be opened and closed.

Further, in the vicinity of the sensitive material feeding mechanism 20, CC
The D camera 50 is installed as shown in FIG. This C
The role of the CD camera 50 will be described later.

(A-2) Electrical Configuration FIG. 3 is a schematic diagram showing the electrical configuration of the drawing system 10. As shown in the figure, the electrical configuration mainly includes the drawing control device 100. Here, the laser-length measuring device 40 outputs the above-mentioned measurement result regarding the relative position of the photosensitive material 1 to the drawing control device 100 as position information Sx (X direction) and Sy (Y direction). Then, the drawing control device 100 creates a scanning signal based on the position information signals Sx and Sy, and also controls the control signal V based on the scanning signal.
Create T, VM, VA, VD. These control signals V
T, VM, VA, and VD are signals that control the operation of each component of the scanning optical system 200.

Further, the measurement data VP of the CCD camera 50 is transmitted to the CPU 310 in the workstation 300 to be subjected to predetermined processing, and the data Di stored in the memory (not shown) is also controlled through the CPU 310. It is transmitted to the device 100.

On the other hand, the motor controller 180 in the drawing control apparatus 100 outputs control signals VC1 and VC2 to the motors 23 and 34, respectively, and the motors 23 and 3 are output.
The rotation of 4 is controlled. Further, the image data SV is given from the figure input device 400 to the drawing control device 100.

A detailed description of a series of operations in the drawing control device 100 and the data processing unit 300 described above will be given later.

(A-3) Basic Principle of Drawing FIG. 4 is an explanatory diagram showing the basic principle of drawing in the drawing system 10. From the drawing head 33, two laser beams LB5a periodically deflected in the (± X) direction,
The LB 5b is irradiated onto the photosensitive material 1. These laser beams LB5a and LB3b are both modulated based on the image signal SV. Then, while moving the photosensitive material 1 in the (-Y) direction, for example, the laser beams LB5a, LB5b.
Exposure. In this case, drawing is performed along the array of the scanning lines L extending in the (± X) direction in the Y direction. Further, the drawing area 2 of the photosensitive material 1 is a parallel stripe 2a,
Are conceptually divided into stripes 2b, ... And drawing is performed for each stripe 2a, 2b ,.

B. Configuration of Scanning Optical System (B-1) Outline of Configuration of Scanning Optical System FIG. 1 is a diagram schematically showing main components of the scanning optical system 200. As shown in this figure, the laser oscillator 21
The laser beam LB oscillated from 0 is the analog A first.
It is incident on the OM 220.

The analog AOM 220 simply attenuates the light amount of the laser beam LB to a light amount suitable for the sensitivity of the photosensitive material 1, and a constant control voltage VA is always applied to the analog AOM 220 during scanning.

Next, the laser beam LB1 is introduced into the beam splitter 230 and the two laser beams LB2.
a and LB2b. The laser beams LB2a and LB2b are respectively digital AOMs.
It is incident on 240 and 250. Where digital AOM
As described above, 240 and 250 are elements for controlling the presence or absence of diffracted light by digital modulation. Therefore, the digital AOMs 240 and 250 are responsive to the digital modulation control signals VDa and VDb issued from the drawing control apparatus 100, respectively, to generate the laser beams LB2a and LB2a.
Turns LB2b on and off. Moreover, both the digital modulation control signals VDa and VDb are signals created based on the image signal SV given from the image input device 400 to the drawing control device 100, and the positional deviation of the drawing point (corresponding to the scanning point). In order to prevent the above, the output timing is a signal appropriately corrected with respect to the timing of the sweep signals VT and VM described later. A detailed description of these points will be given later.

As a result, the laser beam LB2a is diffracted in a certain direction by the ultrasonic wave excited in the digital AOM 240 only when the digital modulation control signal VDa is at the H level, and is emitted from the digital AOM 240 as the laser beam LB3a. Similarly, the laser beam LB2b also has a digital modulation control signal VDb.
Is emitted from the digital AOM 240 as the laser beam LB3b only when is at the H level.

After that, both the beams LB3a and LB3b are introduced into the beam splitter 260, and the laser beam LB
4 is synthesized. That is, the combined laser beam LB4
Is a beam bundle composed of two laser beams LB4a and LB4b which travel at a predetermined distance. Therefore, after the beam splitter 260, the scanning optical system 200 is constructed again by one optical system. A detailed description of such an optical system will be given later.

Next, the laser beam LB4 is emitted from the AOD 27.
Input to 0. This AOD 270 is the drawing control device 1
AO according to the sweep signals VT and VM issued from 00
The deflection angle of the laser beam LB5 diffracted in D270 is controlled. The laser beam LB5 is also a beam bundle of two laser beams LB5a and LB5b,
Both beams LB5a and LB5b are used as a generic term. The operation of the AOD 270 will be described in more detail below.

First, the sweep signal VT is, for example, a sawtooth wave.
It is a signal whose frequency changes continuously and periodically and which is a predetermined signal. As a result, AOD27
0 changes the deflection angle of the laser beam LB5 according to the change in the frequency of the sweep signal VT. As a result, the laser beam LB5 scans the photosensitive material 1 in the ± X directions. However, the deflection angle changes non-linearly with respect to the frequency change of the sweep signal VT. That is, the sweep signal VT itself is
It does not correct the non-linear characteristic itself of the deflection angle of 0.

The other sweep signal VM is a signal having a constant voltage value. Therefore, the intensity of each ultrasonic wave excited in the AOD 270 is always a constant value determined by the voltage value of the sweep signal VM.

Actually, the sweep signals VT and VM are
Instead of being continuously applied to the AOD 270, it is applied to the AOD 270 at predetermined time intervals. Therefore, the laser beam LB5 scans the photosensitive material 1 little by little.

The main constituent elements of the scanning optical system 200 and their schematic operations have been described above. As clearly shown in the above description, the feature of the scanning optical system 200 is that the digital AOMs 240, 2 are not intended to directly correct the nonlinear characteristic of the AOD 270 itself as in the conventional technique.
By correcting the output timings of the control signals VDa and VDb applied to the 50 with respect to the timings of the sweep signals VT and VM, it is intended to correct the positional deviation due to the non-linear characteristic of the AOD 270 itself. A method of optimizing such digital modulation control signals VDa and VDb is also a feature of the present invention, as described later. The basic points of interest that have led to such a configuration are as follows.

That is, of the sweep signal VT of the AOD 270,
When the voltage value for one drawing point is corrected, it is necessary to also correct the value of the sweep signal VT for other drawing points, and it is difficult to correct the value of the sweep signal VT for all the drawing points. This is because even if an ultrasonic wave having a frequency f is excited to deflect the laser beam LB4 in a predetermined direction, the laser beam LB itself is affected by ultrasonic waves having frequencies of about 150 existing in the surroundings. It was as described that it was caused by. From such a viewpoint, the laser beam LB4
When the time of incidence on the AOD 270 is delayed or advanced from the time of excitation of the ultrasonic wave of frequency f (when the sweep signal VT changes), the laser beam LB4 itself changes from ultrasonic waves other than the ultrasonic wave of frequency f. The effect is that it will be completely different from the effect when the incident time and the excitation time are the same. This means that, conversely, the deflection angle of the laser beam LB5 can be controlled by the time of incidence of the laser beam LB4 on the AOD 270. That is, even if the sweep signal VT of the AOD 270 is not corrected, the positional deviation can be corrected by simply controlling the incident time of the laser beam LB4. Such consideration will be further clarified by consideration of FIG. 26 described below.

FIG. 26 schematically shows the value of the frequency f of the sweep signal VT (vertical axis on the right) and the value of the deflection angle θ of the laser beam LB5 (vertical axis on the left) with respect to the scanning time t (horizontal axis). It is the explanatory view shown. In the same figure, the straight line (a) represents the relationship when it is assumed that the frequency f of the sweep signal VT continuously changes. On the other hand, in the straight line (b), the sweep signal VT that continuously changes like the straight line (a) has the AOD2.
When applied to 70, the relationship is shown when the deflection angle θ of the laser beam LB5 is assumed to be proportional to the sweep signal VT, and thus to the scanning time t. The step function (c) represents the actual change of the frequency f of the sweep signal VT.

Therefore, let us consider the deflection angle θ at the time t20 when the frequency f of the sweep signal VT changes from the frequency f2 to the frequency f3. That is, if the deflection characteristic of the AOD 270 has a linear characteristic originally, the time t20
In, the deflection angle θ should be the deflection angle θ20.

However, in actuality, the deflection angle θ is set to the deflection angle θ2 (θ2 <θ2 <due to the non-linearity of the deflection characteristic of the AOD 270.
θ20). As a result, the position of the drawing point deviates from the original position to the scanning start origin side. On the contrary, if viewed from the straight line (b) (from the standpoint of linearity), it is considered that the laser beam LB5 is incident on the AOD 270 at the time t2p and is deflected at the time t2p. Become. That is, at the time t20, the laser beam LB4 apparently appears at the time t20.
It is considered that the state will be equivalent to that which is earlier than time by Δτ. Therefore, at the time t20, on the contrary, if the incident time of the laser beam LB4 can be delayed by the time Δτ (time t2), the deflection angle θ can be set to the deflection angle θ20, and the positional deviation of the drawing point is caused. It will be considered to prevent.

On the contrary, when the position of the drawing point is farther than it should be, it is possible to prevent the position deviation of the drawing point by advancing the incident time of the laser beam LB4. Conceivable.

Therefore, in the present invention, the laser beam LB
As a means for controlling the incident time of 4, the digital AOM
240 and 250 are used, and the output timing of the digital modulation control signals VDa and VDb is delayed or advanced with respect to the output timing of the sweep signal VT.

The present invention also provides a digital AOM 240,
It is based on the fact that the pulse response time at 250 is hundreds of times faster than AOD. For example, the pulse response time in digital AOM is 9 to 12n.
sec. Is. Such a difference is caused by the fact that transverse ultrasonic waves are used in AOD, whereas
This is because ultrasonic waves used in OM are longitudinal waves. This digital AOM240, 2
The feature of 50 is that the digital modulation control signals VDa and VDb
This brings an advantage that the relationship between the timing and the positional deviation of the drawing point can be easily determined.

(B-2) Specific Configuration of Scanning Optical System FIGS. 5 to 7 are optical configuration diagrams clearly showing the specific configuration of the scanning optical system 200. First, a laser beam LB emitted from a laser oscillator 210 composed of an Ar ion laser (wavelength: 488 nm) or the like is incident on the analog AOM 220 via the mirrors M1 to M3 and attenuated to a light amount suitable for the sensitivity of the photosensitive material 1. .. And the emitted laser beam L
B1 is reflected by the mirror M4 and then introduced into the beam splitter 230, where the laser beams LB2a and LB2b.
And are demultiplexed.

Next, the laser beam LB2a is reflected by the mirror M.
5. Digital AOM 24 through the condenser lens L1a
It is incident on 0 and is digitally modulated. The laser beam LB emitted when the digital AOM 240 is turned on
3a is incident on the non-polarization beam splitter 260A via the condenser lens L2a and the mirror M6. On the other hand, the laser beam LB2b is incident on the digital AOM 250 via the condenser lens L1b. Then, the laser beam LB emitted when the digital AOM 250 is turned on.
3b is incident on the non-polarization beam splitter 260A via the condenser lens L2b. Both laser beams LB3a and LB3 incident on the non-polarization beam splitter 260A
b is the non-polarization beam splitter 260A and the polarization beam splitter 2 which is then incident via the mirror M7.
Is synthesized by 60B.

After that, the laser beams LB4a, LB4
The beam interval b is narrowed by the first expander EP1 and then is incident on the AOD 270 (see FIG. 7).
The operation of the AOD 270 is as described above. still,
The inter-lens distances of the rod lens L3 and the cylindrical lens L4 constituting the first expander EP1 are appropriately adjusted to cancel the so-called “AOD cylindrical effect”. This “cylindrical effect of AOD” means that when a parallel beam is incident on the AOD,
When the frequency of the sweep signal of D is constant, the diffracted light also becomes a parallel beam, whereas when the frequency of the sweep signal of AOD changes, the diffracted light does not become a parallel beam and has a certain spread. This is a phenomenon. Therefore, in order to prevent the occurrence of the positional shift of the focus position due to such a phenomenon, the laser beams LB4a and LB4b which are slightly narrowed by changing the distance between the lenses are AO.
Laser beam LB5 incident on D270 and diffracted
a and LB5b are emitted in parallel.

On the other hand, the laser beams LB5a and LB5b emitted from the AOD 270 are the second expander EP.
The beam interval is expanded by 2 and the beam is incident on the scan lens L7. That is, the laser beams LB5a, L
B5b is a rod lens L6 of the second expander EP2
The laser beams LB5a and LB5b emitted from the scan lens L7 have a spread of an angle θ with respect to
The beams are parallel to each other with a beam interval h of about 44 μm (FIG. 7).

Finally, the laser beams LB5a and LB5b emitted from the scan lens L7 are guided to the drawing head 33 described above via the mirror M8. That is, the laser beams LB5a and LB5b are transmitted by the Pechan prism P.
It is incident on the objective lens L9 through Z and the relay lens L8. At this time, the laser beam LB5a, LB5b has a beam interval of about 22 μm.
The beam diameter (diameter) of the LB 5b is 20 μm. This beam interval satisfies the beam interval of 20 μm required for the two beams not to interfere with each other. Thereafter, the beams LB5a and LB5b are irradiated onto the photosensitive material 1 after the beam diameters thereof are reduced in accordance with the predetermined magnification set in the objective lens L9. In this example, three types of magnifications (2 times, 5 times, 1
0 times) is prepared. Here, since the beam diameters of the laser beams LB5a and LB5b upon entering the objective lens L9 are 20 μm, the beam diameters of the respective beams LB5a and LB5b are reduced to 10 μm when the magnification is doubled. Similarly, when the magnification is 5 times and 10 times, the beam diameters are 4 μm and 2 μm, respectively. still,
In the following description, it is assumed that the scanning optical system 200 has a magnification of 10 times (beam diameter 2 μm). In addition, the beam diameter means the diameter of the beam waist of the laser beam.

In the drawing head 33, as shown in FIG. 6, the laser beam L emitted from the objective lens L9.
An autofocus detection system (laser diode LD, B5a, LB5b) for always focusing on the photosensitive material 1
A position detector PSD) is provided.

C. Electrical Configuration of Drawing Control Device (C-1) Overall Configuration of Drawing Control Device FIG. 8 is a block diagram showing the overall configuration of the drawing control device 100 together with peripheral devices. In the figure, digital AOM drivers 241 and 25 are shown as respective drivers of the digital AOMs 240 and 250 and the AOD 270.
1 and the AOD driver 271 are shown in FIG.
In, the description of these drivers was omitted for convenience.

As shown in the figure, the drawing control device 100 is
Clock 110, scanning signal generator 120, digital A
It has OM control units 140 and 150, a sweep signal generation unit 160, a raster conversion unit 170, and a motor controller 180.

First, the system clock SCLK (20 MHz) generated from the clock 110 is used as the scanning signal generator 12
Input to 0.

The scanning signal generator 120 produces the data read signal DR, the data start signal DST and the scan start signal ST based on the system clock SCLK, outputs the data read signal DR to the sweep signal generator 160, and simultaneously starts the data start. The signal DST and the scan start signal ST are output to the digital AOM control units 140 and 150.
The system clock SCLK is also input to the digital AOM control units 140 and 150.

Further, the clock 110 uses the ECL clock CLK (200 MHz) as the digital AOD control unit 14.
Output to 0 and 150.

On the other hand, the CPU 310 controls the CCD camera 50.
The correction data for determining the optimum digital modulation control signals VDa and VDb is created based on the beam position data VPO measured by the digital AOM control units 140 and 1 via the data bus 350.
Send to 50. Further, the CPU 310 also transmits data regarding the sweep signals VT and VM to the sweep signal generator 160 via the data bus 350.

The general operation of each unit 140, 150, 160 is as follows.

First, the digital AOM controller 140
The standby state is set in response to the scan start signal ST. After that, the digital AOM control unit 140 outputs the raster signal SVRa (analog signal) sent from the raster conversion unit 170 to the dot signal DOT1 at a timing determined based on the correction data created by the CPU 310.
(Digital signal). This timing creation will be described later. After that, the dot signal DOT1
Is converted into a digital modulation control signal VDa suitable as a drive signal by the digital AOM driver 241. The digital AOM control unit 150 also performs the same operation. These raster signals SVRa, SVR
b is a signal created by raster conversion of the image signal SV by the raster conversion unit 170.

Further, the sweep signal generator 160 outputs the data concerning the stored sweep signals VT and VM to the data read signal DR.
And the sweep signals VT and VM are created at the same timing, and the sweep signals VT and VM are also read by the AOD driver 27.
Output to 1. The AOD driver 271 uses the sweep signal V
The T and VM are converted into a sweep signal VS suitable as a drive signal.

The general operation of each part is as described above. The detailed configuration of each unit will be described below.

(C-2) Clock FIG. 9 is a block diagram showing the configuration of the clock 110. As shown in the figure, the clock 110 is composed of an ECL oscillator 111 and a frequency divider 112. ECL here
The oscillator 111 uses an ECL clock CLK of 200 MHz.
Is an oscillator that oscillates. And ECL clock CLK
Is input to the frequency divider 112 and is divided into the system clock SCLK of 20 MHz.

(C-3) Scanning Signal Generating Unit FIG. 10 is a block diagram showing the configuration of the scanning signal generating unit 120. The scanning signal generator 120 uses the position information signal Sy (position pulse) sent from the laser length measuring device 40 to scan start signal ST and data read signals DR and D.
It is a unit for creating a DR.

First, the position information signal Sy is subjected to a predetermined correction by the length measurement pulse correction circuit 121, and then the ALU (Arit
hmatic Logic Unit) 122. This ALU1
22 is a conversion rule RUL held in the register 124.
Based on the above, the corrected position information signal Sys expressed in the pulse unit system (time) is converted into a signal expressed in the unit system of length (μm), and then the scanning start pulse SP is output to the scanning signal controller 123. To do.

The position information signal Sy is also sent to the Y-axis position counter 125. This Y-axis position counter 125
Detects the current position of the drawing head 33 in the Y direction from the position information signal Sy, and when detecting that the drawing head 33 has reached a predetermined Y direction drawing start position, one stripe scanning is performed. Y direction start pulse YSP
To the scanning signal controller 123.

Next, the scanning signal controller 123 makes the Y
Only when the direction start pulse YSP is asserted can the scan start pulse SP be output. Then, the scanning signal controller 123 starts to count the scanning signal counter 126 at this point and counts up in synchronization with the timing of the system clock SCLK. Here, the count number of the scanning signal counter 126 corresponds to the address of a memory (described later) included in the sweep signal generation unit 160.

The scanning signal counter 126 has 50
nsec. The sweep signal generator 160 outputs the data read signal DR for reading the data stored in the memory every time in synchronization with the timing of the system clock SCLK.
Output to. Further, the data read signal DR is transmitted to the decoder 127.
It is fed back to the scanning signal controller 123 via.

Further, the decoder 127 outputs the data read signal D
When the count number indicated by R and the count number S previously given to the decoder 127 become equal, the data start signal DST is sent to the digital AOM control units 140 and 1.
Output to 50. Here, the count number S is a value associated with the first drawing point P1.

The scanning signal controller 123 outputs the scanning start pulse SP as the scanning start signal ST to the digital AOM control units 140 and 150.

(C-4) Digital AOM Control Unit FIG. 11 is a block diagram schematically showing the electrical configuration of the digital AOM control unit 140. Digital A
The configuration of the OM control unit 150 is also the digital AOM control unit 14.
Since it is the same as 0, its explanation is omitted.

Here, the digital AOM control unit 140
Data read clock SDR having output timing in which the timing of system clock SCLK is appropriately corrected
Is a unit that converts the rasterized image signal SVRa into a dot signal DOT1 (digital signal). Therefore, in order to output the image signal SVRa serially, the image signal SVRa is first stored in the FIFO memory.

As shown in the figure, the digital AOM control section 140 is composed of a data read clock generation section 141, a data read control section 149 and a FIFO memory 144. The data read control unit 149 includes a data read controller 142, a dot number counter 143, and a dot converter 146.

Here, the data read clock generator 141
Is synchronized with the 200 MHz ECL clock CLK,
It is a circuit that creates a data read clock SDR that is a signal in which the timing of the system clock SCLK is corrected, and is an important component that can be said to be the heart of the digital AOM control unit 140. The correction amount of the timing of the system clock SCLK is determined based on the positional deviation correction data VPOD given to the data read clock generating unit 141 from the data processing unit 300 in advance. Therefore,
First, the detailed configuration and operation of the data read clock generator 141 will be described.

FIG. 12 is a block diagram schematically showing the electrical configuration of the data read clock generator 141. In the figure, the positional deviation correction data memory 1411
The position shift correction data VPOD is stored in advance. The storage process of the position shift correction data VPOD is performed as follows.

That is, the data VPO regarding the 1024 beam positions measured by the CCD camera 50 is transmitted to the CPU 310 via the input interface 330. Then, the CPU 310, based on the data VPO relating to these beam positions, corrects the positional deviation data VPO.
D is calculated, and these data VPOD are stored in the position shift correction data memory 1411 via the output interface 340.
Output to. By the above processing, the positional deviation correction data V
The POD is stored at the corresponding address of the positional deviation correction data memory 1411.

Here, the positional deviation correction data memory 141
One itself has 1400 addresses, and correspondingly, the correction data VPOD is also composed of 1400 data. However, of the 1400 data VPODs,
The meaningful data VPOD is 1024 pieces of data calculated from the measurement result VPO of the CCD camera 50,
These are sequentially stored from the S-th address (equivalent to the S value of the decoder 127) of the positional deviation correction data memory 1411. The details of the method of measuring the beam position by the CCD camera 50 and the method of calculating the positional deviation correction data VPOD will be described later.

Next, the position correction controller 1416 synchronizes with the scanning start signal ST sent from the scanning signal controller 123, and the positional deviation correction data memory 1411.
The data stored in the first address of is read and the data is transmitted to the ECL latch 1412.
In addition, in order to generalize the following story, the data is represented by the symbol VP.
Described as OD (J) (data stored at the Jth address).

The ECL latch 1412 latches the transmitted data VPOD (J) (TTL logic) according to the 200 MHz ECL clock CLK (latch signal), and also the ECL logic data VPOE.
Convert to (J). Then, the ECL latch 1412 is
The data VPOE (J) is set in the ECL counter 1413.

The position correction controller 1416 also outputs an enable signal EN to the counter controller 1417 in response to the scan start signal ST. Then, the counter controller 1417 synchronizes the count enable signal CEN with the ECL counter 1413 and the duty control counter 141 in synchronization with the enable signal EN.
Output to 8. As a result, both counters 1413, 14
18 is in a counting operation start state.

Here, the ECL counter 1413 changes the value of the positional deviation correction data VPOD (J) to be set to 5 nsec. (200 when converted to frequency
MHz) Data read clock SD for an integral multiple of time
The purpose is to shift the rising timing of R. Therefore, the ECL counter 1413 first counts the positional deviation correction data VPOE (J) and determines the carry signal CSJ corresponding to the count number in order to determine the appropriate rising timing of the data read clock SDR for each drawing point Pi. Output to the J terminal of the JK-FF (flip-flop) 1414. The above-mentioned purpose or 5 nsec. The significance of the time unit will be made clearer in the description of the scanning method described later.

Similarly, carry signal CSJ is also set in duty control counter 1418. The duty control counter 1418 is intended to adjust the duty of the data read clock SDR. Therefore, the duty control counter 1418 indicates that the carry signal CSJ is JK.
The carry signal CSDJ rising from the L level to the H level after being set to the J input terminal of the FF1414 (corresponding to the duty) after being set according to the carry signal CSDJ is input to the K input terminal of the JK-FF1414. Output to.

The operation of the JK-FF 1414 is as follows.

First, when the carry signal CSJ is set to the J input terminal, the J input terminal is at the H level,
The K input terminal is at the L level. Therefore, J-K-FF1
The output of 414 is the ECL clock CLK (latch signal)
It is set at the timing of. On the other hand, carry signal CS
When DJ goes from L level to H level, J-K
Since the level of the J terminal of the -FF1414 is at the L level, the output of the JK-FF1414 is the ECL clock C
It is reset at the timing of LK. That is, JKF
Data read clock SDR by setting F1414
Rise timing is determined, and JK-FF141
By resetting 4, the duty of the data read clock SDR is adjusted.

The output signal of the JK-FF 1414 is
In order to improve the synchronization with the ECL clock CLK, the data is first input to the D-FF 1415 and then output as the data read clock SDRJ. Where the symbol SDRJ
Means that it is the data read clock SDR corresponding to the Jth address.

Further, the data read clock SDR is input to the position shift control counter 1419. The position shift control counter 1419 controls the count enable by the latch signal CP from the position correction controller 1416 synchronized with the system clock SCLK. Actually, the count operation is performed by the rising of the data read clock SDRJ, and the counter output ADD is output to the misalignment correction data memory 1411. The counter output ADD is a signal for instructing the next address of the positional deviation correction data memory 1411, that is, the (J + 1) th address.

That is, the positional deviation correction data memory 1411
The positional deviation correction data VPOD (J + 1) stored in the (J + 1) th address of is read according to the timing of the counter output ADD, and is set in the ECL counter 1413 via the ECL latch 1412. After that, the positional deviation correction data VPOE (J + 1)
Similarly, a series of processes is performed for the data read clock SDR (J + 1). Therefore, the rising timing of the data read clock SDR is 5 ns depending on the value of the positional deviation correction data VPOE.
ec. The signal is shifted by an integral multiple of (200 MHz), and the high-level holding time is made constant by the duty control counter 1418. That is, the rising timing and the cycle are controlled by the value of VPOE (J), and the shift amount of the data read clock SDR (J) by the value of VPOE (J) is sequentially accumulated.

The structure and operation of the data read clock generator 141 are as described above. With such a circuit structure, the positional deviation can be corrected and the scan width can be adjusted at the same time. It can be done.

Next, based on FIG. 11, the digital A
The configuration and operation of the OM control unit 140 will be described.

First, the data read clock SDR is input to the data read controller 142. The data read controller 142 uses the data start signal DST for designating the address S corresponding to the first drawing point P1.
The data read start state is synchronized with the data read clock SDR, and at the same time, the image read clock IR synchronized with the data read clock SDR is output to the FIFO memory 144 and the dot converter 146. In addition, the data read controller 14
Similarly, 2 outputs a count-up signal DCNT synchronized with the data read clock SDR to the dot number counter 143 to count the number of dots (the number of drawing points).

Therefore, the raster image signal Ra is sequentially read from the FIFO memory 144 at the timing of the image read clock IR, and the dot converter 146 is read.
Are sequentially transmitted to.

Further, in the dot converter 146, the image signal Ra is sequentially formed into dots at the timing of the image reading clock IR, and the dot converter 146 outputs the dot signal DOT1 which is an ON / OFF signal.

That is, in the dot converter 146, after the image signal Ra from the FIFO memory 144 is read at the timing of the image reading clock IR, the dot converter 146 also reads the image signal Ra at the rising timing of the image reading clock IR. The exclusive OR of (J) and Ra (J + 1) is used to obtain the dot signal DOT1.
Is formed.

(C-5) Sweep Signal Generation Unit FIG. 13 is a block diagram schematically showing the electrical configuration of the sweep signal generation unit 160 together with the data processing unit 300.
As shown in the figure, the sweep signal generator 160 uses the sweep signal V
It is roughly divided into a part that creates T and a part that creates the sweep signal VM.

First, in the linearity correction memory 161, 1400 pieces of linearity correction data VTD relating to the sweep signal VT are previously stored in the CPU via the output interface 340.
It is given by 310. These linearity correction data VTD are signals for correcting the non-linearity of the frequency characteristic with respect to the control voltage in the voltage controlled oscillator (VCO) 166. Then, the linearity correction data VTD corresponds to the timing of the data read signal DR, that is, 50n.
sec. The data is sequentially read from the linearity correction memory 161 for each time, converted into an analog signal by the D / A converter 163a, and then input to the PGA 164a. Where PGA
164a is also the PGA of the analog AOM control unit 130.
Like 133, it has a gain register 167a and an offset register 168a for correcting its gain and its offset according to the standard of AOD270.

After that, the linearity correction data VTD amplified by the PGA 164a is the low-pass filter 165a.
Is input to the VCO 166 via. The frequency characteristic of the VCO 166 is made linear by the linearity correction data VTD, and the sweep signal VT is generated from the VCO 166.

As described above, in the sweep signal generator 160, (70 μse) is reached within one scanning time of the AOD 270.
c. ) 1400 pieces of linearity correction data VTD are sequentially D / A converted at the timing of the system clock SCLK to form the sweep signal VT.

On the other hand, also for the diffraction efficiency correction memory 162, 1400 pieces of diffraction efficiency correction data VMD relating to the sweep signal VM are given in advance by the CPU 310 via the output interface 340.

The diffraction efficiency correction data VMD is also 50n depending on the timing of the data read signal DR.
sec. The data is sequentially read from the diffraction efficiency correction memory 162 for each time, converted into an analog signal by the D / A converter 163b, and then input to the PGA 164b. PG here
The A 164b also has a gain register 167b and an offset register 168b for the same reason. Then, the diffraction efficiency correction data VMD amplified by the PGA 164b is output to the AOD driver 271 together with the sweep signal VT as the sweep signal VM through the low pass filter 165b.

D. Scanning Method FIG. 14 is a flowchart showing a scanning procedure in the drawing system 10. Hereinafter, the scanning procedure for each step will be described with reference to the configuration drawings as appropriate.

(D-1) Step S1 This step, including the following step S2, corresponds to the preparation step for starting scanning.

First, various data required for scanning are set. That is, the linearity correction data VTD and the diffraction efficiency correction data VMD are given from the CPU 310 to the linearity correction memory 161 and the diffraction efficiency correction memory 162, respectively. Further, the image signal SV is input to the raster conversion unit 170 from the image input device 400. Further, the raster-converted image signals SVRa and SVRb are respectively transferred to the FIFO.
It is stored in the memories 142 and 152. Note that these processes (not shown) are controlled by the CPU 310.

(D-2) Step S2 In this step, the positional deviation correction data VPOD is created. This creation is performed according to the procedure shown in FIGS. The details of each procedure will be described below.

Steps S21 to S27 i) First, a 0 value is set to each address J (J: 1 to 1400) of the positional deviation correction data memory 1411 (step S21). That is, the positional deviation correction data VPOD
Is VPOD (J) = 0 (J: 1 to 1400).
This setting itself is, as described above, the CPU 310.
Done by

Ii) Next, the laser beam LB5 is scanned based on the positional deviation correction data VPOD (J) set in step S21, and the actual position X of each drawing point Pi is scanned.
Measure (i). In this case, the positional deviation correction data VP
Since all the OD values are 0, the data read clock generator 141 does not correct the rising timing of the data read clock SDR. Therefore, the digital modulation control signal VD synchronized with the system clock SCLK
a, VDb, digital AOM140,
150 is driven.

Here, the relationship between the drawing point Pi and the address J of the positional deviation correction data memory 1411 is J = i + S-
1 (i: 1 to 1024). Further, the position (beam position) X (i) of the drawing point Pi means the distance in the X direction from the origin of the scan start described above.

FIG. 21 is a diagram showing the concept of such measurement.
Is. In the figure, while fixing the scanning optical system 200, the laser beam LB5a is emitted from the AOD 270 one by one in the direction of the 1024 drawing points Pi, and the CCD camera 50 is moved one by one. By doing so, the actual beam position X at each drawing point Pi
A method for measuring (i) is described.

At this time, the digital AOM 240 has a digital modulation control signal V synchronized with the sweep signals VT and VM.
DS (corresponding to the first digital modulation control signal) is applied. The digital modulation control signal VDS corresponds to the dot signal DOT0 before correction, which will be described later.
Further, sweep signals VT and VM (corresponding to deflection control signals) synchronized with the system clock SCLK are applied to the AOD 270.

However, since the CCD camera 50 is actually fixed, in the present embodiment, as shown below, the scanning optical system 200 is moved one point at a time to reverse each beam position X. (I) is to be measured.
The light receiving surface of the CCD camera 50 is fixed at a position equivalent to that of the photosensitive material 1.

First, the scanning optical system 200 is set to the CCD camera 5
It moves to the light receiving surface of 0 and sets the position of the first drawing point P1 in a measurable state (step S22). In order to generalize the following description, the first drawing point P1 is represented as the drawing point Pi (step S23).

When the above preparation is completed, the light receiving surface of the CCD camera 50 is actually irradiated with the laser beam LB5 (step S24), and the actual position (beam position) X (i) of the drawing point Pi is determined by the CCD. The measurement is performed by the camera 50 (step S25). The signal VPO relating to the beam position X (i) measured by the CCD camera 50 is sent to the CPU 310 via the input interface 330 and stored in the memory 320.

Next, it is judged whether or not the (i + 1) th drawing point Pi is the drawing end point, that is, the 1024th drawing point P1024 (steps S26 and S27).

Here, the drawing point P (i + 1) is the drawing end point P.
When it is determined that it is not 1024, the process proceeds to step S23, and the series of steps S23 to S27 is continued until the drawing point Pi becomes the drawing end point P1024. As a result, the actual beam position X (i) is measured for each drawing point Pi.

Steps S28 to S218 Steps S28 to S218 are all processes performed in the CPU 310.

I) First, in steps S27 and S28, the positional deviation Z (i) regarding the drawing point Pi is calculated. That is, the CPI 310 is based on the ideal beam position XO (i) of the drawing point Pi previously stored in the memory 320 and the actual beam position X (i) of the drawing point Pi also stored in the memory 320. , Z (i) = XO (i) −X (i) (i: 1 to 1024) The positional deviation Z (i) is calculated.

Here, FIG. 19 is a diagram schematically showing the relationship between the ideal beam position XO (i) of each drawing point Pi and the actual beam position X (i). The figure shows a case where the actual beam position X (i) of each drawing point Pi is deviated from the ideal beam position X0 (i) in the (-X) direction. However, at the drawing point P1, the actual beam position X (1) and the ideal beam position XO (1) are equal.
Further, the ideal beam spacing d shown in FIG. 4A is generally a value set to satisfy the relationship of d = n · R (n: natural number) with respect to the beam diameter R. is there.

On the other hand, in the present drawing system 10, the ideal beam interval d is set to be equal to the beam spot diameter R (2 μm) as shown in FIG. 20A (n = 1). .. Therefore, the laser beam LB5a or LB
The ideal arrangement of 5b in the X direction is as shown in FIG.

The frequency of the sweep signal VT is 20 MHz,
Therefore, the cycle is 50 nsec. Therefore, the ideal beam interval d (2 μm) is converted into time units,
50 nsec. Is equivalent to.

Ii) Next, in step S210, the positional deviation accumulated value W (i) for the drawing point Pi is calculated. The positional deviation accumulated value W (i) is a value defined by W (i) = W (i-1) + Z (i-1) -Z (i). However, at the drawing point P1, the positional shift Z (1) has a value of 0 and is defined as W (1) = 0.

For example, the respective positional deviation accumulated values W (2), W (3), W (4) for the drawing points P2, P3, P4 are as follows.

W (2) = W (1) + Z (1) -Z (2) =-Z (2) W (3) = W (2) + Z (2) -Z (3) =-Z (2 ) + Z (2) -Z (3) =-Z (3) W (4) = W (3) + Z (3) -Z (4) =-Z (4) Therefore, the positional deviation accumulated value W (i) Is W (i) =-Z
It will be expressed as (i).

For reference, FIG. 22 shows an example of the positional deviation accumulated value W (i) for each drawing point Pi calculated from the measurement result of the positional deviation Z (i) before correction. This case corresponds to the case where the actual beam positions X (i) of all the drawing points Pi are always displaced from the ideal beam positions XO (i) in the shrinking direction. Depending on the frequency range of the sweep signal VT, the actual beam position X (i) may be the ideal beam position X (i).
It may occur in the direction of extension from O (i).

Iii) Next, the positional deviation accumulated value W (i)
The position shift correction data VPOD (i) is corrected based on the above (steps S211 to S218). This correction is
As will be described below, this is performed by comparing the positional deviation accumulated value W (i) with a value of 1/2 of the minimum positional deviation correction unit M. At this point in time, each positional deviation correction data VPOD
(I) is a state where all are not corrected, that is, VPOD
(I) = 0. Further, an address J [J: 1 to (S-1), (S + 1024) to 140 that does not correspond to the drawing point Pi is used.
0] the positional deviation correction data VPOD stored in
No correction is made for (J).

Here, the minimum unit of positional deviation correction M is generally a value corresponding to 1/10 of the ideal beam interval d. Therefore, in this embodiment, the minimum unit of positional deviation correction M is 0.2 μm. This value is 5 nsec. Is. That is, the minimum unit of positional deviation correction M corresponds to the cycle of the 200 MHz ECL clock CLK. Then, 1/2 of the minimum unit of positional deviation correction M,
That is, 0.1 μm is treated as the limit value of the allowable positional deviation.

First, in step S211, the magnitude of the absolute value | W (i) | of the positional deviation accumulated value W (i) and 1/2 of the minimum positional deviation correction unit M is determined.

If it is determined that the absolute value | W (i) | is less than 1/2 of the minimum unit of positional deviation correction M, the process proceeds to steps S217 and S218. That is, in this case, since the positional deviation of the drawing point Pi is within the allowable range, it is not necessary to correct the positional deviation correction data VPOD (i). Therefore, the positional deviation correction data VPOD
(I) is set to 0 value.

On the other hand, when the absolute value | W (i) | is equal to or larger than 1/2 of the minimum unit M of positional deviation correction, the positional deviation exceeds the allowable range, and therefore the positional deviation correction data VPOD.
The process moves to step S212 for correcting (i).

In step S212, the positional deviation accumulated value W
It is determined whether or not (i) is greater than or equal to the value (-1/2) of the minimum unit of positional deviation correction M. Here, when the determination is “YES”, the processing in the CPU 320 is step S
Move to 213.

In step S213, +1 is added to the positional deviation correction data VPOD (i). Adding +1 corresponds to moving all the actual beam positions X (i) to X (1024) from the drawing point Pi to the drawing point P1024 by 0.2 μm in the + X direction. Details of this point will be described later.

The reason for performing such addition is as follows. That is, when W (i) ≦ (−M / 2), the actual beam position X (i) of the drawing point Pi is in a state of being contracted from the ideal beam position XO (i), This is because it is necessary to move the beam position X (i) at the point Pi toward the ideal beam position XO (i).

Further, after adding +1, the positional deviation accumulated value W
The minimum unit of positional deviation correction M (0.2 μm) is added to (i) (step S214). In this calculation, as the beam position X (i) of the drawing point Pi moves by 0.2 μm in the + X direction, the positional deviation accumulated value W (i) is set within the allowable range (−).
This is a process for setting a value within M / 2) to (+ M / 2). Then, the addition result is the new positional deviation cumulative value W.
It is stored in the memory 320 as (i).

On the other hand, "NO" in the step S212.
When it is determined that, that is, when W (i) ≧ (+ M / 2), the processing in the CPU 320 moves to step S215.

In step S215, a process of adding -1 to the positional deviation correction data VPOD (i) is performed. That is, when W (i) ≧ (+ M / 2), the actual beam position X (i) of the drawing point Pi is, on the contrary, advanced to the + X direction from the ideal beam position XO (i). Therefore, the beam position X (i) of the drawing point Pi is made to approach the ideal beam position XO (i) by adding -1. By this addition of -1, the actual beam positions X (i) to X (1024) from the drawing point Pi to the drawing point P1024 all move in the -X direction by 0.2 μm.

After further adding -1, the positional deviation accumulated value W
The minimum unit of positional deviation correction M (0.2 μm) is subtracted from (i) (step S216). This subtraction is the drawing point Pi
As the beam position X (i) of No. moves by 0.2 μm in the −X direction, the positional deviation accumulated value W (i) is set to a value within the allowable range (−M / 2) to (+ M / 2). This is processing for. Then, the addition result is stored in the memory 320 as a new positional deviation accumulated value W (i).

After that, when the correction processing of the positional deviation correction data VPOD (i) for the drawing point Pi is completed, the positional deviation correction data VPOD for the next drawing point Pi + 1 is obtained.
The correction process of (i + 1) is performed (step S217,
S218). This series of correction processing is performed at the drawing end point P10.
The correction process for 24 is performed until it is completed.

The method of creating the positional deviation correction data VPOD (i) has been described above, but in order to further clarify the understanding of such a creating method, it will be specifically described below.

I) Here, FIG. 23 is a diagram showing an example of the positional deviation accumulated value W (i) before and after the correction. In the figure, the black dots on the polygonal line (a) indicate the position deviation accumulated value W (i) before correction, and the white dots on the polygonal line (b) indicate the position after correction based on the result of the polygonal line (a). Deviation cumulative value W
It represents (i). Therefore, these results will be described for each drawing point Pi.

First, regarding the drawing points P (1) and P (2), the absolute values of the positional deviation accumulated values W (1) and W (2) are both within the allowable range (less than M / 2). , No correction is made. Therefore, the correction data VPOD (1), VPOD
A zero value is set in (2).

At the next drawing point P (3), the drawing point P (2)
Since the position deviation of the drawing point P (3) itself is accumulated as it is in the position deviation generated in the above, the position deviation accumulated value W (3) is equal to the lower limit value -M / 2 of the allowable range. Will be exceeded. Therefore, the positional deviation correction data VPOD
+1 is added to (3), and the minimum unit of positional deviation correction M is added to the positional deviation cumulative value W (3) before correction. As a result, the corrected positional deviation accumulated value W (3) is 0.
It becomes the value within the range from to + M / 2.

On the other hand, at the next drawing point P (4), the drawing point P
Since the position deviation of the drawing point P (4) itself is accumulated with respect to the position deviation allowed in the position deviation correction in (3), the position deviation cumulative value W (4) is the lower limit of the allowable range. It does not exceed the value -M / 2. Therefore, the drawing point P (4)
A zero value is set in the positional deviation correction data VPOD (4). Hereinafter, the drawing points P (5) to P (7) are similarly processed, and the positional deviation correction data VPOD (5) to VPOD (7) of the drawing points P (5) to P (7) are respectively set. +1, 0, 0 values are set.

Ii) Next, FIG. 24 shows a timing chart of each signal when each positional deviation correction data VPOD (i) is created based on the example of FIG. In the figure, (a) shows the system clock SCLK and (b) shows
Is the position deviation correction data before correction VPOD (i) (VPO
Dot signal DOT0 created based on D (i) = 0)
(Corresponding to the first digital modulation control signal) (for reference), (c) is the corrected positional deviation correction data VPOD
The dot signal DOT1 (corresponding to the second digital modulation control signal) created based on (i), (d) represents the time axis, and (e) represents the drawing point. Below, Figure 1
1, the drawing point P (1) will be sequentially described with reference to FIG.

First, at time t1, the drawing point P (1)
Of the positional deviation correction data VPOD (1) is read from the positional deviation correction data memory 1411, and the ECL counter 1
413 is set. However, since this positional deviation correction data VPOD (1) has a 0 value, JK-FF14
The rising timing of the data read clock SDR output from 14 is the same as the rising timing of the system clock SCLK. That is, for the drawing point P (1), the dot signal DOT1 is the system clock SC.
It is synchronized with LK. Further, since the positional deviation correction data VPOD (2) of the drawing point P (2) also has a value of 0, the dot signal DOT1 is also synchronized with the system clock SCLK at the drawing point P (2) (time t2). ..

Next, regarding the drawing point P (3), the positional deviation correction data VPOD (3) is +1. Therefore, E
The CL counter 1413 displays the positional deviation correction data VPOD.
By counting (3), the count number is incremented from 0 to 1, and the J terminal of the JK-FF 1414 is set to the H level at the time t3 delayed by the time Δt (5 nsec.) From the time t30. To As a result, the data read clock SDR and hence the dot signal D
The rising timing of OT1 is time t3. The delay time Δt of the rising timing becomes the delay time of the incident time of the laser beam LB5 on the AOD 270 as it is, and the beam position X of the drawing point P (1) is obtained.
The positional deviation of (3) is within the allowable range.

Next, regarding the drawing point P (4), the positional deviation correction data VPOD (4) has a zero value. Therefore, E
The count number of the CL counter 1413 is still 1. As a result, the rising timing of the dot signal DOT1 also becomes the time t4 delayed by the time Δt from the time t40, and the positional deviation of the beam position X (4) of the drawing point P (4) is also within the allowable range.

On the other hand, regarding the drawing point P (5), since the positional deviation correction data VPOD (5) is +1, ECL
The count number of the counter 1413 is 2. as a result,
The rising timing of the dot signal DOT1 is time t
The time t5 is delayed from 50 by a time 2Δt.

Hereinafter, for the drawing points P (6) and P (7), the positional deviation correction data VPOD (6) and VPOD
Since both (7) are 0 values, the delay time of the rising timing of the dot signal DOT1 is time 2Δt.

Iii) Finally, FIG. 25 illustrates the beam spot when the laser beam LB5a is scanned using the dot signal DOT1 shown in FIG. 24 together with the case where the dot signal DOT0 is used. That is, FIG. 25 (a)
Is the result when the dot signal DOT0 is used (corresponding to the conventional example). However, a circle described by a broken line shows a beam spot when the beam is in an ideal beam position. or,
FIG. 25B shows the result when the dot signal DOT1 is used. 25 (c) means the X coordinate axis.

As is clear from the figure, the dot signal DOT1
When using, the beam positions X (1) to X (5) of the drawing points P (1) to P (5) are all within the allowable range. That is, the positional deviation between the drawing points P (1) to P (5) is less than 0.1 μm. This positional deviation is an amount that can be sufficiently ignored, and it can be said that the laser beam LB5a is always accurately and accurately irradiated onto the ideal beam position X0 (i) during scanning.

(D-3) Step S3 The optimum digital modulation control signal VD is obtained in the previous step.
Since a and VDb (corresponding to the second digital modulation control signal) are determined, scanning is started in this step.

FIG. 27 shows the movement path positions of the photosensitive material 1 and the drawing head 33 in the (± X) direction when drawing is performed on the photosensitive material 1 using the drawing system 10.

First, as shown in (a) of the figure, the drawing head 33 starts the scanning start position (scanning origin) near the lower left corner of the photosensitive material 1.
, The photosensitive material 1 is moved in the Y direction. Then, the scanning is started.

(D-3) Steps S4 to S5 The scanning is carried out by scanning the laser beams LB5a and LB5b in the X direction while sending the photosensitive material 1 in the (± Y) directions. The drawing system 10 is set so that the drawing points Pi are formed in a line without deviating from the scanning line L even when the photosensitive material 1 is scanned in the X direction while being fed in the + Y direction or the -Y direction. Has been done. Since such a technique is disclosed in the document of Japanese Patent Application No. 1-140099 filed by the present applicant, its explanation is omitted here.

First of all, the photosensitive material 1 (-Y
Direction). FIG. 27 shows such a state.
It is (b). Therefore, the drawing for the first stripe proceeds in the Y direction, and when the feeding in the -Y direction on the photosensitive material 1 is completed, the state shown in FIG. 27C is obtained.

Next, the drawing head 33 moves in the X direction by a predetermined distance ΔX, as shown in FIG. 27 (d). This distance ΔX is set to a distance equal to the mutual arrangement interval between stripes.

After that, the photosensitive material 1 is sent in the Y direction in reverse, whereby the drawing of the second stripe is completed (FIG. 27 (e)).

Thereafter, the same reciprocal scanning is repeated for the other stripes (FIG. 27 (f)), and finally the desired image is recorded in the drawing area, and the scanning is completed (step S5). ).

E. Modified Example (E-1) In this embodiment, the analog AOM 220 is used to adjust the light amount of the laser beam LB to a light amount value suitable for the sensitivity of the photosensitive material 1, but the present invention is not limited to this. .. For example, an optical attenuator, an optical filter, or the like can be used instead of the analog AOM 220.

(E-2) In this embodiment, when the laser beam LB5 is scanned at high speed (scanning time 70 μse
c. ). However, it is obvious that the present invention is not limited to the high speed scanning, and can be applied to the case of the low speed scanning which is the range of application of the prior art. Also in this case, it is possible to prevent the occurrence of positional deviation at each drawing point.

(E-3) In this embodiment, the case where two laser beams LB5a and LB5b are scanned has been described, but the present invention is not limited to this. That is, 1
Drawing may be performed only by scanning the laser beam LB5a of the book. In this case, a beam splitter or the like becomes unnecessary. Also, a plurality of laser beams of three or more may be scanned. In this case, it is necessary to prepare as many digital AOMs as the number of laser beams.

[0174]

According to the present invention, the output timing of the first digital modulation control signal is corrected based on the beam position measured for each scanning point, and the corrected signal is output as the second signal.
Since the signal is applied to the light modulation element as the digital modulation control signal, the light beam scanned on the photosensitive material by the light deflection element is always applied to the ideal scanning position. That is, the present invention has an effect that it is possible to prevent the occurrence of positional deviation at all scanning points.

[Brief description of drawings]

FIG. 1 is a block diagram showing main components of a scanning optical system.

FIG. 2 is a perspective view showing a mechanical configuration of a drawing system which is an embodiment of the present invention.

FIG. 3 is a configuration diagram schematically showing an electrical configuration of a drawing system.

FIG. 4 is an explanatory diagram showing the basic principle of drawing in the drawing system.

FIG. 5 is a configuration diagram specifically showing an optical configuration of a scanning optical system.

FIG. 6 is a configuration diagram specifically showing an optical configuration of a scanning optical system.

FIG. 7 is a configuration diagram specifically showing an optical configuration of a scanning optical system.

FIG. 8 is a block diagram showing an overall configuration of a drawing control device.

FIG. 9 is a configuration diagram of a clock.

FIG. 10 is an electrical configuration diagram of a scan signal generator.

FIG. 11 is an electrical configuration diagram of a digital AOM control unit.

FIG. 12 is an electrical configuration diagram of a data read clock generator.

FIG. 13 is an electrical configuration diagram of a sweep signal generator.

FIG. 14 is a flowchart showing a scanning procedure in the drawing system.

FIG. 15 is a flowchart showing a method for determining a digital modulation control signal.

FIG. 16 is a flowchart showing a method for determining a digital modulation control signal.

FIG. 17 is a flowchart showing a method for determining a digital modulation control signal.

FIG. 18 is a flowchart showing a method of determining a digital modulation control signal.

FIG. 19 is an explanatory diagram showing a relationship between an ideal beam position and an actual beam position.

FIG. 20 is an explanatory diagram showing a relationship between a beam interval and a beam spot diameter.

FIG. 21 is an explanatory diagram showing a beam position measuring method using a CCD camera.

FIG. 22 is an explanatory diagram showing an example of the calculation result of the positional deviation accumulated value.

FIG. 23 is an explanatory diagram specifically showing a method of calculating a positional deviation accumulated value.

FIG. 24 is a timing chart showing a relationship between a dot signal before correction and a dot signal after correction.

FIG. 25 is an explanatory diagram showing a relationship between a beam spot before correction and a beam spot after correction.

FIG. 26 is an explanatory view schematically showing in order to clarify the technical point of view of the present invention.

FIG. 27 is an explanatory diagram showing the relative movement of the photosensitive material and the drawing head.

FIG. 28 is an explanatory diagram showing a relationship between ultrasonic waves and an incident laser beam in AOD.

FIG. 29 is a block diagram showing a configuration of a conventional scanning optical system.

[Explanation of symbols]

 1 Sensitive Material 100 Drawing Control Device 200 Scanning Optical System 210 Laser Oscillator 220 Analog AOM 270 AOD LB5 Laser Beam VD Digital Modulation Control Signal VDa Digital Modulation Control Signal VDb Digital Modulation Control Signal VT Sweep Signal VM Sweep Signal CLK ECL Clock SCLK System Clock 1411 Position shift correction data memory 50 CCD camera 161 Linearity correction memory 162 Linearity correction memory VPOD Position shift correction data

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Norio Morita, No. 465, Kuze Tsukiyama-cho, Minami-ku, Kyoto 1 Dainichi Moto Screen Manufacturing Co., Ltd., Kuze Plant (72) Inventor Sadahide Adachi, Kuze Tsukiyama-cho, Minami-ku, Kyoto No. 465, 1 Dainichi Honshu Manufacturing Co., Ltd., Kuze Factory

Claims (1)

[Claims] 1. A light beam scanning method for scanning a light beam onto a light-sensitive material by injecting the light beam into a light deflecting element via a light modulating element, comprising: (a) the light deflecting element and the light modulating element. A predetermined deflection control signal and a first digital modulation control signal synchronized with the predetermined deflection control signal are applied to the respective elements, and the beam position of the light beam emitted from the light deflection element on the photosensitive material is adjusted. A step of measuring each scanning point within the scanning range; and (b) a positional deviation determined by a difference between the beam position measured in the step (a) and an ideal position corresponding to the beam position. And (c) correcting the output timing of the first digital modulation control signal based on the calculation result, and correcting the corrected first digital modulation control signal to the second value. Determining a digital modulation control signal, and (d) applying the second digital modulation control signal and the predetermined deflection control signal to the light modulation element and the light deflection element, respectively, to sense the light beam. And a step of scanning on the material.