JPS62189867A - Pulse generating device - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

Detailed Description of the Invention [Technical Field of the Invention] The present invention relates to a pulse generator, which is driven based on the scanning position of a line sensor, for example.
This invention relates to a pulse generator suitable for an imaging device that reads a target image.7 [Backbone of the invention] Reads a target image using a line sensor in which a large number of image pickup devices such as CODs are arranged in a one-dimensional manner. An example of an imaging device that performs this is the one shown in Figure 8g2.

In this figure, a test sample 12 to be imaged is placed on an X-Y stage 10 that is movable in the X and Y directions of the figure. On the sides of this X-Y stage 10, an X-axis interferometric needle receiver (hereinafter simply referred to as "X-interference ν-bar") 14 and a Y-axis interferometer receiver (hereinafter simply referred to as "Y-bar") 14 are installed.
Interference receivers (referred to as interference receivers) 16 are respectively disposed so that the amount of movement of the XY stage 10 can be measured by the interference effect of laser light. That is, when the X-Y stage 10 moves in the X direction, pulses are output from the VcX interference receiver 14 at predetermined pitches, and the pulses are output in the Y direction 1o1.
vcX When the X stage 10 moves, the Y interference receiver 16 outputs a pulse force at every predetermined pitch. In addition, the interferometer receiver 14°16 is actually VC
is placed on the test sample 12, which includes a photoelectric detector that outputs a two-phase signal with a 90° phase shift in the form of a pseudo sine wave, and a pulse frequency converter that outputs a full pulse for every unit phase shift change from the photoelectric detector. is an optical system 1 such as an appropriate objective lens.
A line sensor 20 is disposed via the line sensor 8, and an image on the test sample 12 is read by moving the line sensor 20 in the X-Y direction. Note that the optical system 18 has the function of enlarging the image on the inspection sample 12 and projecting it onto the line sensor 20VC.

In order to perform two-dimensional imaging using the one-dimensional line sensor 20 in the above-described imaging device, the X-Y strip 10 is scanned in a predetermined direction, for example, the Y direction, and the line sensor 20 is imaged. It is necessary to supply a timing pulse for reading for each line.

One method for this is to supply timing pulses from the line sensor 20 at regular intervals,
It is conceivable to scan so that the speed of 0 becomes constant accordingly.

However, in this method, even if the movement of the XX stage 10 is minute, it must be performed at a constant speed, which is impossible to implement.

Another method is to move the XX stage 10 at a constant speed, measure up and down pulses from an interferometer, and drive the line sensor 20 based on this. The operation in this case will be explained below. Note that for ease of understanding, the explanation will be given assuming that the magnification of the optical system 18 is "1". When the X-X stage 10 moves in the Y direction, as shown in FIG.
It may be considered that the line sensor 20 moves relatively on the line sensor 2 in the Y direction. This movement in the Y direction is performed continuously, and a pulse is output from the Y interference receiver 16 every time the movement is a predetermined amount, as shown in FIG. 4(A)K. Let the moving pitch of this pulse output be 'kLPA.O On the other hand, if the moving pitch of the line sensor 20 corresponding to the width of the line sensor 20 in the Y direction is LPB (see Fig. 3), then LPB=N-LPA...・・・・・・・・・・・・
- When (1) is reached, the line sensor 20 has relatively moved by that width, as shown in FIG. 5 (5). In other words, when N pulses are output from the Y interference receiver 16, the line sensor 20 moves by one pixel. Therefore, at this point, as shown in FIG. 4 (E9), the sensor drive signal is not input to the line sensor 20, and the image of the test sample 12 is read.

By the way, in order to operate well as described above, the pitch LPB of the line sensor 20, that is, its width, must be an integral multiple of the moving pitch LPA of the XX stage 10, which corresponds to one pulse of the output of the Y interference receiver 16. It is necessary that

However, in reality, such conditions are not met. - To give an example, the pitch LPB of the line sensor 20 is, for example, 18 μm. However, if the magnification of the optical system 18 (see Figure 2) is 40 times, then the test sample 1
2, 18/40 = 0.45 μm.

On the other hand, the moving pitch LPA of the x-yi stage 10 is used in the Y interference receiver 16, and its measurement accuracy is determined by the wavelength of the h laser beam.
It is 0.0197756175 μm per one pulse of the output of No. 6.

Determining N in equation (1) from the above values, N= LPB
/LPA=0.45/0.0197756175=
22.7552907・・・・・・・・・・・・・・・
...(2) and is not necessarily an integer. For this reason, when feeding the XX stage 10 times over the long dimension, errors below the decimal point accumulate, and as shown in Figure 5 (day or ()
Attach the sensor to the line sensor 20 at the position shown by the broken line in Q.
The F drive signal will be input. In other words, X-
Image enlargement or reduction often occurs in the feeding direction of the X stage 10.

[Purpose of the invention]

The present invention has been made in view of the above, and it is an object of the present invention to provide a pulse generator that outputs a pulse signal for driving an imaging means within the minimum error range of detecting movement of an imaging target. [Summary of the Invention] According to the non-explosion #IK, the reference data for which #reference data can be stored in the data storage means (64) is a pulse burst in the detection unit amount of physical change amount. This is the data that is missing the change meter to be generated, and is the data with the required number of digits including the decimal part.

Reference data is 1ru *means t36.38.42)
K! Then, the necessary number of digits including the decimal number N1 are added in order.

On the other hand, the detection means (14, 16) detects all physical changes 1 and outputs a pulse for each unit amount of change. This pulse is counted by the counting means (44), and this counted value and the integer part of the addition I@ value of the adding means are compared by the comparing means (44).
0). When the two match, the adding operation of the adding means becomes weak and a pulse is generated every predetermined amount of change that is larger than the unit amount.

[Embodiment] Hereinafter, embodiments of the present invention will be explained with reference to the accompanying drawings. Fig. 1 shows an embodiment of the pulse generator according to the present invention. , the host computer 30 is connected to a microcomputer 62, and this microcomputer 62 is connected to a first 32-bit register 54. This first register 34Fi, 3
is connected to a second 32-bit register 68 via a 2-bit adder 36;
The output side of is also connected to the adder 36 and the register 34°38
data is added.

The upper 8 bits of the next KS register 38 are input to a 24-bit humparator 40.

Counters 42 and 44 are connected to this comparator 40, and the upper 8 bits and 1
The output of the 6-bit counter 42 and the output of the 24-bit counter 44 are compared by a comparator 40.

Further, a reset signal is inputted from the outside to the register 68 and counter 42 and 44 input signals CL.In addition, the match signal which is the comparison output of the comparator 40 is used to drive the line sensor 20. It is used as a signal and is also input to the clock terminal of the register 38 and counter 42.

Among the components of each part mentioned above, the register 64 has (1
), that is, the number of output pulses from the Y interference receiver 16 required to drive the line sensor 20t--(hereinafter referred to as the "number of drive timing pulses") is stored. The drive timing pulse number N is calculated in advance based on the device configuration as shown in equation (2), and is stored in the register 34 from the host computer 60 via the microcomputer 32. The number of drive timing pulses has, for example, 62 effective bits, of which 24 bits are below the decimal point.

Next, the counter 42 connected to the adder 66 counts up when there is a carry in the addition by the adder 56, and the counter 44 receives the Y interference receiver 16 color pulses. The pulse count increases due to human power.

Next, the overall operation of the above embodiment will be explained.

First, as described above, the number N of drive timing pulses expressed as a decimal point (for example, the number N 22 as shown in equation (2))
．． 7552907) is stored in the register 34.

Then, the X-Yx-f-ji 10 is sent in the Y direction at a substantially constant speed, and a reset signal is input at the starting position of the inspection effective area of the inspection sample 12, and the register 68
, the contents of counters 42 and 44 are cleared and both are “
0'' □ At this point, the input of the comparator 40 will eventually become ``0'' and match, and a match signal will be output.

On the other hand, the adder 56 is input with the one-time data of the registers 34 and 38, but since the contents of the register 38 are "0", the contents of the register 34, that is, the number of drive timing pulses N, are input from the adder 36. VC will be output from the register 38. In this state, when the match signal of the humpator 40 is input to the register 38, the output of the adder 66, that is, the number N of drive timing pulses will be set in the register 38. As a result of this operation, the inputs of the comparator 40 no longer match, so the output of the match signal is stopped.At the same time, the X-Y stage 10 is sent in the Y direction. Along with this movement of the X-Y stage 10,
As shown in FIG. 4, the interference receiver 16 outputs pulses at predetermined pitches (0,0197756175 μm). This pulse is input to a counter 44, and the counter 44 counts up every time a pulse is input. It should be noted that the counter 42 has no carry especially due to the addition of the adder 56, so its content is "0".

As the count up of the counter 44 progresses and the count value reaches r22JK&, it will match the integer part "22" of the drive timing pulse number N stored in the register 38. As mentioned above, the counter 42 counts up. Since the content is "0", the inputs of the comparator 40 match υ, a match signal is output from the comparator 40, and the line sensor 20 is driven.

This situation! Considering lA in more detail, originally the number of drive timing pulses N = 22.7552907te5
Although the in-sensor 20 should be driven, it is driven with 22 pulses. To illustrate, the line sensor 20 is driven at the line sensor position as shown by the broken line in FIG. The area [1,7552907] is less than one pulse output from the X interference receiver 16. In scale, one pulse of the X interference receiver 16 is 0.0197756175 μm as described above. From, a7552
For 907 pulses, 0.0197756175X O,
7552907-0,015C#m]・Song・(3).

, when a match signal is output from the hump parator 40, it is input to the register 38, and the addition result of the adder 36 is set in the register 38. At this time, the adder 36 contains the drive timing pulse number N, which is the content of the register 34.
and the number N of ms timing pulses, which are the contents of the register 68, are respectively input. Therefore, the output of the adder 36 becomes 2N, which is set in the register 38.On the other hand, in the same way as in the case described above, the output pulse of the X interference receiver 16 due to the movement of the X-Y stage 10 The counter 44 counts up to 0, and the count value of the counter 44 becomes the set value W1 of the register 38. When the value s matches the integer part of 2N, a match signal is output from the comparator 40, and the line sensor 20 is driven.

Considering this case in detail, first the value of 2N is
It becomes 45.5105814. This is the contents of register 68. Note that no particular carry occurs in the addition of 2N, and the integer part of 2N is "45". Therefore, when the value of the counter 44 becomes "45", the
A coincidence signal is output from 0 to the line sensor 20. In this case, 0.510581 for the ideal position
The line sensor 20 is driven at a position 4 pulses less.
This is less than one pulse of the output pulse.

The above operation is repeated and the line sensor 20 is driven. In other words, in terms of the Y interference and the number of pulses of the interference receiver 16, the reading drive of the line sensor 20 is performed first with the 22nd pulse, and then the driving of the line sensor 20 is performed with the 23rd pulse. Comparing the driving position of the line sensor 20 in the example with the conventional one, it is as shown in the table below◇
Note that the position of the line sensor 20 is similar to that of the X interference receiver 16.
〇-if, N=22 expressed as the number of pulses
, TeL4 Udo, m5rXJ (C) (D Liketonashi, N
= 23, the result will be as shown in Figure 5 (■). Since N is not an integer to begin with, the side force of the line sensor 20 will not be applied at the ideal position even in this example, but the error will be is at most one output pulse of the Y interference receiver 16, specifically 0.0197756175
It is less than μm.

That is, the line sensor 20 is driven within a range of -0.0197756175 μm from the normal position.

Therefore, the light amount of the line sensor 20 is within the range in which the light amount can be corrected, and the positional error can be sufficiently ignored. That is, since the time interval of the sharpening pulses applied to the line sensor 20 (the imaging time for one line) changes by only one pulse from the interferometer receiver, the image signal for each line has I/(time and sight product)
Moreover, there is no large error, and the light amount correction operation by the AGC circuit in the processing circuit of the line sensor 20 is not significantly affected.

Note that the present invention is not limited to the above embodiments in any way. For example, in the above embodiment, the number of drive timing pulses N
The decimal part of the drive timing pulse number N is ignored and the integer part is compared with the output pulse version of the Y interference receiver 16, but the decimal part of the drive timing pulse number N is rounded off and compared. 10% pulse i.e. ±0.0('!988
If the line 7/20 is driven in the range of 7530875 μm and the decimal part is rounded up, the new J of +0.0197756175 μm for the normal digit i
The line sensor 20 will be driven by the recess.Also, in the above embodiment, only the Y direction will be described, and the same applies to the case where the hole is in the X direction. Further, although registers, counters, etc. are actually displayed in binary rather than decimal, the above embodiment assumes that they are displayed in decimal for convenience of explanation.

Although the above embodiment is an example applied to an imaging device, the invention may be applied to any other device.

〔Effect of the invention〕

As explained above, according to the pulse generator according to the present invention, pulse ? It has the effect of being able to output.

[Brief explanation of drawings]

Fig. fI51 is a block diagram showing one embodiment of the invention,
The figure is a schematic perspective view showing an example of an imaging device, Figure 5 is an explanatory diagram showing the scanning direction of the line sensor, Figure f24 is a time chart showing the output pulse of the interference receiver and the sensor drive signal @7, Figure 5 is the line It is an explanatory view showing an example of a pivotal position of a sensor. [Explanation of symbols of main parts] 10...X-Y stage, 12... Test sample, 14
゜16...Laser interferometer receiver, 20...Line sensor, 34.38...Register, 56...γder, 40...Comparator, 42.44...Counter agent patent attorney Masaru Sato Figure 2, Figure 4, Figure 5

Claims (1)

[Claims] A counting means counts output pulses from a detection means that detects the amount of physical change and generates and outputs a pulse for each unit amount of the change, and based on this counted value, the target is detected. In a pulse generator that generates a reference pulse every predetermined amount of change in the amount, reference data obtained by dividing the value of the predetermined amount of change by the value of the unit amount is stored in the necessary number of significant digits including the decimal part. a data storage means for adding the reference data of the data storage means with the required number of significant digits including a decimal part; and a count value of the counting means and an integer part of the added value of the addition means. 1. A pulse generating device comprising a comparing means for comparing the two numbers and causing the adding means to perform an addition operation and generating the reference pulse when the two match.