JPS6184505A - Position detecting device - Google Patents

Abstract

PURPOSE:To measure the position of a surface to be detected irrelevant to the surface shape without contacting by splitting luminous flux into two by using a light source which has short coherence distance, reflecting one piece of the luminous flux by the surface to be detected and passing the other piece through a reference optical system, and superposing the both again on each other and reading interference fringes. CONSTITUTION:Light from a light source 8 is collimated by a collimator 12 into parallel light, which is split into two through a polarization beam splitter 18. One light beam is converged on the object surface of a lens 1 to be detected by adjusting an objective lens 12 and then reflected. The other light beam is reflected by a corner cube 20. The corner cube 20 is movable in the y-direction and the quantity of the movement is measured. Two reflected light beams are superposed on each other and interference fringes are measured. The corner cube 20 is moved and when the output of a photoelectric element 33 peaks, the two split light beams become equal to optical path length.

Description

DETAILED DESCRIPTION OF THE INVENTION (Field of Application of the Invention) The present invention relates to a position detection method that detects the relative position between a surface to be inspected that reflects light, such as a lens surface, and the apparatus of the present invention by using optical interference. This invention relates to a detection device.

(Background of the Invention) Conventionally, as a device for measuring the position of a lens surface and the center thickness of a lens, there is a length measuring machine as shown in FIG. 74, which has a built-in linear encoder. 1 is a lens having a lens surface which is a surface to be inspected; 2 is a vertically movable spindle; 3 is an optical linear scale attached to the spindle 2; 4 is a mount; 5 is a mount; A sensor reads the amount of displacement of the scale 3; 6 is a bearing that supports the spindle 3 so that it can move accurately and smoothly; and 7 is a table surface. In addition, linear scale 3
, sensor 5 and an electronic circuit that processes the output of sensor 5 and displays the amount of displacement of linear scale 3 is called a linear encoder.

When measuring the position or center thickness of the upper surface of the lens 1, first lower the spindle 2 to the stage surface 7 with the lens l not placed on the stage surface 7, and then set the displayed value of the linear encoder to zero. set. Next, raise the spindle 2 sufficiently high, place the lens 1 on the stage surface 7, and
lower it to the top of lens 1. The value displayed by the linear encoder at that time indicates the position of the upper surface of the lens 1, that is, the center thickness of the lens 1 in the case of a biconvex lens as shown in FIG.

The problem with the length measuring machine shown in FIG. 7 is that it is a contact type, so it is easy to scratch the surface to be measured, and there is elasticity due to contact pressure between the tip of the spindle 2 and the surface to be measured or the stage surface 7. Deformation occurs, causing measurement errors, and furthermore, the inner surface of the object cannot be measured.

For example, it is impossible to measure the distance between the bonded surface and the surface of a bonded lens, or the air gap in a lens system that combines multiple lenses.

On the other hand, non-contact methods include light sectioning, triangulation, and methods that measure the amount of movement of the lens position that focuses on the test surface, but all of these methods measure the inner surface of the test object as the test surface. In some cases, it always depends on the shape of the surface of the object to be inspected, and there is a major problem in that measurement cannot be performed unless the shape is accurately known.

(Object of the Invention) An object of the present invention is to solve the above-mentioned problems and to provide a position detection device that does not come into contact with the surface to be measured and does not essentially depend on the shape of the surface of the surface to be measured.

(Characteristics of the Invention) In order to achieve the above object, the present invention applies the principle of Michelson's interferometer to detect the position of a surface to be measured, and in its application, uses an idea opposite to that of Michelson's interferometer. , using a light source with a short coherence length, that is, a somewhat wide spectral width, so that the visibility curve of the interference fringes is sensitive to changes in the optical path length of the reference optical system; A focus-adjustable objective lens is provided that focuses light onto the surface to be inspected and converts the reflected light beam from the surface to parallel light beams again.
The optical path length of one optical path changes only depending on the position of the test surface, the reference optical system changes the optical path length of the other divided parallel beam, and the amount of change in the optical path length of the reference optical system is measured. A length measuring means is provided, and while changing the optical path length of the reference optical system, the amount of change in the optical path length at a specific point of visibility, that is, the peak or inflection point, is measured, and from the amount of change, the amount of change in the optical path length is determined from the amount of change. It is characterized by detecting the position.

(Embodiment of the invention) As a method for measuring the width of the emission spectrum line of a light source, a visibility curve with respect to the optical path difference is obtained using a Michelson interferometer.
The method of calculating it has been known for a long time. The present invention is devised so that this principle can be applied to position detection.

FIG. 1 shows an optical system according to an embodiment of the present invention.
To explain the direction of movement, etc., the orthogonal coordinate system shown in FIG. 1 will be used. That is, the two axes are on the right, the y-axis is on the top, and the X-axis is on the bottom of the page. 8 is a light source, 9 is a heat ray absorption filter, 10 is a condenser lens, 11 is a pinhole, 12 is a collimator, and 13 is a bandpass filter for adjusting the spectral width and shape of the light source 8.

Light source 8, heat absorption filter 9, and bandpass filter 1
3 constitutes a light source having a short coherence length, that is, a light beam having a somewhat wide spectral width. A light source with a short coherence length is expressed in comparison with a laser, and while the coherence length of a laser is usually 100 mm or more, the coherence length of the light source used in the present invention is, for example, about 1 mm or less. In comparison with wavelength, it is about 1000 times or less of one wavelength.

When the spectrum is Gaussian and its center wavelength is 550 nm, the half width of the spectrum is 27 nm in order to set the optical path difference at which the visibility is 1/2 to be 5 pm. In other words, a suitable light source can be easily obtained using a halogen bulb and an interference filter. However, since only one light source is used, it is preferable to use a light emitting diode with high brightness and high luminous efficiency.

14 is a polarizing plate whose polarization direction is 45'' with respect to the X-axis and the y-axis; 15 is a beam splitter that slightly reflects light in order to monitor variations in the amount of light from the light source 8; 16 is a lens; 17 is a photoelectric element; A polarizing beam splitter 19 splits the incident light beam into two parallel light beams: a transmitted light beam whose polarization direction is in the y-axis direction and a reflected light beam whose polarization direction is in the X-axis direction. A quarter-wave plate 20 is arranged to be movable in the y-axis direction and has a driving means and a length measuring circuit for measuring the amount of movement thereof; 21 is a corner cube that measures the direction in which the phase advances; 45 for the x or y axis
The 174-wavelength plate 22.23 is a lens, which constitutes an objective lens 24 that focuses a parallel light beam on the surface to be measured of the lens 1 in the form of a spot. The lens 23 is movable independently in two axial directions, and the lenses 22 and 23 are movable together in the X-axis direction, and are adjusted to focus on the surface to be inspected. 25 is a beam splitter that guides some light flux to a monitor optical system that monitors whether the parallel light flux forms a spot on the test surface using the objective lens 24;
26, 27, and 28 are lenses, a beam bending mirror, and an eyepiece constituting the monitor optical system, 29 is a 174-wave plate whose phase direction is set at 45 degrees with respect to the X axis or the z axis, and 30 is a condenser. 31 is a beam splitter, 32 is a polarizing beam splitter, 33.34 is a photoelectric element, 35 is a polarizing beam splitter, and 36.37 is a photoelectric element.

The polarizing beam splitter 32 and the photoelectric elements 33,34 are
It is integrated and can be rotated around the y-axis,
The optical path length of the reference optical system consisting of the quarter-wave plate 19 and the corner cube 20, the quarter-wave plate 21, and the objective lens 24.
The rotation angle is adjusted such that the output signal of the photoelectric element 33 reaches its peak when the optical path length of the object optical system consisting of the test surface of the lens L and the test surface of the lens L is equal, and at that position, the photoelectric element 33 detects the polarization component of the polarization direction O0. , and the photoelectric element 34 detects the polarization direction 9
Detects the 0° polarization component. Polarizing beam splitter 35
The photoelectric elements 36 and 37 are integrally rotatable around two axes, and the photoelectric elements 36 and 37 can be rotated in the polarization direction 4.
The photoelectric element 37 detects the polarization component of 5°, and the polarization direction 13 is detected.
The rotation angle is adjusted to detect the 5' polarization component.

Regarding the polarization direction of the polarizing plate 14, if the ratio of the reflectance of the reference optical system to the reflectance of the object optical system is 1, the polarization direction is preferably 45°, but if the ratio is 1 or more or 1 or less, It is preferable that the polarization direction can be rotated so that (amount of incident light×reflectance) of the reference optical system and the object optical system are equal.

The signals output from the photoelectric elements 17, 33, 34, 36.37 and the length measurement circuit 38 (Fig. 2) that measures the distance of movement of the corner cube 20 are processed by a signal processing system. An example of the processing system is shown in FIG.

To explain the measurement procedure, first, the reference optical system is adjusted by moving the zero-order corner cube 20, which is adjusted while observing from the eyepiece 28, so that the focus of the objective lens 24 is on the test surface of the lens 1. and the optical path length of the object optical system, and the movement amount y0 of the corner cube 20 at that time is
Measure the length.

This amount of movement yo indicates the position of the surface to be inspected.

Next, the operation will be explained.

The configuration from the light source 8 to the bandpass filter 13 creates a parallel light beam with a short coherence length. This parallel light flux is 45° with respect to the X-axis and the y-axis by the polarizing plate 14.
It becomes a polarized light beam with the direction of . Polarizing beam splitter 1
The light beam reflected by 8 has a polarization direction in the X-axis direction, and 17
The light is circularly polarized by the four-wavelength plate 19, and the corner cube 20
The light is reflected by the 1/4 wavelength plate 19 and becomes linearly polarized light again, but since its polarization direction is in the X-axis direction, it passes through the polarizing beam splitter 18 . On the other hand, the light beam transmitted through the polarizing beam splitter 18 has a polarization direction in the y-axis direction, and becomes circularly polarized light by the 1/4 wavelength plate 21.
It passes through the objective lens 24, is reflected by the test surface of the lens 1, and passes through the 1/4 wavelength plate 21 again, becoming linearly polarized light again. However, since the polarization direction is in the X-axis direction, the polarizing beam splitter 18 reflect.

When the two orthogonal linearly polarized light beams emitted from the polarizing beam splitter 18 enter the 1/4 wavelength plate 29, they become two clockwise and counterclockwise circularly polarized lights, and their combination simply becomes linearly polarized light. The orientation changes depending on the optical path difference between the reference optical system and the object optical system. The ratio (intensity) of linearly polarized light to the total amount of light is proportional to visibility, and therefore changes depending on the optical path difference. The intensity and orientation of this linearly polarized light are detected by photoelectric elements 33, 34, 36, 37.

Note that since polarization directions 0° and 180° indicate exactly the same direction, the phase of the output signal from the photoelectric element 33, etc. is twice the polarization direction.
Corresponds to 600. Therefore, the phase difference between the output signals of the photoelectric elements 33.34 and the output signals of the photoelectric elements 36.37 is 180°. The phase difference is 90
°.

In the signal processing system shown in FIG.
The photoelectric element 34 receives the polarized light intensity signal A1 of the 180° phase, the photoelectric element 36 receives the polarized light intensity signal B1 of the 90° phase, and the photoelectric element 37 receives the polarized light intensity signal B2 of the 270° phase. , respectively, and a length measuring circuit 38 such as a linear encoder measures the amount of movement y of the corner cube 20 from the zero point and outputs the length. The zero point is the position where the optical path length between the corner cube 20 and the polarizing beam splitter 18 is equal to the optical path length between the position detection reference plane 39 (FIG. 1) of this device and the polarizing beam splitter 18. .

Each signal is offset and gain adjusted by amplifiers 40-44. As a result, if the voltage V of the polarized light intensity signal A1 amplified by the amplifier 41 is taken as the vertical axis, and the amount of movement y of the corner cube 20 or the time t when the corner cube 20 is moved at a constant speed is taken as the horizontal axis, It will look like Figure 3. When the movement amount yo with the largest amplitude is the optical path difference blade,
Coincident with peak visibility. On the other hand, if you put a sine wave on the horizontal axis and a cosine wave on the vertical axis on an oscilloscope, the trajectory will be the fourth one.
It is well known to draw a circle as shown in the figure. Since the sine wave and the cosine wave are out of phase by 90 degrees, the method for determining the amplitude of the polarization intensity signal A1 is to use the polarization intensity signal A,
By using the polarized light intensity signal B1 having a phase shift of 90° and squaring each of them and adding them, the square of the amplitude can be obtained.

However, as can be seen from FIG. 3, the center of vibration of the polarized light intensity signals A and B1 is not O, but ■.

It has become. This is because the amount of light does not have negative energy. A simple way to set the DC component v0 to 0 is to set a constant voltage v from the polarized light intensity signals A, B1.

However, this method is not wise because the DC component v0 depends not only on fluctuations in the light intensity of the light source 8, but also on the focus adjustment of the objective lens 24, the reflectance of the surface to be measured, etc.
Therefore, in FIG. 2, the polarization intensity signal A2 having a phase shift of 1800 degrees is subtracted by the subtraction circuit 45 from the polarization intensity signal Al. Similarly, the subtraction circuit 46 subtracts the polarized light intensity signal B2 having a phase shift of 1800 from the polarized light intensity signal B1.

Next, the division circuits 47 and 48 produce (A+-A2).

(B1-82) are respectively divided by the light amount fluctuation monitor signal C, and the output signals of the zero division circuit 47 and 48 that correct the light amount fluctuation are designated as A and B. Multiplying circuit 4 for signals A and B respectively
By squaring it by 9.50 and adding it by the addition circuit 51, a mountain-shaped curve having a peak at the movement amount yo where the optical path difference becomes zero is obtained, as shown in FIG. This corresponds to the visibility curve squared.

After the signal (A 2 +332) passes through the filter 52 for removing noise, it is converted to the signal shown in FIG. 6 (A) by the differentiating circuit S3.
It is differentiated into a waveform like . to is a time corresponding to the amount of movement yo, and is a zero crossing point. Pulse generation circuit 5
4 generates a pulse PO shown in FIG. 6(B) at time t0. On the other hand, the waveform shaping circuit 55 outputs a signal (A 2 +B
When the level of 2) is above the threshold value Lt, a high level signal is sent to the gate 56 to open it. This causes the pulse Po to pass through the gate 56.

During the detection operation, the corner cube 20 is moved in a direction that makes the optical path difference between the reference optical system and the object optical system zero, and the amount of movement y is
is measured and outputted by the length measuring circuit 38, but it is not input to the memory 58 because the gate 57 is normally closed. When the optical path difference between the reference optical system and the object optical system becomes zero, a pulse PO is generated and this pulse PO opens the gate 57, so the movement amount y0 at that time is input to the memory 58 and stored as the position of the surface to be inspected. be done.

When measuring the center thickness of the lens 1, the objective lens 24 is then focused on the rear lens surface of the lens 1, and the movement amount y1 is measured by the same operation as described above and stored in the memory 58. let If the refractive index of lens l is n, then
The center thickness d is determined by the calculation circuit 59 using the following calculation formula.

d= (y+ -yo)/n The method described above measures the amount of movement yo when the peak of visibility is detected, but the amount of movement Vo can be measured by using the inflection point of the visibility curve. It can also be measured. 55th!
If we differentiate the visibility squared curve shown in J twice, we get the 6th
The curve becomes the one shown in Figure (C), which crosses zero at the two inflection points of the squared visibility curve. This zero cross point LO1
Generate pulse P 01 + P 02 at +t02,
Movement amount at each point yor + '! By memorizing 02 and calculating yo = (Vot+Yoz)/2, the amount of movement 'lo can be obtained.

According to this embodiment, since it is a non-contact method in which a spot light is applied to the surface to be inspected, it is possible to prevent the surface to be inspected from being scratched or elastically deformed. Furthermore, it is possible to measure the position of the bonded surfaces of a bonded lens and the air spacing between multiple lenses. Furthermore, even when the surface to be measured is the inner surface of a lens, the optical path length of the object optical system is not affected by the radius of curvature of the lens surface, making the measurement simple.

(Correspondence between the invention and the embodiments) Light source 8, heat ray absorption filter 9, and bandpass filter 1
3 corresponds to a light source with a short coherence length of the present invention, the condenser lens 10, pinhole 11, and collimator 13 correspond to a collimating optical system, and the polarizing beam splitter 18 corresponds to a beam splitting means and a beam combining means. 1/4 wavelength plate 29
The area from to the gate 56 (excluding the length measuring circuit 38 and the position detection reference surface 39) corresponds to interference fringe reading means.

(Modification) The interference fringe reading means is not limited to the one shown in the illustrated embodiment, and may be replaced by various known means. Of course, a method that does not use polarized light may also be used.

Alternatively, finding the peak visibility may be performed using the human eye. In that case, a human eye is placed in place of the photoelectric element 33, and while the corner cube 20 is moved in the y direction, the length measurement value of the length measurement circuit 38 is read when the visibility is the best. Although it is not a good idea for accuracy improvement or automation, it has the advantage of being a simple and inexpensive device.

In the illustrated embodiment, a homodyne interference type is used in which the wavelength of light is only one with a certain spectral width, but it is possible to use a heterogain interference type in which the wavelength is slightly changed between the reference optical system and the object optical system. can.

Although the corner cube 20 is used as the reference optical system, it can be replaced with a cat's eye made of a lens and a concave or convex surface.

In order to change the optical path length of the reference optical system, the refractive index of the medium in the optical path may be changed. In particular, when measuring glass thickness, the optical path length can be changed by inserting glass of standard thickness as it is. The amount of mechanical change can be reduced.

Although the polarization beam splitter 18 serves both as a beam splitting means and a beam combining means, separate means may be used.

In the illustrated embodiment, the optical path length of the object optical system on the apparatus side is not changed by moving the focus-adjustable objective lens 24 only in the optical axis direction to focus the light on the test surface. Replace with another objective lens of known length,
The resulting change in optical path length may be corrected by calculation, and the objective lens 24 may be equipped with an automatic focusing mechanism.

(Effects of the Invention) As explained above, according to the present invention, by using a light source with a short coherence length, that is, a somewhat wide spectrum width, the visibility curve of interference fringes changes depending on the change in the optical path length of the reference optical system. The object lens is designed to be sensitive, and is equipped with an adjustable objective lens that focuses one of the split parallel light beams onto the test surface and returns the reflected light flux from the test surface to parallel light fluxes. The optical path length of one optical path changes only depending on the position of the surface, the reference optical system changes the optical path length of the other divided parallel beam, and the length measurement measures the amount of change in the optical path length of the reference optical system. By providing a means to measure the amount of change in the optical path length at a specific point of visibility while changing the optical path length of the reference optical system, and detecting the position of the test surface from the amount of change. The position of the test surface can be detected without contacting the test surface and essentially independent of the surface shape of the test surface.

[Brief explanation of drawings]

FIG. 1 is a layout diagram showing an optical system according to an embodiment of the present invention, and FIG.
The figure is a block diagram showing a signal processing system according to an embodiment of the present invention.
Figure 3 shows the waveform of the polarized light intensity signal, Figure 4 shows the amplitude of the polarized light intensity signal, Figure 5 shows the squared visibility curve, and Figure 6 shows the waveform during signal processing. The figure shown in FIG. 7 is a diagram showing a conventional length measuring machine. 1...Lens, 8...Light source, 9...
... Heat ray absorption filter, 10 ... Condensing lens, 11 ... Pinhole, 12 ... Collimator, 13 ... Bandpass filter, 14.
...Polarizing plate, 17, 33, 34.36.37...
...Photoelectric element, 18,32.35...Polarizing beam splitter, 19,21.29...1/
4 wavelength plate, 20... Corner cube, 24...
...Objective lens, 38... Length measurement circuit.

Claims (1)

[Claims] 1. A light source with a short coherence length, a collimating optical system that converts the light beam of the light source into a parallel beam, a beam splitting means that splits the parallel beam of the collimating optical system into two, and a beam splitting means that splits the parallel beam of the collimated optical system into two, and a focus-adjustable objective lens that focuses light onto the test surface and converts the reflected light beam from the test surface into a parallel light beam;
a reference optical system that changes the optical path length of the other divided parallel beam; a length measuring means that measures the amount of change in the optical path length of the reference optical system; and a parallel beam that has passed through the objective lens and the reference optical system, respectively. A position detection device comprising a beam combining means for re-superimposing the beams, and an interference fringe reading means for finding a specific point of visibility of the interference fringes of the overlapping parallel beams.