JP4111251B2 - Optical system manufacturing method and mold member manufacturing method - Google Patents

Description

The present invention relates to a method of manufacturing an optical system mounted to an image projection apparatus (projector) and the like, and to a method of manufacturing a mold member used in the manufacture of optical science system (mold, etc.).

  In recent years, optical elements formed of a resin such as plastic have been used for projection optical systems and laser scanning optical systems. This is because such an optical element made of resin is cheaper and lighter and more excellent in mass productivity than an optical element made of a glass material.

  The resin optical element is manufactured by a molding method such as injection molding or injection compression molding using a mold (mold member). Therefore, the resin-made optical element has an advantage that a curved surface (optical surface) having an aspherical shape or a free-form surface shape can be easily formed as compared with the glass-made optical element.

  However, when an optical element having an optical surface with a complicated shape such as an aspherical shape or a free-form surface is molded with a mold, surface defects are likely to occur due to uneven cooling and shrinkage on the optical surface. For this reason, it is generally performed to measure such a defective surface shape and correct the mold based on the measurement result (Patent Document 1).

And as an example of the measurement of a surface shape, patent document 2 is mentioned. In the method of Patent Document 2, the optical performance of an optical element is obtained by performing polynomial approximation of an aspheric surface of an optical surface using data obtained by three-dimensional measurement of the optical surface, and further incorporating such polynomial approximation surface in an optical system. (Aberration change) is evaluated. Then, by evaluating the optical performance, the optical surface (optimum shape optical surface) required to exhibit the desired optical performance is found, and the mold is corrected so that it becomes the optimum shape optical surface. May be.
Japanese Patent Laid-Open No. 7-60857 JP 2004-361274 A

  However, in the evaluation method of Patent Document 2, an error between optical surface measurement data (three-dimensional measurement data) and predetermined design value data is approximated by a polynomial (error polynomial) as a whole surface. Therefore, when local waviness (waviness portion) exists on the measured optical surface, if an approximation expression of higher order is used to appropriately represent the waviness portion, a place other than waviness (non-waviness portion) High-order undulation will occur. On the other hand, if a low-order approximation expression is used to appropriately represent the non-swelled part, the swelled part cannot be represented properly.

  Further, in the optical element molding method of Patent Document 1, in order to appropriately correct the waviness portion, the gold corresponding to the optical surface is canceled so that the error between the measurement data of the optical surface and the predetermined design value data is offset. The cavity surface of the mold is corrected and the optical element is reshaped.

  However, in such a molding method, in the case of an optical system having a plurality of optical elements, it is necessary to correct the mold of each optical element with the aim of the design value shape. Therefore, at least the number of mold corrections is required as many as the number of optical surfaces of the optical element, and it takes a very long time to manufacture the optical system.

The present invention has been made in view of the above situation. An object of the present invention is a method for manufacturing an optical system (more specifically, an optical system including a plurality of optical elements) in a short time and at a low cost while being efficient , and a mold used for manufacturing the optical system. The object is to provide a method for manufacturing a member (mold or the like) .

  The present invention is a method of manufacturing an optical system including a plurality of optical surfaces by molding using a mold member. In this optical system manufacturing method, the optical performance of the entire system is determined from the step of setting the approximate surface of all the optical surfaces of the optical system including the optical surface formed by the initial molding, and the approximate surface of all the optical surfaces in the optical system. A first optical performance evaluation step for evaluating the above, and at least one of the approximate surfaces of all the optical surfaces formed by the mold member as a change approximate surface, while being molded by the mold member of the approximate surfaces of all the optical surfaces A change amount calculation step for obtaining a change amount of a change approximate surface when at least one surface is made a non-change approximate surface instead of a change approximate surface and the optimum optical performance as the entire system is exhibited, and a change approximation based on the change amount The first correction processing step for obtaining a correction processing amount for the cavity surface of the mold member corresponding to the surface to perform correction processing to produce a new cavity surface, and the correction mold member processed in the first correction processing step is used for optical processing. First molding process to mold the element and correction A second forming step of forming an optical element using a mold member other than the member include.

  According to this manufacturing method, not all the cavity surfaces in all the mold members are corrected. That is, in this manufacturing method, for example, a desired optical surface can be used as a change approximate surface, and the corresponding cavity surface can be preferentially corrected. Therefore, the number of corrections of the mold member is relatively reduced, and the optical system is manufactured in a short time and at a low cost while being efficient.

  The first molding step includes the step of setting an approximate surface based on the measurement result of the surface shape of the optical surface corresponding to the new cavity surface in the optical element molded with the mold member having the new cavity surface, A second correction processing step for correcting the new cavity surface so as to cancel out the shape error between the approximate surface of the optical surface corresponding to the surface and the new cavity surface, and a correction mold member processed in the second correction processing step And forming an optical element using the method.

  The second correction processing is performed only on the cavity surface (new cavity surface) that has already been corrected. For this reason, the second correction processing has only a light burden as compared with the burden in the case of correcting all the cavity surfaces. In addition, the optical performance of the optical system falls within an allowable range if correction processing that cancels the shape error is performed with high accuracy. Then, since this lightly burdened second correction process exists, the optical system manufacturing method can easily manufacture an optical system that exhibits high optical performance.

  By the way, in the step of setting the approximate surface, it is desirable that the approximate surface based on the measurement surface shape of the optical surface formed by the initial molding is set as the optical surface formed by the molding using the mold member. .

  In the step of setting the approximate surface, when the optical surface is a polished surface, it is desirable that design data of the optical surface is set as the approximate surface on the polished surface.

  Further, it is desirable that an approximate surface having the largest difference between the approximate surface of each optical surface in the optical system and the design data of the optical surface corresponding to each approximate surface is a change approximate surface.

  Further, it is desirable that the approximate surface of the optical surface closest to the optical surface having the largest difference between the approximate surface of each optical surface in the optical system and the design data of the optical surface corresponding to each approximate surface is a change approximate surface.

  In addition, it is desirable that the change approximation plane is at most two planes.

  Further, in the step of setting the approximate surface, a space division setting step of dividing a plane perpendicular to a predetermined reference axis on the optical surface of the optical element into a plurality of pieces and setting a plurality of spaces with the divided planes as a bottom. It is desirable that an approximate expression having continuity is used at the boundary between spaces.

  In this manufacturing method, it is desirable that the approximate expression is a function of at least a third order. Note that continuity means that the second derivative of the approximate expression is continuous at the boundary between the divided spaces. An example of such an approximate expression is a spline function.

  The optical system preferably includes at least one lens and at least one mirror as elements molded by the mold member, and at least the lens surface is used as a change approximation surface.

  Moreover, it is desirable that at least one of the lens and the mirror included in the optical system is formed by glass molding.

  In addition, the optical system preferably includes a non-rotationally symmetric surface, and in the change amount calculating step, an approximate surface corresponding to the non-rotationally symmetric surface is preferably a change approximate surface.

  The optical system includes at least one lens and at least one mirror as elements molded by the mold member, and at least one lens surface is a non-rotationally symmetric surface. It is desirable that the non-rotationally symmetric surface be a change approximation surface.

  In the case where at least one lens and at least one mirror are included as optical elements in the optical system (in particular, at least one of the lens and mirror included in the optical system is glass-molded. It is desirable that at least one lens surface is used as the change approximation surface.

  According to the method for manufacturing an optical system of the present invention, it is possible to preferentially correct a cavity surface corresponding to an optical surface that easily affects optical performance. Therefore, the number of corrections of the mold member is relatively reduced, and the optical system is manufactured in a short time and at a low cost while being efficient.

These are the flowcharts which show the process of the manufacturing method of a projection optical system. These are top views which show the measurement condition of the lens surface in a 2nd lens. These are top views which show the division space in a lens surface. These are spot diagrams of the projection optical system including only the initial product. These are spot diagrams of a projection optical system including a reshaped product. Is a spot diagram of a projection optical system that includes an optical element made of glass and an optical element made of resin and that is composed only of an initial product (however, the negative half of the Z-axis direction on the screen surface) . FIG. 8 is a spot diagram showing a half on the positive side in the Z-axis direction on the screen surface, unlike FIG. Is a spot diagram of a projection optical system that includes a glass optical element and a resin optical element, and of which a remolded product is included (however, on the negative side of the Z-axis direction on the screen surface) Half). FIG. 9 is a spot diagram showing a half on the positive side in the Z-axis direction on the screen surface, unlike FIG. These are the block diagrams of an image projector.

Explanation of symbols

PS Projection optical system (optical system)
M1 first mirror (optical element)
L1 first lens (optical element)
M2 second mirror (optical element)
L2 Second lens (optical element)
M3 third mirror (optical element)
M4 4th mirror (optical element)
FM flat mirror (optical element)
SCN Screen PDS Image Projector F Optical element design value data f Initial cavity surface data in mold G Initial surface data in initial product D New cavity surface data

[Embodiment 1]
An embodiment of the present invention will be described below with reference to the drawings.

[1. About Image Projection Device]
FIG. 10 shows a screen SCN, an illumination optical system (not shown), a light modulation element MD that modulates light from the illumination optical system, and light (image light) modulated by the light modulation element MD. 1 shows an image projection apparatus PDS including a projection optical system PS leading to The image projection apparatus PDS is a rear projection type that obliquely enlarges and projects light from the light modulation element MD (reduction side) toward the back of the screen SCN (enlargement side).

  The projection optical system PS includes a first mirror (spherical mirror) M1, a first lens (rotational asymmetric free-form surface lens) L1, and a second mirror according to the order of light traveling from the light modulation element MD to the screen SCN. (Rotationally symmetric aspherical mirror) M2, second lens (rotationally asymmetric free-form surface lens) L2, third mirror (rotationally asymmetric freeform surface mirror) M3, fourth mirror (rotationally asymmetric free-form surface mirror) M4, and plane mirror FM It is arranged.

  A protective cover glass CG is disposed on the front surface of the light modulation element MD (emitted surface of the modulated light), and a part of the light is interposed between the first mirror M1 and the first lens L1. An optical stop ST for shielding light is disposed. Further, the plane mirror FM is a folding mirror that reflects and reflects the light from the fourth mirror M4 to the screen SCN.

[2. Projection optical system manufacturing process]
Here, a method for manufacturing the projection optical system (optical system) PS will be described with reference to the flowchart of FIG. 1 (operations STEP1 to STEP22). Since the flat mirror FM is not a curved reflecting surface, it is assumed that another manufacturing method is used instead of resin molding which is advantageous for the subsequent curved surface production. Further, since the flat mirror FM hardly affects the optical performance, the flat mirror FM is not included in the projection optical system PS in the subsequent optical performance evaluation or the like.

<STEP1: Optical design process for all optical elements>
First, six (8 surfaces) optical designs corresponding to the first mirror M1, the first lens L1, the second mirror M2, the second lens L2, the third mirror M3, and the fourth mirror M4, which are optical elements, are performed. Is called.

  Specifically, the reflecting surface S (M1) of the first mirror M1, which is the optical surface (S), the reduction side lens surface S (L1s) and the enlargement side lens surface S (L1e) of the first lens L1, and the second mirror. The reflection surface S (M2) of M2, the reduction side lens surface S (L2s) and the enlargement side lens surface S (L2e) of the second lens L2, the reflection surface S (M3) of the third mirror M3, and the fourth mirror M4 The reflecting surface S (M4) is expressed by a polynomial (the surface expressed in this way may be called a design optical surface).

  However, such a design optical surface is set so as to exhibit desirable optical performance as the entire projection optical system PS (entire system). The polynomial representing the design optical surface is “F”, and the polynomial F corresponding to each optical surface is listed below. The design optical surface indicated by the polynomial F may be referred to as “design data”.

Design optical surface corresponding to the reflecting surface S (M1) of the first mirror M1: F [S (M1)]
Design optical surface corresponding to the reduction side lens surface S (L1s) of the first lens L1: F [S (L1s)]
Design optical surface corresponding to the enlargement side lens surface S (L1e) of the first lens L1: F [S (L1e)]
Design optical surface corresponding to the reflecting surface S (M2) of the second mirror M2: F [S (M2)]
Design optical surface corresponding to the reduction side lens surface S (L2s) of the second lens L2: F [S (L2s)]
Design optical surface corresponding to the enlargement side lens surface S (L2e) of the second lens L2: F [S (L2e)]
Design optical surface corresponding to the reflective surface S (M3) of the third mirror M3: F [S (M3)]
Design optical surface corresponding to the reflective surface S (M4) of the fourth mirror M4: F [S (M4)]

ただし、
X ：高さHの位置でのX方向の基準面からの変位量（面頂点基準）
H ：X軸に対して垂直な方向の高さ[H=√(Y^２+Z^２)]
Co ：面頂点での曲率
ε ：２次曲面パラメータ
Ai ：i次の非球面の係数
Gjk：Yのj次、Zのk次の自由曲面の面係数
である。As an example of the polynomial representing the design optical surface, the following equation (Equation 1) using local orthogonal coordinates (X, Y, Z) with the surface vertex of each design optical surface as the origin can be given.

However,
X: Displacement from the reference plane in the X direction at the height H position (plane vertex reference)
H: Height in the direction perpendicular to the X axis [H = √ (Y ² + Z ² )]
Co: curvature at the surface vertex ε: quadric surface parameter
Ai: i-th order aspheric coefficient
Gjk: The surface coefficient of the free-form surface of the jth order of Y and the kth order of Z.

<STEP 2: Mold design process for all optical elements>
Next, a mold (first mold to eighth mold) corresponding to each optical element is designed. For this purpose, machining data (Numerical Control Data; NC data) corresponding to the design data of all optical surfaces is created. However, the processing tool is a diamond cutter having a tip having an R shape. Therefore, the contact point of the diamond cutter tip with respect to the processing surface changes with the change in the shape of the workpiece in the mold. Therefore, machining data is created by calculating machining point coordinates in consideration of the shape of the machining tool.

  Note that the cavity surface of the mold (the surface of the mold for molding the optical surface) is plated with phosphorus-containing nickel. Accordingly, the plated surface is processed with a diamond cutter, and the entire surface of the plated surface is evenly polished to complete the cavity surface.

And the cavity surface of each metal mold | die (1st metal mold | die-8th metal mold | die) can also be expressed using polynomial "f". Therefore, the polynomial f corresponding to each cavity surface (T) is listed below. The cavity surface represented by the polynomial f may be referred to as “initial cavity surface data”.

The cavity surface T (M1) of the first mold corresponding to the reflecting surface S (M1) of the first mirror M1
: F [S (M1)]
The cavity surface T (L1s) of the second mold corresponding to the reduction side lens surface S (L1s) of the first lens L1
: F [S (L1s)]
The cavity surface T (L1e) of the third mold corresponding to the magnification side lens surface S (L1e) of the first lens L1
: F [S (L1e)]
-Cavity surface T (M2) of the fourth mold corresponding to the reflective surface S (M2) of the second mirror M2.
: F [S (M2)]
-Cavity surface T (L2s) of the fifth mold corresponding to the reduction side lens surface S (L2s) of the second lens L2.
: F [S (L2s)]
-Cavity surface T (L2e) of the sixth mold corresponding to the enlargement side lens surface S (L2e) of the second lens L2.
: F [S (L2e)]
-Cavity surface T (M3) of the seventh mold corresponding to the reflective surface S (M3) of the third mirror M3
: F [S (M3)]
-Cavity surface T (M4) of the eighth mold corresponding to the reflective surface S (M4) of the fourth mirror M4
: F [S (M4)]

<STEP3: Initial molding process of all optical elements>
Then, all the optical elements are manufactured by injection molding the resin, which is the optical element material, using the first to eighth molds (this initial optical element is referred to as “initial product”). This molding may be referred to as “initial molding”).

  However, reproducible molding must be performed under molding conditions with less variation in surface accuracy of the molded optical surface. For this purpose, the most reproducible results are obtained by testing various molding conditions (parameters) such as the temperature at which the optical element material is melted (resin temperature), the mold temperature of the mold, and the injection speed and injection pressure when the optical element material is injected. Good molding conditions are set.

For example, the molding conditions of the second lens L2 are as follows. Note that the second lens L2 manufactured under the following molding conditions has a thickness (core thickness) of 3.5 mm and a maximum effective area EA (see FIGS. 2 and 3 described later) having a diameter of 52 mm.
・ Resin temperature: 285 ℃
-Mold temperature: 135 ° C
・ Injection speed: 15 mm / s
Injection pressure: 1050 kg / cm ²

<STEP 4: Surface shape measurement process for all initial products>
Next, the shape measurement of the optical surface in all initial products is performed. In addition, although the apparatus which measures a surface shape is not limited, For example, Matsushita Electric Industrial's ultrahigh precision three-dimensional measuring machine UA3P, rank tailor Hobson's form talissurf, etc. are mentioned.

  The measurement state is as shown in FIG. 2 (however, FIG. 2 measures the reduction side lens surface S (L2s) of the second lens L2). FIG. 2 shows a local right-handed orthogonal coordinate system (X, Y, Z) in which the surface vertex of the lens surface S (L2s) is the origin and the normal direction of the optical surface is the X axis (reference axis). It is illustrated on the basis. Specifically, a YZ plane (a plane composed of a Y-axis direction and a Z-axis direction) is illustrated.

And this measuring method performs line measurement along one direction (Z-axis direction) in the YZ plane. However, in this measurement method, line measurement is performed at a pitch interval of 0.15 mm (measurement interval in the Z-axis direction). The measurement interval in the Y-axis direction is 1.0 mm, and the measured point may be referred to as “measurement point MP”.

Further, if the measurement point MP does not exist around the light ray passage area (effective area EA) on the lens surface S (L2s), the accuracy of the approximate surface of the initial product, which will be described later, is reduced. For this reason, the line measurement is performed in an area wider than the effective area EA (for example, a range wider than the effective area EA in the Y-axis direction and the Z-axis direction).

  Note that the raw data when measuring the optical surface includes the amount of shift in the X-axis, Y-axis, and Z-axis directions and the X-axis, Y-axis, and Z-axis rotations in addition to the amount of change in molding. It includes setting errors when measuring rotation. Therefore, the measurement data composed of the measurement points MP is calculated in consideration of these setting errors.

<STEP5: Approximate surface setting process for all initial products>
Subsequently, the approximate surface of the optical surface in all initial products is set using the measurement data. However, for such an approximate plane setting, as shown in FIG. 3, the YZ plane (plane perpendicular to the X axis) covering the effective area EA is equally divided into 25 (Y axis direction 5 divisions × Z axis direction). A plurality of spaces (spaces in which the thickness of the divided surface extends in the X-axis direction; divided spaces) with the divided YZ plane (divided surface DA) as the bottom is set [space division setting step ].

  Then, the optical surface (initial surface) of the initial product is approximated by a polynomial in consideration of the plurality of divided spaces. An example of a function suitable for such approximation is a spline function (B-spline function or the like). Therefore, the spline function will be described below for explanation. The spline function is defined as follows (however, it is a case of a quintic spline function).

The k-direction vector in the X direction when the section is divided into n in any X direction is expressed as (x0, x0, x0, x0, x0, x0, x1, x2, x3, x4, ..., x (n-1), xn, The basis function of the B-spline defined when xn, xn, xn, xn, xn) is expressed by the following equation (Equation 2).

Similarly, the k-direction vector in the Y direction when it is divided into m in the Y direction is (y0, y0, y0, y0, y0, y0, y1, y2, y3, y4, ..., y (m-1), ym, ym , ym, ym, ym, ym), the B-spline basis function defined is expressed by the following equation (Equation 3).

In such a case, the surface function f (x, y) of the fifth-order B-spline function is defined by a linear sum (Equation 4) of (n + 5) × (m + 5) spline bases.

ｂi,k（x）は、下記式（数７）でのみ値を持ち、それ以外では“０”である。

However, the k-th order B-spline basis function bi, k (x) is a function represented by the following equation (Equation 5 / Equation 6):

bi, k (x) has a value only in the following equation (Expression 7), and is “0” otherwise.

  Subsequently, the spline functions G corresponding to the optical surfaces of the optical elements of the initial product are listed. The approximate surface indicated by the spline function G may be referred to as “initial surface data”.

-Approximate surface corresponding to the reflective surface S (M1) of the first mirror M1 of the initial product: G [S (M1)]
-Approximate surface corresponding to the reduction-side lens surface S (L1s) of the first lens L1 of the initial product: G [S (L1s)]
-Approximate surface corresponding to the magnification side lens surface S (L1e) of the first lens L1 of the initial product: G [S (L1e)]
-Approximate surface corresponding to the reflective surface S (M2) of the second mirror M2 of the initial product: G [S (M2)]
-Approximate surface corresponding to the reduction side lens surface S (L2s) of the second lens L2 of the initial product: G [S (L2s)]
-Approximate surface corresponding to the magnification side lens surface S (L2e) of the second lens L2 of the initial product: G [S (L2e)]
-Approximate surface corresponding to the reflective surface S (M3) of the third mirror M3 of the initial product: G [S (M3)]
-Approximate surface corresponding to the reflective surface S (M4) of the fourth mirror M4 of the initial product: G [S (M4)]

  Table 1 and Table 2 list approximate surface coefficients of the spline function G [S (L2e)] indicating the approximate surface corresponding to the magnification side lens surface S (L2e) of the second lens L2 as an example. .

《Knot vector coefficient》

《Coefficient of function basis of surface》

<STEP6: Optical performance evaluation process for the entire system consisting of initial products>
The initial surface data (G [S (M1)], G [S (L1s)], G [S (L1e)], G [S (M2)], G [S (L2s)], G [S ( L2e)], G [S (M3)], G [S (M4)]), the optical simulation apparatus evaluates the optical performance of the projection optical system PS [Optical performance evaluation of the initial product type optical system Step (first optical performance evaluation step)]. Accordingly, initial surface data corresponding to all optical surfaces is input to the optical simulation software in the optical simulation apparatus.

  Various optical simulation apparatuses exist in the prior art, and are not particularly limited. Also, the item of optical performance evaluation is not particularly limited. For example, various aberration evaluations, magnification evaluations, MTF (Modulation Transfer Function) evaluations, or spot diagram evaluations may be used.

  Therefore, FIG. 4 shows a spot diagram as an example of optical performance evaluation. The spot diagram of FIG. 4 is a spot diagram calculated from the initial plane data, and is imaged by overlapping spot diagrams of three wavelengths (460 nm, 546 nm, and 620 nm) at 25 evaluation points on the screen surface SCN. The characteristics are shown (the scale is expressed as ± 1 mm). The coordinates (Y, Z) in the figure are local coordinates (Y, Z; mm) of the screen surface SCN indicating the projection position of the spot centroid of each evaluation point. Since the projection optical system PS is an optical system that is plane-symmetric with respect to the XY plane of the screen surface, the spot diagram shows only the negative half of the Z-axis direction on the screen surface SCN, and the other half is omitted. ing.

<STEP7: Judgment process of optical performance evaluation result of projection optical system consisting of initial product>
Here, determination of the optical performance evaluation result, that is, whether or not the optical performance of the projection optical system PS made of the initial product is within an allowable range is determined. The subsequent steps are determined based on the determination result.

  For example, when the optical performance of the projection optical system PS made of an initial product is within an allowable range, the projection optical system PS has sufficiently satisfactory optical performance. Therefore, the mold does not have to be corrected. Therefore, the production of the projection optical system PS may be completed (STEP 7 → STEP 20).

  On the other hand, when the optical performance of the projection optical system PS made of the initial product is out of the allowable range, the projection optical system PS has optical performance that cannot be fully satisfied. For this reason, the mold must be corrected. Therefore, the process of obtaining new cavity surface data (referred to as “new cavity surface data”) of the mold and performing re-molding with the corrected mold according to the new cavity surface data. I will explain.

<STEP8: Correction amount calculation step for specific optical surface>
Usually, initial surface data and design data in each optical element are often different. However, it is not necessary to correct the molds of all the optical elements, and a projection optical system capable of achieving sufficient optical performance by newly setting the mold shape of the optical surface (specific optical surface) of the specific optical element PS can be produced.

  Specifically, the initial surface data other than the specific optical surface is fixed, the specific optical surface is changed, and the projection optical system PS is redesigned. However, although the specific optical surface may be any surface, it is preferable that the surface (maximum shape error surface) having the largest deviation between the initial surface data and the design data is the specific optical surface. This is because the initial surface data of the remaining optical surfaces other than the specific optical surface has little deviation from the design data, and as a result of redesigning the specific optical surface, there is little deviation from the design performance of the projection optical system PS. is there.

  The specific optical surface may be a surface closest to the maximum shape error surface. There are cases where it is not desired to correct the mold shape, for example, when the maximum shape error surface is a flat surface. In such a case, the specific optical surface may be the surface closest to the maximum shape error surface. This is because the state of the light incident on the surface closest to the maximum shape error surface is close to the state of the light incident on the maximum shape error surface, and as a result of redesigning the specific optical surface, there is little change from the design performance. is there.

  Hereinafter, the specific optical surface is the reduction-side lens surface S (L2s) of the second lens L2, and the cavity surface T (L2s) of the fifth mold corresponding to the reduction-side lens surface S (L2s) of the second lens L2 is determined. A case where correction processing is performed will be described as an example. The correction processing to the cavity surface T (L2s) corresponding to the specific optical surface may be referred to as “first correction processing”.

First, the new cavity surface data of the cavity surface T (L2s) subjected to the first correction processing is defined as “D” and defined as follows.
D = f [S (L2s)] + H
However,
f [S (L2s)]: Initial cavity surface data of cavity surface T (L2s)
H: A function indicating the correction amount for the existing cavity surface T (L2s) (for example,
, Spline function)
It is.

  Then, if the coefficient included in the function H (hereinafter described as an example spline function H) is determined, the new cavity surface data D is determined. Therefore, optical performance evaluation is performed using initial surface data of the reduction-side lens surface S (L2s) of the second lens L2 in consideration of the spline function H, that is, “G [S (L2s)] + H”.

  Specifically, initial surface data (G [S (M1)], G [S (L1s)], G [S (L1e)], G [S (M2)], G [S (L2s)] + H · G [S (L2e)], G [S (M3)], G [S (M4)]) is referred to so that the optical simulation apparatus evaluates the optical performance of the projection optical system PS. The evaluation may be referred to as “search optical performance evaluation”, and the result of this search optical performance evaluation may be referred to as “search result”).

  More specifically, the retrieval optical performance evaluation is performed by design data (F [S (M1)], F [S (L1s)], F [S (L1e)], F [S (M2)], F in the projection optical system PS. [S (L2s)], F [S (L2e)], F [S (M3)], F [S (M4)] We are changing and optimizing. The optimization here does not necessarily mean finding the best value by searching for a so-called local minimum. And the result (search result) which performed optimization should just have desired performance. When the optimum result is realized by the search optical performance evaluation, the spline function H is determined.

  Accordingly, the initial surface data G [S (L2s)] of the reduction-side lens surface S (L2s) of the second lens L2 is changed (that is, G [S (L2s)] + H is set), thereby projecting. The amount of change (that is, the spline function H) of the initial surface data G [S (L2s)] of the reduction side lens surface S (L2s) of the second lens L2 when the optimum performance as the optical system PS is exhibited is obtained. It can be said that [change amount calculation step]. Instead of directly changing the initial plane data G [S (L2s)] and obtaining the change amount as the difference, optimization may be performed using the correction amount as a variable.

  Tables 3 and 4 show coefficients in the spline function H determined thereafter. When such a coefficient is determined, the correction processing amount to be applied to the cavity surface T (L2s) of the fifth mold is also determined according to the coefficient.

《Knot vector coefficient》

《Coefficient of function basis of surface》

<STEP9: Mold correction process corresponding to specific optical surface>
Then, the cavity surface T (L2s) of the fifth mold is first corrected according to the determined correction processing amount [first correction processing step]. Therefore, this first correction processing can be said to be correction processing for realizing the cavity surface T (L2s) according to the new cavity surface data D.

<STEP10: Molding process with corrected mold>
Next, a new second lens L2 is molded (re-molded) with the first correction-processed fifth mold. The second lens L2 reshaped in this way may be referred to as a “reshaped product”.

<STEP11: Surface shape measurement process of remolded product>
Subsequently, as in STEP 4, the reduction-side lens surface S (L2s) of the new second lens L2 is measured by a device for measuring the surface shape. Such measurement data may be referred to as “re-molded product measurement data”.

<STEP12: Approximate surface setting process of remolded product>
Further, as in STEP 5, the approximate surface (approximate surface of the specific optical surface) of the reduction side lens surface S (L2s) of the second lens L2 of the reshaped product is set using the reshaped product measurement data. The approximate surface of the lens surface S (L2s) set in this way may be referred to as “re-formed product surface data” (G ′ [S (L2s)]).

<STEP 13: Optical performance evaluation process of projection optical system including remolded product>
The initial surface data (G [S (M1)], G [S (L1s)], G [S (L1e)], G [S (M2)], G [S (L2e)], G [S ( M3)], G [S (M4)]) and the remolded product surface data (G ′ [S (L2s)]), the optical simulation apparatus, like STEP 6, performs the optical simulation of the projection optical system PS. Perform performance evaluation [Optical performance evaluation process of remolded product-containing optical system].

  That is, the projection optical system PS includes the first mirror M1, the first lens L1, the second mirror M2, the third mirror M3, and the fourth mirror M4 that are initial products, and the second lens L2 that is a reshaped product. Optical performance evaluation will be performed. The result of the optical performance evaluation is a spot diagram shown in FIG. 5 (this FIG. 5 is expressed in the same way as FIG. 4).

<STEP 14: Judgment Step of Optical Performance Evaluation Result of Projection Optical System>
Then, it is determined whether or not the result of the optical performance evaluation of the projection optical system PS including the reshaped product (such as the spot diagram in FIG. 5) is within an allowable range. If the optical performance of the projection optical system PS including the reshaped product is within an allowable range, the projection optical system PS has sufficiently satisfactory optical performance. Therefore, the manufacturing may be completed (STEP 14 → STEP 20).

<STEP 15: Drive-in process to new cavity surface data>
However, when the optical performance of the projection optical system PS including the reshaped product is out of the allowable range, the projection optical system PS has optical performance that cannot be sufficiently satisfied. Such insufficient optical performance can be attributed to the fact that the cavity surface T (L2s) of the fifth mold subjected to the first correction processing is not processed according to the new cavity surface data D, or optical This may be due to uneven cooling on the surface, shrinkage during cooling of the optical element, molding conditions are not optimized, and the like.

  Therefore, correction processing (follow-up processing) is performed to offset the shape error between the new cavity surface data D and the reshaped product surface data G ′ [S (L2s)] determined in STEP 12 [second correction processing] Process].

<STEP 16: Molding process with die after follow-up processing>
Then, the second lens L <b> 2 is molded again (follow-up molding) with the fifth mold having the cavity surface T (L <b> 2 s) subjected to the follow-up process. The second lens L <b> 2 that has been subjected to the follow-up molding may be referred to as a “drive-in product”.

<STEP 17: Surface shape measurement process of driven-in products>
Subsequently, as in STEP 4, the reduction-side lens surface S (L2s) of the second lens L2 that is a driven-in product is measured by a device that measures the surface shape. Such measurement data may be “follow-up product measurement data”.

<STEP18: Approximate surface setting process for driven-in products>
Further, similar to STEP 5, the approximate surface of the reduction side lens surface S (L2s) of the second lens L2 of the driven-in product is set using the driven-up product measurement data. The approximate surface of the set lens surface S (L2s) may be referred to as “run-up product surface data” (G ″ [S (L2s)]).

<STEP19: Optical performance evaluation process of projection optical system including follow-up product>
The initial surface data (G [S (M1)], G [S (L1s)], G [S (L1e)], G [S (M2)], G [S (L2e)], G [S ( M3)], G [S (M4)]) and the driven surface data (G ″ [S (L2s)]), the optical simulation apparatus, as in STEP 6, performs the optical simulation of the projection optical system PS. Perform performance evaluation [Optical performance evaluation process of driven-in type optical system]. That is, the optical system of the projection optical system PS includes the first mirror M1, the first lens L1, the second mirror M2, the third mirror M3, and the fourth mirror M4, which are initial products, and the second lens L2, which is a driven product. Performance evaluation will be performed.

<STEP 14: Judgment Step of Optical Performance Evaluation Result of Projection Optical System>
Then, from the result of the optical performance evaluation of the projection optical system including the driven-in product, it is determined whether or not the optical performance is within an allowable range (STEP 19 → STEP 14). If the optical performance of the projection optical system PS including the driven-in product is within an allowable range, the projection optical system PS has sufficiently satisfactory optical performance. Therefore, the manufacturing may be completed (STEP 19 → STEP 14 → STEP 20).

  However, when the optical performance of the projection optical system PS including the driven product is out of the allowable range, the projection optical system PS has optical performance that cannot be sufficiently satisfied. Therefore, the follow-up process is performed again (re-follow process; multiple times STEP14 → STEP15).

  Further, the surface shape of the second lens L2 (re-dried product) molded by the fifth die after the re-rolling process is measured, and an approximate surface is set with the measurement data to evaluate the optical performance ( STEP16-STEP19). Then, from the result of the optical performance evaluation (from the result of the optical performance evaluation of the retracked product-containing optical system), if the optical performance of the projection optical system PS including the retracked product is within the allowable range, the manufacturing is completed. (STEP 19 → STEP 14).

  However, if the optical performance of the projection optical system PS including the re-driving product is out of the allowable range, the cavity surface T (L2s) of the fifth mold may be re-rolled again. That is, the follow-up process or the like is continued until the optical performance as the projection optical system PS is within the allowable range in STEP 14 (from STEP 19 to YES in STEP 14) (STEP 15 to STEP 19 are repeated).

  In STEP 14, a mold capable of forming an optical element having an optical performance within the allowable range as the projection optical system PS is referred to as a correction mold (correction mold member).

<STEP 20-22: Completion of optical system>
Finally, the optical system uses an optical element produced by molding using a correction mold (first molding step; STEP 20) and molding using a mold (mold other than the correction mold) that is not corrected (first mold). (2) Molding step: The optical element produced by STEP 21) is completed by assembling (STEP 22).

[3. Summary]
The manufacturing method described above is manufactured while manufacturing each optical element (initial product) in the projection optical system PS having a plurality of optical elements by molding an optical element material with a mold (mold member). An approximate surface (initial surface data) of each optical surface is set by determining an approximate expression based on the measurement result of the surface shape of the optical surface in each optical element.

  In addition, the manufacturing method evaluates the optical performance of the entire system from the approximate surfaces of all the optical surfaces in the projection optical system PS, and further, at least one of the approximate surfaces of the entire optical surfaces {for example, the initial product By changing the reduction-side lens surface S (L2s) of the second lens L2, which is referred to as “change approximate surface”}, the amount of change of the change approximate surface when the optimum optical performance as the entire system is exhibited is obtained. (The approximate surface that is not changed is referred to as a non-changeable approximate surface). Then, based on the amount of change, a correction processing amount for the cavity surface of the mold corresponding to the change approximate surface is obtained and corrected to produce a new cavity surface.

  This manufacturing method does not correct all the cavity surfaces in all molds. For example, such a manufacturing method can preferentially correct the cavity surface corresponding to the maximum shape error surface. Therefore, the number of mold corrections is relatively small, and the projection optical system PS is manufactured in a short time and at a low cost while being efficient.

  Furthermore, this manufacturing method performs the correction | amendment process of a metal mold | die based on the optical performance evaluation as projection optical system PS. That is, the cavity surface is not corrected to form an optical surface having the same data as the design data F of the optical surface, but a specific optical surface is used to exhibit optimum optical performance as the projection optical system PS. The cavity surface {for example, the cavity surface T (L2s) of the fifth mold} corresponding to is corrected.

  Therefore, even if there is a shape error (an error with the design data F) on other optical surfaces other than the specific optical surface {for example, the reduction-side lens surface S (L2s) of the second lens L2 of the initial product) There is no problem if the optical surface has a surface shape that exhibits optimum optical performance as the projection optical system PS. Therefore, it can be said that such a manufacturing method is hardly affected by the shape error of each optical surface in the projection optical system PS. Because, when the cavity surface is corrected to form an optical surface having the same data as the design data F of the optical surface, all the cavity surfaces are accurately corrected to optimize the optical performance of the projection optical system PS. This is because it must be processed.

  However, the first correction processing (first correction processing; production of a new cavity surface) to the cavity surface corresponding to the specific optical surface may not be performed with high accuracy. In this case, projection optics including an optical element (for example, a second lens L2 as a remolded product) molded with a mold having a new cavity surface and an optical element (initial product) molded with an existing mold. The optical performance of the system PS is outside the allowable range.

  Therefore, such a manufacturing method measures the surface shape of an optical surface corresponding to a new cavity surface in an optical element (for example, the second lens L2 which is a remolded product) molded with a mold having a new cavity surface. An approximate surface is set by determining an approximate expression based on the result. In addition, this manufacturing method uses a shape error between the approximate surface of the optical surface corresponding to the new cavity surface and the new cavity surface (for example, reshaped product surface data G ′ [S (L2s)] and new cavity surface data. The new cavity surface is corrected so as to cancel out the error with respect to D).

  Such correction processing [second correction processing] is performed to eliminate the shape error, but is performed only on the cavity surface (new cavity surface) that has already been corrected. For this reason, the second correction processing has only a light burden as compared with the burden in the case of correcting all the cavity surfaces.

  However, if the second correction process (second correction process) to the cavity surface corresponding to the specific optical surface is performed with high accuracy, an optical molded with a mold having a cavity surface corrected multiple times. The optical performance of the projection optical system PS including an element (for example, the second lens L2 as a follow-up product) and an optical element (initial product) molded with an existing mold is within an allowable range. Then, since this lightly-burden second correction process exists, the manufacturing method can easily manufacture the projection optical system PS that exhibits high optical performance.

  Note that the manufacturing method described above is particularly effective for a manufacturing method of a reflective optical system including a plurality of eccentric surfaces, particularly an optical system including a free-form surface. Usually, an error during molding is likely to cause an asymmetric error (for example, an as component) with respect to a design value. For this reason, if a non-rotationally symmetric surface (especially a free-form surface) is not a rotationally symmetric surface, a forming error can be easily corrected. The eccentric surface means a surface that does not have a rotationally symmetric axis, or a surface that has a symmetric axis that is largely deviated from the center of the effective surface even if it has one.

  By the way, in the optical performance evaluation, an approximate surface (initial surface data, reshaped product surface data, driven product surface data, etc.) of the optical surface of the optical element in the projection optical system PS is used. Specifically, the optical performance evaluation is performed by a ray tracing simulation using power obtained from the surface shape of the approximate surface. Therefore, how to set the approximate surface becomes important.

  For example, when local waviness (waviness portion) exists on the measured optical surface, if an approximation expression of higher order is used to appropriately represent the waviness portion, a place other than waviness (non-waviness portion) High-order undulation will occur. On the other hand, if a low-order approximation expression is used to appropriately represent the non-swelled part, the swelled part cannot be represented properly.

  Therefore, the approximate surface of the optical surface (initial surface data, reshaped surface data, driven surface data, etc.) is represented by a function that can efficiently represent the waviness portion of the optical surface, for example, a spline function. And desirable. As an example, a plane (YZ plane; see FIG. 3) perpendicular to a predetermined reference axis (for example, the X axis) on the optical surface of the optical element is divided into a plurality, and these divided planes are defined as bottoms. A plurality of spaces (divided spaces) to be set are set, and a spline function having continuity at the boundary between the divided spaces is exemplified.

  However, it is desirable that the order of the spline function or the like is at least the third order (in the above description, the fifth order spline functions are listed and described). With such a cubic or higher-order spline function, an approximate surface corresponding to each divided space can be set using the measurement point MP in the divided space. In particular, when a function of at least a third order (for example, a spline function) is used, the second derivative of the approximate expression is continuous at the boundary between the divided spaces, and a step or the like is generated on the approximate surface at the boundary of the divided space. First, ray tracing simulation is possible because the power of the approximate surface is continuous.

  Desirably, the order of the spline function or the like is 4 to 8th. This is because a local undulation portion that greatly affects the optical performance of the projector field is expressed by a 4th to 8th order function. In addition, by creating an approximate surface with a 4th to 8th order function, it is possible to express a local undulation shape that greatly affects the optical performance, and further removes higher-order undulation portions that do not easily affect the optical performance. It can be said that it also functions as a filter function.

[Embodiment 2]
The present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention.

  For example, the optical performance caused by the reduction side lens surface S (L2s) and the enlargement side lens surface S (L2e) in the second lens L2 is changed by the shape change of the reduction side lens surface S (L2s) of the second lens L2. Although an example of correction is given, it is not limited to this. For example, the optical performance due to the enlargement side lens surface S (L2e) of the second lens L2 is changed by the shape change of the reduction side lens surface S (L2s) of the second lens L2 or the reflection surface S (M3) of the third mirror. It may be corrected.

  In short, in order to optimize the optical performance as the projection optical system PS, the mold may be corrected so that the shape of at least one of the optical surfaces included in the projection optical system PS changes. However, if the image height of light passing through an optical surface that affects the optical performance and a shape change in the optical surface through which light having an image height approximately equal to that of the light pass, the optical performance is effectively improved. Therefore, for example, it can be said that it is desirable to correct the optical performance caused by the enlargement side lens surface S (L2e) of the second lens L2 by the shape change of the reduction side lens surface S (L2s) of the second lens L2.

  Further, when setting a divided space (in the case of a space dividing step), the relationship between the number of divided spaces and the number of measurement points in the divided space becomes important. For example, if the number of measurement data MP is relatively large even though the number of divided spaces is relatively small, the shape error is almost equivalent to expressing the shape error by an error polynomial for the entire surface. Then, local waviness is not expressed on the approximate surface.

  On the other hand, for example, when the number of measurement data MP is relatively small although the number of divided spaces is relatively large, the local undulation of the optical surface is approximated, but the measurement data MP in the direction perpendicular to the measurement direction ( In FIG. 3, the number of measurement data MP) in the Y-axis direction perpendicular to the Z-axis direction is reduced in the divided space. As a result, the approximation accuracy in the direction of the reduced measurement data MP (the Y-axis direction in FIG. 3) decreases.

  In view of the above, it would be desirable to have a relatively large number of divided spaces and a relatively large number of measurement data MP. However, in such a case, it takes a very long time to measure the surface shape, resulting in the temperature drift of the measuring instrument due to changes in environmental temperature and humidity, and the change in the optical surface of the initial product over time. Decreases. Moreover, since the number of points of the measurement data MP is relatively large, the measurement efficiency is also lowered.

  Therefore, a divided space that can improve the measurement efficiency while ensuring the approximate accuracy of local waviness of the optical surface and the approximate accuracy of the optical surface due to the reduction of the measurement points MP in the divided space. And the number of measurement points in the divided space are desirable. As an example, there is a measurement method as shown in FIG. 3 in which the number of divided spaces is 25 and five lines are measured for each divided space.

  When creating an approximate surface, the greater the number of measurement lines for each divided space, the better the approximation accuracy. However, for example, in the case where an approximate surface is created with a quintic function that affects the optical performance of the projector field, in order to create an approximate surface with high accuracy, 5 lines are normally desirable for each divided space. If there are at least three lines, an approximate surface can be created with high accuracy.

  In FIG. 3, line measurement is performed along the Z-axis direction, but line measurement may be performed along the Y-axis direction. Further, line measurement (matrix measurement) along both the Y-axis direction and the Z-axis direction may be performed. However, although the measurement efficiency of the matrix is reduced, the measurement efficiency is improved as compared with the measurement of the matrix if the line measurement is performed in one direction (Y-axis direction or Z-axis direction).

  Further, the polynomial representing the design data F and the initial cavity surface data F is not particularly limited, but may be a spline function.

  Moreover, not all optical elements may be molded articles. For example, the first mirror M1 that is a spherical surface may be a polished product. Further, glass molding may be used instead of resin molding. Further, it is not necessary to obtain approximate surfaces for all optical elements and perform optical performance evaluation using the approximate surfaces of all optical elements. For example, when the first mirror M1 is a glass polished product, a shape almost as designed is obtained, so design data may be used as the approximate surface of the first mirror M1.

[Embodiment 3]
A third embodiment will be described. In addition, about the member which has the same function as the member used in Embodiment 1 * 2, the same code | symbol is attached and the description is abbreviate | omitted.

  Examples of materials for optical elements such as mirrors and lenses include various materials (glass, resin, etc.). However, the temperature required for molding a glass optical element is higher than the temperature required for molding a resin optical element. Due to the high temperature, a glass optical element (particularly a glass optical element having a relatively large external size) is used. ) Is likely to be lower than the molding accuracy of the resin optical element. Therefore, the glass optical element is difficult to be formed into a surface shape as designed.

  In general, when the effective luminous flux width of light reaching the surfaces of the optical elements (for example, the mirror surface and the lens surface) is substantially the same, if the error amount of the surface shape of each surface is the same, The optical performance deterioration due to the surface shape error is larger than the optical performance deterioration due to the surface shape error of the lens surface (that is, the mirror surface is higher than the lens surface in terms of sensitivity to the optical performance). Therefore, in the projection optical system PS in which the glass optical element and the resin optical element are mixed, when the glass optical element is a mirror, an error in the surface shape of the mirror is likely to occur, and the projection optical system PS The optical performance is greatly affected.

  Examples of such optical performance include the spot diagrams of FIGS. These spot diagrams show a projection optical system PS including a first mirror M1 and a second mirror M2 made of glass, and a first lens L1, a second lens L2, a third mirror M3, and a fourth mirror M4 made of resin. It is calculated from the initial surface data. However, the deterioration of the optical performance shown in these spot diagrams is caused by the fact that the reflecting surface S (M2) of the second mirror M2 is not formed into a desired surface shape.

Note that the spot diagrams in these figures show imaging characteristics by overlapping spot diagrams of three wavelengths (460 nm, 546 nm, and 620 nm) at 45 evaluation points on the screen surface SCN (scale is ± 1 mm). Notation). 6 shows only the half on the positive side in the Z-axis direction on the screen surface SCN, FIG. 7 shows the negative side in the Z-axis direction, which is the other half, and the coordinates (Y, Z) in FIG. And the expression is the same as in FIG. 5 ( ^{note that} “e ⁻ⁿ ” in the figure is “10 ⁻ⁿ ”).

  In order to prevent optical performance deterioration as shown in FIGS. 6 and 7, the manufacturing method described above may be used. That is, an optical element material (glass or resin) is molded with a mold to manufacture each optical element (initial product) in the projection optical system PS having a plurality of optical elements, and the optical in each manufactured optical element. The approximate surface (initial surface data) of each optical surface is set by determining an approximate expression based on the measurement result of the surface shape of the surface.

  Furthermore, the optical performance of the entire system is evaluated from the approximate surfaces of all the optical surfaces, and at least one of the approximate surfaces of all the optical surfaces (change approximate surface) is changed, so that the optimal optical performance as the entire system is achieved. The amount of change of the approximated surface is calculated when performing the process, and based on the amount of change, the correction processing amount for the cavity surface of the mold corresponding to the changed approximate surface is obtained, and the first correction processing is performed to produce a new cavity surface. do it.

  However, in the projection optical system PS including the mirror and the lens, it is preferable to use at least one lens surface as the change approximate surface. After the first correction processing is performed based on the change amount H of the initial surface data, when re-molding is performed with the new cavity surface data D as a target value, the new cavity surface data D and the re-molded product surface data If a shape error occurs with G ′, the optical performance deteriorates. However, the sensitivity of the lens surface is lower than that of the mirror surface (mirror reflection surface), and therefore, if the change approximate surface is a lens surface, the influence on the optical performance is reduced.

  Therefore, the initial surface data (G [S () is obtained by using the reduction-side lens surface S (L1s) of the first lens L1 and the reduction-side lens surface S (L2s) of the second lens L2 in the projection optical system PS as change approximate surfaces. L1s)], G [S (L2s)]) were determined. Further, based on the amount of change, a correction machining amount for the mold cavity surface {second mold cavity surface T (L1s), fifth mold cavity surface T (L2s)} corresponding to the change approximate surface is obtained. The first correction processing was performed to produce a new cavity surface.

  Then, after re-molding the new first lens L1 and the second lens L2 with the second mold and the fifth mold subjected to the first correction processing, the reduction-side lens surface S of the new first lens L1. (L1s) and the surface shape of the reduction lens surface S (L2s) of the second lens L2 are measured, and further, approximate surface data (reformed product surface data; G ′ [S (L1s)) ], G ′ [S (L2s)]).

  8 and 9 show the reshaped product surface data (G ′ [S (L1s)], G ′ [S (L2s)]) and the initial surface data (G [S (M1)], G [S (L1e). )], G [S (M2)], G [S (L2e)], G [S (M3)], G [S (M4)]) (refer to FIG. 8). And FIG. 9 has the same representation as FIG. 6 and FIG.

  That is, these spot diagrams include the first mirror M1, the second mirror M2, the third mirror M3, and the fourth mirror M4 that are initial products, and the first lens L1 and the second lens L2 that are remolded products. The optical performance evaluation of the projection optical system PS including it is shown. These optical performance evaluation results could be determined to be within the allowable range.

  In particular, the second mirror M2 in the projection optical system PS is a glass-made rotationally symmetric aspherical mirror. Usually, when a rotationally symmetric aspherical mirror is to be formed by glass molding, a rotationally symmetric mirror surface is hardly formed due to a relatively low molding accuracy. That is, an error in the asymmetric surface shape is likely to occur on the mirror surface (from the viewpoint of the asymmetric surface shape error, FIGS. 6 to 9 show the positive side and the negative side in the Z-axis direction on the screen surface SCN. ).

  However, it is difficult to perform free-form correction processing on the cavity surface of the mold corresponding to the mirror surface. This is because the correction processing for the cavity surface of the glass molding die for high temperature is more difficult than the correction processing for the cavity surface of the resin molding die.

  Then, the optical performance deterioration caused by the surface shape error on the reflection surface S (M2) of the second mirror M2 is caused by the cavity surface T (M2) of the fourth die (glass molding die) corresponding to the second mirror M2. ), The cavity surfaces {T (L1s), T (L2s) of the second mold and the fifth mold (resin molding mold) corresponding to the first lens L1 and the second lens L2 made of resin. )} Is easy to correct.

  In addition, since it is easy to correct the free curved surface shape to the cavity surfaces {T (L1s), T (L2s)} of the second mold and the fifth mold, the reflecting surface S (M2) of the second mirror M2 It is easy to correct (cancel) the error of the asymmetrical surface shape generated in ().

  From the above, in the projection optical system PS including at least one mirror and at least one lens, it can be said that at least one lens surface is a change approximate surface from the viewpoint of sensitivity to optical performance. In particular, the surface shape error of a glass molded mirror, more specifically, the optical performance deterioration caused by the surface shape error generated on the reflection surface (mirror surface) which is a rotationally symmetric aspherical surface is the case where the lens surface is a change approximate surface It can be said that it is easily corrected.

  However, the present invention is not limited to this, and the optical performance deterioration due to the surface shape error of the lens surface of the glass molded lens, more specifically, the surface shape error generated on the lens surface which is a rotationally symmetric aspheric surface is also caused by the lens surface (desirably Can be said to be easily corrected when the lens surface of the resin lens is used as a change approximate surface. This is because the lens surface is less sensitive than the mirror surface, so that even if a reshape product has a shape error, the influence on the optical performance is small.

  Finally, it goes without saying that embodiments obtained by appropriately combining the techniques disclosed above are also included in the technical scope of the present invention.

Claims (21)

In a method of manufacturing an optical system including a plurality of optical surfaces by molding using a mold member,
Setting an approximate surface of all optical surfaces of the optical system including the optical surface formed by initial molding;
A first optical performance evaluation step for evaluating the optical performance of the entire system from the approximate surface of the entire optical surface in the optical system;
Of the approximate surfaces of all the optical surfaces, at least one surface molded by the mold member is a change approximate surface, and among the approximate surfaces of all the optical surfaces, at least one surface molded by the mold member is the change approximate surface. A change amount calculating step for obtaining a change amount of the change approximate surface when the optimal optical performance as the entire system is exhibited,
A first correction processing step of determining a correction processing amount to the cavity surface of the mold member corresponding to the change approximate surface based on the change amount, performing correction processing, and producing a new cavity surface;
A first molding step of molding an optical element using the correction mold member processed in the first correction processing step;
A second molding step of molding an optical element using a mold member other than the correction mold member;
The manufacturing method of the optical system containing this. The first molding step is as follows.
A step of setting an approximate surface based on the measurement result of the surface shape of the optical surface corresponding to the new cavity surface in the optical element molded by the mold member having the new cavity surface;
A second correction processing step for correcting the new cavity surface so as to offset the shape error between the approximate surface of the optical surface corresponding to the new cavity surface and the new cavity surface;
Forming an optical element using the correction mold member processed in the second correction processing step;
The manufacturing method of the optical system of Claim 1 containing this. In the step of setting the approximate surface,
The optical system manufacturing method according to claim 1, wherein an approximate surface based on a measurement surface shape of the optical surface formed by the initial molding is set on the optical surface formed by molding using the mold member. In the step of setting the approximate surface,
2. The method of manufacturing an optical system according to claim 1, wherein when the optical surface is a polished surface, design data of the optical surface is set as an approximate surface on the polished surface.   The method of manufacturing an optical system according to claim 1, wherein an approximate surface having the largest difference between the approximate surface of each optical surface in the optical system and the design data of the optical surface corresponding to each approximate surface is a change approximate surface.   The approximate surface of the optical surface closest to the optical surface having the largest difference between the approximate surface of each optical surface in the optical system and the design data of the optical surface corresponding to each approximate surface is defined as a change approximate surface. The manufacturing method of the optical system.   The method of manufacturing an optical system according to claim 1, wherein the change approximate surface is at most two surfaces. In the step of setting the approximate surface,
A space division setting step is included in which a plane perpendicular to a predetermined reference axis on the optical surface of the optical element is divided into a plurality of spaces, and a plurality of spaces having the divided planes as a bottom are set. The method of manufacturing an optical system according to claim 1, wherein an approximate expression having continuity at a boundary is used.   The method of manufacturing an optical system according to claim 8, wherein the approximate expression is a function of at least a third order.   The optical system manufacturing method according to claim 9, wherein the continuity is that the second derivative of the approximate expression is continuous at a boundary between the spaces.   The optical system manufacturing method according to claim 8, wherein the approximate expression is a spline function. The optical system includes at least one lens and at least one mirror as an element molded by a mold member,
The method for manufacturing an optical system according to claim 1, wherein at least a lens surface is used as the change approximate surface.   The method for manufacturing an optical system according to claim 12, wherein at least one of the lens and the mirror included in the optical system is formed by glass molding. The optical system includes a non-rotationally symmetric surface,
The method of manufacturing an optical system according to claim 1, wherein in the change amount calculating step, an approximate surface corresponding to the non-rotationally symmetric surface is a change approximate surface. The optical system includes at least one lens and at least one mirror as an element molded by a mold member,
The method for manufacturing an optical system according to claim 14, wherein at least a lens surface is used as the change approximate surface.   The method for manufacturing an optical system according to claim 15, wherein at least one of the lens and the mirror included in the optical system is formed by glass molding.   In a method for manufacturing a mold member for molding an optical system including a plurality of optical surfaces,
  Setting an approximate surface of all optical surfaces of the optical system including the optical surface formed by initial molding;
  A first optical performance evaluation step for evaluating the optical performance of the entire system from the approximate surface of the entire optical surface in the optical system;
  Of the approximate surfaces of all the optical surfaces, at least one surface molded by the mold member is a change approximate surface, and among the approximate surfaces of all the optical surfaces, at least one surface molded by the mold member is the change approximate surface. A change amount calculating step for obtaining a change amount of the change approximate surface when the optimal optical performance as the entire system is exhibited,
  A first correction processing step of determining a correction processing amount to the cavity surface of the mold member corresponding to the change approximate surface based on the change amount, performing correction processing, and producing a new cavity surface;
The manufacturing method of the mold member containing this.   In the step of setting the approximate surface,
  The method for manufacturing a mold member according to claim 17, wherein an approximate surface based on a measurement surface shape of the optical surface formed by the initial molding is set on the optical surface formed by molding using the mold member.   In the step of setting the approximate surface,
  18. The method for manufacturing a mold member according to claim 17, wherein when the optical surface is a polished surface, design data of the optical surface is set as an approximate surface on the polished surface.   18. The method for manufacturing a mold member according to claim 17, wherein an approximate surface having the largest difference between the approximate surface of each optical surface in the optical system and the design data of the optical surface corresponding to each approximate surface is a change approximate surface.   The optical system includes a non-rotationally symmetric surface,
  The mold member manufacturing method according to claim 17, wherein in the change amount calculation step, an approximate surface corresponding to the non-rotationally symmetric surface is a change approximate surface.

Images