JP2010140038A - Method for manufacturing optical element - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method of manufacturing an optical device in which the beam shift precision in each optical device is easily secured by devising a means of the laminating means of respective parallel plates having the variation of the thickness to suppress the misalignment of the contact surface of the finally resultant optical device. <P>SOLUTION: A plurality of parallel plates are laminated and cut and then the cut laminate bodies are piled and cut to form a plurality of optical elements. In the laminating step of the parallel plates, the parallel plate having a thickness of a reference value or more and the parallel plate having a thickness below the reference value are selected from among a plurality of ranks and laminated such that the misalignment of the contact surface in each cutting surface falls within a prescribed range in every cutting position. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to an optical element manufacturing method for manufacturing an optical element formed by laminating a plurality of transparent media by repeatedly laminating and cutting a plurality of transparent parallel plates.

  Conventionally, as a manufacturing method of a prism used for an optical pickup, for example, there is a manufacturing method disclosed in Patent Document 1. This manufacturing method will be briefly described with reference to FIGS. 2 (a) to 2 (e) showing manufacturing steps in the manufacturing method of the present invention. First, as shown in FIG. 2 (a), an optical thin film for beam splitting is used. A plurality of parallel flat plates 11 subjected to the above are laminated stepwise using an optical adhesive (for example, an ultraviolet curable adhesive) to form a laminate 1. Then, the laminate 1 is cut at an angle of 45 degrees. Thereby, as shown in FIG.2 (b), the some lamination | stacking division body 12 is formed.

  Then, as shown in FIG.2 (c), the several laminated division body 12 is piled up again again, and is fixed with temporary joining or an auxiliary board, etc., and these are cut | disconnected by the orthogonal | vertical direction. Then, by releasing the temporary joint portion, as shown in FIG. 2 (d), a plurality of rod-like bodies 13 formed by bonding two transparent media are formed. Finally, as shown in FIG. 2E, the individual cube-shaped prisms 14 are obtained by cutting each rod-like body 13 along a plane perpendicular to the longitudinal direction. The release of the temporary joint portion may be performed at another stage (for example, after cutting in FIG. 2 (e)).

  According to this manufacturing method, in addition to using a relatively inexpensive plate-like glass material (parallel plate 11), a large amount of prisms 14 can be obtained through a series of processing steps. That is, the prism 14 can be mass-produced at a low cost.

JP 2000-143264 A

  By the way, although the thickness of each parallel plate 11 usually varies, in the manufacturing method of Patent Document 1, each parallel plate 11 is not considered at all in the process of FIG. Are stacked. In the step of FIG. 2 (c), a stack of a plurality of laminated divided bodies 12 is cut with, for example, a wire saw, but the wire pitch of the wire saw is constant, and can be easily changed and adjusted. Can not do it. For this reason, if the thickness of each parallel plate 11 is biased when the plurality of parallel plates 11 in FIG. 2A are laminated, two transparent media (parallel) are cut at the time of cutting in the step of FIG. The positions of the joining surfaces 21 of the flat plates 11 and 11) are shifted at both ends of the wire saw.

  For example, if each parallel plate 11 is biased to a thin one, when the laminated divided body 12 is cut at a predetermined pitch with a wire aligned with the cutting position at the rightmost end, the maximum length shown in FIG. With the rightmost prism 14 and the leftmost prism 14 shown in FIG. 23 (a), the positions of the bonding surfaces 21 are shifted as shown in FIGS. 23 (a) and 23 (b). On the other hand, when each parallel plate 11 is biased toward a thicker one, when the laminated divided body 12 is cut at a predetermined pitch with the wire aligned with the cutting position on the rightmost end, it is shown in FIG. With the rightmost prism 14 and the leftmost prism 14 shown in FIG. 23 (c), the positions of the joint surfaces 21 are shifted as shown in FIGS. 23 (c) and 23 (d). As a result, when the prism 14 is applied to an optical pickup, as shown in FIG. 24, depending on the prism 14, the light incident on the incident surface center of the prism 14 is reflected by the optical thin film formed on the bonding surface 21. When emitted, a beam shift occurs in which the position of the emitted light is shifted from the center of the emission surface.

  That is, in the manufacturing method of Patent Document 1 described above, each parallel plate 11 is laminated and cut without considering the variation in thickness. Therefore, the thickness is biased only to the thicker one or the thinner one. There is a possibility that the parallel plates 11 are stacked and cut. In that case, a positional shift of the joint surface occurs in each prism 14 finally obtained, and it becomes difficult to ensure the beam shift accuracy in each prism 14.

  The present invention has been made in order to solve the above-described problems, and an object of the present invention is to finally obtain an optical element by devising a method of laminating parallel plates having variations in thickness. It is an object of the present invention to provide a method of manufacturing an optical element capable of easily ensuring the beam shift accuracy in each optical element while suppressing the positional deviation of the joint surface.

  The method for manufacturing an optical element of the present invention includes a first step of forming a laminate by laminating parallel plates made of a plurality of transparent media via an adhesive, a predetermined inclination angle, and a predetermined inclination. Cutting the laminated body at a pitch of the second step of forming a plurality of laminated divided bodies, and stacking the plurality of laminated divided bodies so as to be perpendicular to the cut surface in the second step and in parallel A third step of forming a plurality of optical elements formed by laminating transparent media by cutting at a predetermined pitch in a direction in which the flat plates are laminated, In the step, any one of the cut surfaces in the third step is set at a predetermined pitch with reference to a cutting reference position determined so as to satisfy a predetermined positional relationship with respect to the joint surface that is a boundary of the parallel plates. Cutting the laminated segment After the previous step of dividing the plurality of parallel flat plates into a plurality of ranks according to the difference between the thickness of the parallel flat plate and a reference value of the thickness, in the first step, each cutting in the third step The parallel plate belonging to a rank having a thickness that is greater than or equal to the reference value or greater than the reference value and the thickness is the reference value so that the positional deviation of the joint surface with respect to the surface is within a predetermined range at any cutting position. The parallel plates belonging to a rank less than the value or below the reference value are selected and laminated.

  In the optical element manufacturing method of the present invention, each rank is associated with an integer value that monotonously decreases or increases in descending order of the maximum value or the minimum value of the parallel plate thickness allowed by each rank. And the integer value of the rank to which the parallel plate whose thickness is greater than or equal to the reference value and the rank to which the parallel plate whose thickness is less than or less than the reference value belongs is opposite to each other. In the first step, the parallel values from the respective ranks are set such that the accumulated value of the integer value corresponding to each parallel plate to be stacked is within a predetermined range regardless of the accumulated value up to any parallel plate. A flat plate may be selected and laminated.

  In the optical element manufacturing method of the present invention, each rank is associated with an integer value that monotonously decreases or increases in descending order of the maximum value or the minimum value of the parallel plate thickness allowed by each rank. And the integer value of the rank to which the parallel plate whose thickness is greater than or equal to the reference value and the rank to which the parallel plate whose thickness is less than or less than the reference value belongs is opposite to each other. In the first step, the average value of the integer values corresponding to the ranks to which the predetermined number of parallel flat plates that are successively stacked belongs is the average value of the predetermined number of parallel flat plates that are continuously stacked. You may make it laminate | stack by selecting a parallel plate from each rank so that it may be settled in a predetermined range for every different combination.

  In the method for manufacturing an optical element of the present invention, among the plurality of ranks to which the parallel plate having a thickness equal to or greater than the reference value or larger than the reference value belongs, the rank with the smallest minimum value of the allowed parallel plate thickness is: While being associated with the first integer value (for example, 1), each of the other ranks is the same as the first integer value in order from the smallest of the minimum parallel plate thicknesses allowed by each rank. It is associated with an integer value (for example, 2, 3,...) That has a sign and whose absolute value increases by a predetermined value from the absolute value of the first integer value, and the thickness is less than the reference value. Or, among the ranks to which the parallel plates below the reference value belong, the rank having the largest minimum value of the allowed parallel plate thickness is the second integer value having a different absolute value from the first integer value (for example, -1), while other ranks Is the same sign as the second integer value, and the absolute value increases by a predetermined value from the absolute value of the second integer value, in order from the largest of the minimum values of the parallel plate thickness allowed by each rank. It may be associated with such an integer value (for example, -2, -3,...).

  In the optical element manufacturing method of the present invention, each rank is associated with an integer value that monotonously decreases or increases in descending order of the maximum value or the minimum value of the parallel plate thickness allowed by each rank. In addition, the above-mentioned integer value is 0 for a rank to which a parallel plate whose thickness falls within a predetermined range from the reference value, and is a rank outside the rank, and the thickness is larger than the reference value. When the rank to which the parallel plate belongs and the rank to which the parallel plate whose thickness is smaller than the reference value have integer values opposite to each other, the first step corresponds to each parallel plate to be laminated. The parallel plate may be selected from each rank and stacked so that the accumulated value of the integer values to be within the predetermined range is the accumulated value up to any parallel plate.

  In the optical element manufacturing method of the present invention, each rank is associated with an integer value that monotonously decreases or increases in descending order of the maximum value or the minimum value of the parallel plate thickness allowed by each rank. In addition, the above-mentioned integer value is 0 for a rank to which a parallel plate whose thickness falls within a predetermined range from the reference value, and is a rank outside the rank, and the thickness is larger than the reference value. When the rank to which the parallel plate belongs and the rank to which the parallel plate whose thickness is smaller than the reference value have integer values opposite to each other, in the first step, the predetermined number of sheets that are continuously stacked The parallel plate is selected from each rank so that the average value of the integer values corresponding to each rank to which the parallel plate belongs is within a predetermined range for each different combination of a predetermined number of parallel plates stacked successively. It may be a layer.

  In the optical element manufacturing method of the present invention, an allowable parallel plate thickness among a plurality of ranks to which a parallel plate having a thickness greater than a reference value belongs and which is outside the rank of the integer value 0. The rank with the smallest minimum value is associated with the first integer value (for example, 1), while each of the other ranks has the smaller minimum value of the parallel plate thickness allowed by each rank. In order from the first integer value, it is associated with an integer value (for example, 2, 3,...) That has the same sign as the first integer value and whose absolute value increases by a predetermined value from the absolute value of the first integer value. Among the multiple ranks to which the parallel plate having a thickness smaller than the reference value belongs, the smallest value of the allowable parallel plate thickness is the highest. A higher rank is a second integer that differs in absolute value from the first integer value. While each of the other ranks is associated with a value (for example, -1), each of the other ranks has the same sign as the second integer value in order from the largest minimum value of the parallel plate thickness allowed by each rank, In addition, the absolute value may be associated with an integer value (for example, −2, −3,...) Such that the absolute value increases by a predetermined value from the absolute value of the second integer value.

  In the optical element manufacturing method of the present invention, in the first step, parallel flat plates belonging to ranks having different signs of integer values may be alternately laminated.

  In the method for producing an optical element of the present invention, in the first step, parallel flat plates belonging to the same rank of absolute values of integer values may be alternately laminated.

  In the method of manufacturing an optical element of the present invention, when the number of parallel plates to be laminated in the first step is an even number, in the first step, the parallel plate whose integer value sign belongs to a positive rank, The same number of parallel flat plates belonging to negative ranks may be stacked.

  In the method of manufacturing an optical element of the present invention, when the number of parallel plates to be laminated in the first step is an odd number, in the first step, in the first step, the parallel plates whose integer value sign belongs to a positive rank, The same number of parallel flat plates belonging to negative ranks of numerical values may be stacked, and an odd number of parallel flat plates belonging to ranks having an integer value of 0 may be stacked.

  In the method of manufacturing an optical element of the present invention, the cutting reference position when cutting the plurality of laminated division bodies at a predetermined pitch in the third step is the center in the direction of taking the cutting pitch in each laminated division body or substantially the same. The center position is desirable.

  The optical element manufacturing method of the present invention may further include a fourth step of forming an optical thin film on the surface of each parallel plate. However, the fourth step is performed before the first step.

  The method for producing an optical element of the present invention may further include a fifth step of polishing at least one of the surfaces that have become cut surfaces at the time of cutting in the second step, in each of the laminated divided bodies. However, the fifth step is performed before the third step.

  In the method of manufacturing an optical element of the present invention, in the third step, after stacking a plurality of laminated division bodies and cutting them at a predetermined pitch, at least one of the surfaces that became the cut surfaces at the time of cutting is polished. It may be.

  According to the present invention, in the first step, no matter what value n can take (for example, when m = 11, n takes any value such as 1 to 10 or 2 to 9). The thickness is greater than or equal to the reference value or the reference value so that the displacement of the joint surface with respect to each cut surface in the step is within a predetermined range (for example, within the displacement standard of the joint surface) at any cutting position. A parallel flat plate belonging to a higher rank and a parallel flat plate belonging to a rank having a thickness less than or equal to the reference value are selected from the parallel flat plates divided into a plurality of ranks and stacked. For example, the former parallel plate and the latter parallel plate are alternately laminated, or a predetermined number of the former parallel plates are continuously laminated, and the latter parallel plate is formed thereon. For example, a predetermined number of sheets are stacked.

  By roughly laminating and laminating two types of parallel flat plates from a plurality of ranks in this way, in the third step, the parts to be cut of each stacked divided body are respectively determined at each cutting position of a predetermined pitch. Each stacked divided body can be cut in a state close to. As a result, for each optical element obtained by cutting at a predetermined pitch, it is possible to suppress the displacement of the bonding surface.

  At this time, when the optical element manufactured by the manufacturing method of the present invention is applied to, for example, an optical pickup, an optical thin film (for example, a polarization separation film) is formed on the bonding surface of one transparent medium. The accuracy directly affects the accuracy of the beam shift in which the light incident on the optical element and reflected by the optical thin film is shifted in the direction perpendicular to the traveling direction (optical axis direction). Therefore, according to the above-described manufacturing method of the present invention that can suppress the displacement of the bonding surface, the beam shift accuracy can be easily ensured for each optical element to be obtained.

It is a flowchart which shows the flow of each process in the manufacturing method of the optical element which concerns on one Embodiment of this invention. (A)-(e) is explanatory drawing which shows the manufacturing process in the said manufacturing method. It is explanatory drawing which shows an example of the rank division | segmentation of several parallel flat plates, and the matching of the integer value to each rank. It is explanatory drawing which shows the relationship between the position of the joint surface of two transparent media, and the cutting position by a wire. It is explanatory drawing which shows the simulation result of Example 1. FIG. It is explanatory drawing which shows the simulation result of Example 2. FIG. It is explanatory drawing which shows the simulation result of Example 3. FIG. It is explanatory drawing which shows the simulation result of Example 4. FIG. It is explanatory drawing which shows the simulation result of Example 5. It is explanatory drawing which shows the other example of rank division | segmentation of a some parallel flat plate, and matching of the integer value to each rank. It is explanatory drawing which shows the simulation result of Example 6. FIG. It is explanatory drawing which shows the simulation result of Example 7. FIG. It is explanatory drawing which shows the simulation result of Example 8. FIG. It is explanatory drawing which shows the simulation result of Example 9. FIG. It is explanatory drawing which shows the simulation result of Example 10. FIG. It is explanatory drawing which shows the simulation result of Example 11. FIG. It is explanatory drawing which shows the simulation result of Example 12. FIG. It is explanatory drawing which shows the simulation result of Example 13. FIG. It is explanatory drawing which shows the further another example of the ranking of several parallel flat plates, and the matching of the integer value to each rank. It is explanatory drawing which shows the further another example of the ranking of several parallel flat plates, and the matching of the integer value to each rank. It is explanatory drawing which shows the further another example of the ranking of several parallel flat plates, and the matching of the integer value to each rank. It is explanatory drawing which shows the further another example of the ranking of several parallel flat plates, and the matching of the integer value to each rank. (A), (b) and (c), (d) is explanatory drawing which respectively shows the example of the position shift of the joint surface in a prism. It is explanatory drawing which shows typically a mode that a beam shift occurs by the position shift of a joint surface.

  An embodiment of the present invention will be described below with reference to the drawings.

(1. Manufacturing method flow)
First, a general flow of the optical element manufacturing method of the present invention will be described with reference to FIG. 1 and FIGS. 2 (a) to 2 (e).

  First, an optical thin film is formed on the surface of a parallel plate 11 made of a transparent medium (for example, glass) shown in FIG. 2A (S1: thin film forming step, fourth step). This optical thin film can be, for example, a polarization separation film that transmits or reflects incident light according to its polarization state. Next, a plurality of parallel flat plates 11 are laminated stepwise via a first translucent adhesive (for example, an ultraviolet curable adhesive) to form a laminate 1 (S2: laminating step, first Step 1). Details of the stacking process of S2 will be described later.

  Next, the laminated body 1 is cut at a predetermined inclination angle (for example, 45 degrees) at a predetermined pitch with respect to the surface on which the optical thin film is formed on the parallel plate 11 or the surface coated with the adhesive. Then, a plurality of laminated division bodies 12 shown in FIG. 2B are formed (S3: cutting step, second step). And in each laminated division 12, at least 1 (at least one of front and back) which became a cut surface at the time of cutting | disconnection by S3 is grind | polished (S4: grinding | polishing process, 5th process).

  Subsequently, as shown in FIG. 2 (c), the plurality of laminated division bodies 12 are again stacked upward (S5: lamination step, third step), and these are cut in a vertical direction at a predetermined pitch (S6: Cutting step, third step). In S5, the plurality of laminated division bodies 12 are temporarily joined with the second adhesive or fixed to an auxiliary plate or the like. Further, the cutting pitch in S6 may be a cutting pitch such that there is one bonded portion (bonding surface 21) of the transparent medium per optical element finally obtained, or a plurality of cutting pitches. A clear cutting pitch may be used. Thereafter, at least one of the surfaces (at least one of the front and back surfaces) that became the cut surface at the time of cutting in S6 is polished (S7: polishing step, third step).

  Next, the temporary joining portion is released (dissolving the second adhesive or releasing the fixing with an auxiliary plate or the like), and as shown in FIG. 13 is formed (S8: provisional joint releasing step, third step). Note that the temporary joint portion may be released after the next S9. And finally, as shown in FIG.2 (e), each rod-shaped body 13 is cut | disconnected by a predetermined | prescribed pitch by the surface perpendicular | vertical to a elongate direction, and several optical element which consists of bonding of several transparent media As a cube-shaped prism 14 (S9: cutting step, third step).

  According to the above manufacturing method, a large amount of prisms 14 can be obtained by a series of processing steps in which the steps such as lamination, cutting, and polishing are repeated. Further, by forming an optical thin film on the surface of each parallel plate 11 before the stacking step of S2, the optical performance of the optical thin film is obtained as an optical element finally obtained through the above-described steps. The prism 14 can be obtained. In particular, by using the optical thin film as a polarization separation film, the finally obtained prism 14 can be used as a polarization separation element of an optical pickup. The optical thin film may be a thin film that selectively transmits or reflects incident light according to its wavelength, and an optical thin film in which the ratio between the transmitted light amount and the reflected light amount is 1: 1. It may be.

  Further, in the polishing step of S4, at least one of the surfaces that became the cut surfaces at the time of cutting in the third cutting step is polished in each stacked divided body 12. Similarly, also in the polishing step of S7, after laminating the plurality of laminated division bodies 12 and cutting them at a predetermined pitch, at least one of the surfaces that became the cut surfaces at the time of cutting is polished. In addition, even if it is a cut surface, it is not necessary to grind | polish about the surface which is not used as an optical surface in an optical element.

(2. Details of S2)
Next, the details of the above-described S2 stacking step will be described. Here, m is a natural number of 2 or more, n is a natural number of m or less, and in S2, m parallel flat plates are stacked to form the stacked body 1.

  In S2, regardless of the value of n within a predetermined section, the length of n parallel plates 11 from the cutting reference position in the direction perpendicular to the cutting surface of the laminated laminate 12 in S6, and the above cutting Thickness is greater than or equal to the reference value from a plurality of ranks divided according to the difference between the thickness and the reference value so that the difference from the reference position to the nth cutting position is within a predetermined range. The parallel plate 11 and the parallel plate 11 having a thickness less than the reference value are selected and laminated. Specific examples of such a stacking method will be described below as specific examples 1 and 2.

  For convenience of explanation below, the parallel plate 11 having a thickness equal to or greater than the reference value is also referred to as a parallel plate 11a, and the parallel plate 11 having a thickness less than the reference value is also referred to as a parallel plate 11b. In this definition, the parallel flat plate 11 having a thickness equal to the reference value becomes the parallel flat plate 11a, but may be put in the parallel flat plate 11b. That is, the parallel plate 11 having a thickness larger than the reference value may be used as the parallel plate 11a, and the parallel plate 11 having a thickness equal to or less than the reference value may be used as the parallel plate 11b.

(2-1. Specific Example 1)
First, in Example 1, the number of divisions (number of ranks) when dividing the parallel plate 11 according to the difference between the thickness and the reference value is set to four, and the parallel plates 11a and 11b are selected from the four ranks. A case of selecting and stacking will be described.

  In Example 1, as shown in FIG. 3, when manufacturing a finished product (prism 14) having an outer size of 4 mm square, the reference value of the thickness of the parallel plate 11 to be used is 2.828 mm. Each parallel plate 11 was divided into four ranks with values as boundaries. Further, the dimension classification of each rank (difference between the maximum value and the minimum value of the thickness of the parallel plate 11 allowed by each rank) is set to 0.01 mm, and the required positional deviation standard (accuracy) of the joining surface 21 is 0. .05 mm. The positional deviation standard of the joining surface 21 is that the beam shift accuracy in the optical pickup (the emitted light when the light incident on the center of the incident surface of the prism 14 is reflected and emitted by the optical thin film formed on the joining surface 21). Corresponds to a value (absolute value) indicating how much the position of is shifted from the center of the exit surface.

  Therefore, in the specific example 1, according to each thickness, the parallel plate 11 has a difference between the thickness and the reference value of −0.02 or more and less than −0.01, −0.01 or more and less than 0.00, It is classified into one of four ranks of 0.00 or more and less than +0.01 and +0.01 or more and less than +0.02. The parallel plate 11 that takes the minimum difference between the thickness and the reference value in each rank may be placed in the next lower rank. That is, according to each thickness, the plurality of parallel flat plates 11 have a difference between the thickness and the reference value of greater than −0.02 to −0.01 or less, greater than −0.01 to 0.00 or less, 0 It may be divided into any of the four ranks of greater than 0.00 and +0.01 or less and greater than +0.01 and +0.02. In addition, the unit of the numerical value which shows the difference of above-mentioned thickness and a reference value is all mm. The same applies to the following.

  Further, in the first specific example, the integer values 2, 1, -1, and -2 correspond to the ranks in descending order of the minimum value (or the maximum value) of the thickness of the parallel plate 11 permitted by each rank. It is attached. That is, each rank has an integer value that decreases monotonically in order from the largest minimum (or maximum) thickness of the parallel plate 11 allowed by each rank, and the parallel plate 11a, the parallel plate 11b, Are associated with integer values having opposite signs.

  In Specific Example 1, integer values of −2, −1, 1 and 2 correspond to each rank in order from the smallest minimum value (or maximum value) of the parallel plate 11 allowed by each rank. It may be attached. That is, each rank has an integer value that monotonously increases in order from the largest minimum (or maximum) thickness of the parallel plate 11 allowed by each rank, and includes the parallel plate 11a and the parallel plate 11b. , Integer values having opposite signs may be associated with each other.

  Hereinafter, the Example of the specific example 1 is demonstrated below as Examples 1-5. In Examples 1 to 5 (excluding Example 2), the positional deviation of the joint surface 21 in the case where a total of 11 parallel flat plates 11 were selected and laminated from each rank was evaluated by simulation. In Example 2, the number of parallel plates 11 to be stacked was 10 and the simulation was performed in the same manner.

  As shown in FIG. 4, when the thickness of the parallel plate 11 selected from each rank is t1, t2,..., In this embodiment, the laminate 1 is cut at an angle of 45 degrees to obtain a plurality of Since the laminated divided body 12 is formed and the stacked ones are cut perpendicularly this time, the length of each optical element in the direction perpendicular to the cut surface of the laminated divided body 12 is T1 = √2. t1, T2 = √2 · t2,... Therefore, if the distance between wires (wire pitch) is L, the difference D1 = T1-L, the difference D2 = (T1 + T2) -2L,... The positional deviation of the joint surface 21 can be evaluated.

  Here, in order to simplify the calculation, for example, the thickness of the first adhesive, the thickness (thickness) of the wire, the cutting margin with the wire, and the polishing margin are not considered. Instead, there is no problem if the absolute values of the differences D1, D2,... Are within 80% of the positional deviation standard (0.05 mm) of the joint surface 21 (within 0.04 mm in absolute value). evaluated.

  5 to 9 show the simulation results of Examples 1 to 5, respectively. In addition, integer value accumulation (SIGMA) A in a figure shows the accumulation value from the cutting reference position (wire insertion reference position) of the integer value A corresponding to the rank to which each parallel plate 11 belongs. The moving average refers to an average value of integer values A corresponding to each rank to which a predetermined number (here, for example, itself and three preceding and following) parallel flat plates 11 belong, which are successively stacked. Further, the plate thickness t indicating the thickness of the parallel plate 11 is a value obtained by using a random function so as to be a completely random value in the rank.

  In addition, the length T in the figure is the length of each parallel plate 11 from the cutting reference position in the direction perpendicular to the cut surface of the laminated divided body 12 in S6 (the above T1, T2,... And ΣT indicates the length of n parallel plates 11 from the cutting reference position (integrated value). Further, the integrated wire distance ΣL indicates the distance (n · L) from the cutting reference position to the nth cutting position. In Examples 1 to 5, the cutting reference position is the end position among a plurality of cutting positions into which wires are inserted. The difference D indicates the difference between the integrated ΣT and the integrated ΣL at each cutting position. ABS-MAX is the difference between the maximum value of the difference D (D-MAX) and the minimum value of the difference D (D-MIN). This indicates the value with the larger absolute value.

  In the first embodiment, as shown in FIG. 5, the ABS-MAX when the parallel plates 11 belonging to the corresponding rank are stacked so that the integer values are alternately “−1” and “1” is simulated. Asked. In Example 1, ABS-MAX is 0.011 (mm), and the target within 80% of the positional deviation standard of the joint surface 21 is cleared with a margin. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability close to 100%. It was. Therefore, the method of laminating the parallel flat plates 11 as in Example 1 is actually (when each laminated division body 12 is cut with a wire to finally obtain one optical element) It can be said that the effect of suppressing the positional deviation of the bonding surface 21 within the standard for each element is extremely high.

  In the second embodiment, as shown in FIG. 6, the ABS-MAX in the case where the parallel plates 11 belonging to the corresponding rank are laminated so that the integer values are alternately “−2” and “2” is simulated. Asked. In Example 2, ABS-MAX is 0.022 (mm), and the target within 80% of the positional deviation standard of the joint surface 21 is cleared. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability of 93%. It was. Therefore, it can be said that the method of laminating the parallel flat plates 11 as in Example 2 is also highly effective in suppressing the positional deviation of the bonding surface 21 within the standard for each optical element.

  In Example 3, as shown in FIG. 7, the parallel plates 11 belonging to the corresponding ranks were stacked so that five integer values “−1” were continued and then six integer values “1” were continued. The ABS-MAX of the case was determined by simulation. In Example 3, ABS-MAX is 0.026 (mm), and the target within 80% of the positional deviation standard of the joint surface 21 is cleared. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability of 91%. It was. Therefore, it can be said that the method of laminating the parallel flat plates 11 as in Example 3 is also highly effective in suppressing the positional deviation of the bonding surface 21 within the standard for each optical element.

  In Example 4, as shown in FIG. 8, ABS-MAX in the case where parallel plates 11 belonging to the corresponding rank are stacked so that integer values “2”, “−1”, and “−1” are alternated. Was obtained by simulation. In Example 4, ABS-MAX is 0.019 (mm), and the target within 80% of the positional deviation standard of the joint surface 21 is cleared with a margin. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability of 89%. It was. Therefore, it can be said that the method of laminating the parallel plates 11 as in Example 4 is actually highly effective in keeping the positional deviation of the bonding surface 21 within the standard for each optical element.

  In Example 5, as shown in FIG. 9, when the parallel plates 11 belonging to the corresponding rank are stacked so that the integer values “−1”, “2”, “−2”, “1” are alternated. ABS-MAX was obtained by simulation. In Example 5, ABS-MAX is 0.029 (mm), and the target within 80% of the positional deviation standard of the joint surface 21 is cleared. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability of 90%. It was. Therefore, it can be said that the method of laminating the parallel plates 11 as in Example 5 is actually highly effective in suppressing the positional deviation of the bonding surface 21 within the standard for each optical element.

(2-2. Specific Example 2)
In Example 2, the number of sections (number of ranks) when the parallel plate 11 is divided according to the difference between the thickness and the reference value is five, and the parallel plates 11a and 11b are selected from the five ranks. Will be described.

  In Specific Example 2, as shown in FIG. 10, when manufacturing a finished product having an outer size of 4 mm square, the reference value of the thickness of the parallel plate 11 to be used is set to 2.828 mm, and this reference value is the central value. Each parallel plate 11 was divided into five ranks so that the following ranks exist. Moreover, the dimension classification of each rank was 0.008 mm, and the required positional deviation standard (accuracy) of the joining surface 21 was 0.05 mm.

  Therefore, in the specific example 2, the plurality of parallel flat plates 11 have a difference between the thickness and the reference value of −0.02 or more and less than −0.012, or −0.012 or more and less than −0.004, depending on each thickness. , −0.004 or more and less than +0.004, +0.004 or more and less than +0.012, and +0.012 or more and less than +0.02. The parallel plate 11 that takes the minimum difference between the thickness and the reference value in each rank may be placed in the next lower rank. That is, according to each thickness, the plurality of parallel flat plates 11 have a difference between the thickness and the reference value of greater than −0.02 to −0.012 or less, greater than −0.012 to −0.004 or less, It may be classified into any of the five ranks of greater than −0.004, +0.004 or less, greater than +0.004, +0.012 or less, and greater than +0.012 and +0.02 or less.

  Moreover, in the specific example 2, the integer values of 2, 1, 0, −1, and −2 are assigned to each rank in order from the largest minimum value (or maximum value) of the thickness of the parallel plate 11 allowed by each rank. Are associated. That is, each rank is associated with an integer value that monotonously decreases in order from the largest minimum (or maximum) thickness of the parallel plate 11 allowed by each rank. Moreover, the integer value is 0 for the rank to which the parallel plate 11 whose thickness falls within a predetermined range (herein, within ± 0.004) from the reference value, and is a rank outside this rank, The ranks to which the parallel flat plates 11a whose thickness is larger than the reference value belong and the ranks to which the parallel flat plates 11b whose thickness is smaller than the reference value belong have mutually opposite integer values.

  In Specific Example 2, each rank has an integer value of −2, −1, 0, 1 and 2 in order from the smallest minimum (or maximum) thickness of the parallel plate 11 allowed by each rank. May be associated with each other. That is, an integer value that monotonously increases may be associated in order from the largest minimum value (or maximum value) of the thickness of the parallel plate 11 that is allowed by each rank. However, the integer value is 0 for the rank to which the parallel flat plate 11 whose thickness falls within the predetermined range from the reference value belongs, and is a rank outside the rank and having a thickness larger than the reference value. The ranks to which the flat plate 11a belongs and the ranks to which the parallel flat plate 11b having a thickness smaller than the reference value belong are integer values with opposite signs.

  Hereinafter, the Example of the specific example 2 is demonstrated below as Examples 6-13. In Examples 6 to 13 (excluding Example 8), a total of 11 parallel flat plates 11 were selected and laminated from each rank, and the positional deviation of the joining surface 21 was evaluated by simulation. In Example 8, the number of parallel plates 11 to be stacked was 10 and the simulation was performed in the same manner. FIGS. 11-18 has shown the simulation result of Examples 6-13, respectively.

  In Example 6, as shown in FIG. 11, ABS-MAX in the case where the parallel plates 11 belonging to the corresponding rank are laminated so that the integer value “0” is continuous was obtained by simulation. However, even in this case (even if parallel plates 11 belonging to the same rank are selected), as a result, a parallel plate 11a having a thickness equal to or larger than a reference value (2.828 mm) and a parallel plate 11b smaller than that are selected. This is the same as the other embodiments. In Example 6, ABS-MAX is 0.017 (mm), and clears the target within 80% of the positional deviation standard of the joint surface 21 with a margin. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability close to 100%. It was. Therefore, it can be said that the method of laminating the parallel plates 11 as in Example 6 is very effective in suppressing the positional deviation of the bonding surface 21 within the standard for each optical element.

  In Example 7, as shown in FIG. 12, the ABS-MAX in the case where the parallel plates 11 belonging to the corresponding rank are laminated so that the integer values are alternately “−1” and “1” is simulated. Asked. In Example 7, ABS-MAX is 0.026 (mm), which clears the target within 80% of the positional deviation standard of the joint surface 21. In addition, when a similar simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability of 99%. It was. Therefore, it can be said that the method of laminating the parallel plates 11 as in Example 7 is actually very effective in keeping the positional deviation of the bonding surface 21 within the standard for each optical element.

  In Example 8, as shown in FIG. 13, the ABS-MAX in the case where the parallel plates 11 belonging to the corresponding rank are stacked so that the integer values are alternately “−2” and “2” is simulated. Asked. In Example 8, ABS-MAX is 0.034 (mm), which clears the target within 80% of the positional deviation standard of the joint surface 21. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability of 95%. It was. Therefore, it can be said that the method of laminating the parallel plates 11 as in Example 8 is actually very effective in suppressing the positional deviation of the bonding surface 21 within the standard for each optical element.

  In the ninth embodiment, as shown in FIG. 14, the ABS-MAX in the case where the parallel plates 11 belonging to the corresponding rank are stacked so that the integer values “2”, “−1”, and “−1” are alternated. Was obtained by simulation. In Example 9, ABS-MAX is 0.022 (mm), and clears the target within 80% of the positional deviation standard of the joint surface 21. In addition, when a similar simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability of 99%. It was. Accordingly, it can be said that the method of laminating the parallel flat plates 11 as in Example 9 is actually extremely effective in keeping the positional deviation of the bonding surface 21 within the standard for each optical element.

  In Example 10, as shown in FIG. 15, when the parallel plates 11 belonging to the corresponding rank are stacked so that the integer values “−1”, “2”, “−2”, “1” are alternated. ABS-MAX was obtained by simulation. In Example 10, ABS-MAX is 0.021 (mm), and the target within 80% of the positional deviation standard of the joint surface 21 is cleared. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability of 98%. It was. Therefore, it can be said that the method of laminating the parallel plates 11 as in Example 10 is actually very effective in suppressing the positional deviation of the bonding surface 21 within the standard for each optical element.

  In Example 11, as shown in FIG. 16, the parallel plates 11 belonging to the corresponding rank are laminated so that the integer values “−2”, “2”, “−1”, and “1” are alternated, And ABS-MAX at the time of laminating | stacking the parallel plate 11 which belongs to the rank corresponding to integer value "0" on the outermost side was calculated | required by simulation. In Example 11, ABS-MAX is 0.029 (mm), which clears the target within 80% of the positional deviation standard of the joint surface 21. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability of 88%. It was. Therefore, it can be said that the method of laminating the parallel plates 11 as in Example 11 is also highly effective in suppressing the positional deviation of the bonding surface 21 within the standard for each optical element.

  In the twelfth embodiment, as shown in FIG. 17, the parallel plate 11 having ranks of integer values “−1”, “1”, “−1”, “1”, “−1” has a rank of integer value “0”. The ABS-MAX when the parallel plates 11 were stacked so as to be stacked symmetrically with respect to the parallel plates 11 was obtained by simulation. In Example 12, ABS-MAX is 0.014 (mm), which clears the target within 80% of the positional deviation standard of the joint surface 21. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, a result that the ABS-MAX was within 80% of the positional deviation standard of the joint surface 21 was obtained with a probability close to 100%. It was. Therefore, it can be said that the method of laminating the parallel flat plates 11 as in Example 11 is actually extremely effective in keeping the positional deviation of the bonding surface 21 within the standard for each optical element.

  In the thirteenth embodiment, as shown in FIG. 18, the parallel plates 11 with the integer values “2”, “−2”, “2”, “−1”, and “−1” are ranked with the integer value “0”. The ABS-MAX when the parallel plates 11 were stacked so as to be stacked symmetrically with respect to the parallel plates 11 was obtained by simulation. In Example 13, ABS-MAX is 0.031 (mm), and the target within 80% of the positional deviation standard of the joint surface 21 is cleared. In addition, when the same simulation was performed by changing the random value of the thickness t of the parallel plate 11, the ABS-MAX exceeded 80% of the positional deviation standard of the joint surface 21 as compared to the above-described embodiments. I found many. However, it can be said that the lamination method as in Example 13 has an effect of suppressing the positional deviation of the bonding surface 21 within the standard for each optical element by appropriately selecting the thickness t of the parallel plate 11.

  In the above description, it is assumed that a plurality of parallel plates 11 are classified (ranked) according to their thicknesses, and integer values are associated with each rank at the time of the stacking process of S2. However, when these divisions and associations are not performed, such a division process and association process may be performed before S2 (lamination process) or S1 (thin film formation process). .

(3. Effect)
In the laminating method of the specific examples 1 and 2 described above, the difference D between the integrated ΣT (the length of n parallel plates 11) and the integrated wire distance ΣL (the distance from the cutting reference position to the nth cutting position) is The parallel plates 11a and 11b are selected and stacked from a plurality of ranks so that any value of n can be within a predetermined range. That is, when laminating 11 parallel flat plates 11, n takes any value from 1 to 10, and when laminating 10 parallel flat plates 11, any value from 1 to 9 is taken, The parallel plates 11a and 11b are selected and laminated so that the difference D is within 0.05 mm which is the positional deviation standard of the joining surface 21. When 11 parallel flat plates 11 are laminated, the range of values that n can take is 2 to 9, and the optical elements at both ends may be discarded. Further, when 11 parallel flat plates 11 are laminated, and the cutting reference position in S6 is positioned in the middle of the direction of taking the cutting pitch as described later, the range of values that n can take is 1-5 may be sufficient.

  By such a laminating method, in the cutting step of S6, each laminated division body 12 can be cut in a state in which the portion to be cut of each laminated division body 12 is brought close to each cutting position of a predetermined pitch. . Accordingly, it is possible to suppress the positional deviation of the bonding surface 21 for each optical element obtained by cutting at a predetermined pitch. As a result, when the optical element manufactured by the manufacturing method of the present invention is applied to an optical pickup, the beam shift accuracy can be easily ensured for each obtained optical element.

  In particular, the parallel plate 11a having a thickness greater than or equal to the reference value and the parallel plate 11b having a thickness less than the reference value or less than the reference value are selected and laminated from a plurality of ranks. By laminating the parallel plates 11a and 11b, it is possible to reliably reduce the increase in the positional deviation of the joining surface 21 in the positive direction or the minus direction, and the above-described effect of suppressing the positional deviation of the joining surface 21 is ensured. be able to.

  Further, according to the present invention, since the positional deviation of the joint surface 21 can be suppressed by utilizing the variation in the thickness of each parallel plate 11, the tolerance of the single parallel plate 11 to be procured can be widened, and the optical element It is also possible to reduce the manufacturing cost.

  In Specific Example 1 described above, each rank is associated with an integer value A that monotonously decreases or increases in descending order of the maximum value or minimum value of the thickness of the parallel plate 11 allowed by each rank. And the rank to which the parallel plate 11a whose thickness A is greater than the reference value or greater than the reference value belongs and the rank to which the parallel plate 11b whose thickness is less than or less than the reference value belong to each other. In the case of the opposite sign, the accumulated value (integer value accumulated ΣA) in the range of values (for example, 1 to 10) that can be taken by the integer value A corresponding to each parallel plate 11 to be laminated is which parallel plate. The parallel plates 11 are selected and stacked from each rank so that the accumulated value up to 11 falls within a predetermined range (for example, within a range of −5 to 5). The predetermined range may be appropriately set according to, for example, the required positional deviation standard of the bonding surface 21 and the width (dimension category) of each rank.

  In Specific Example 2 described above, each rank is associated with an integer value A that monotonously decreases or increases in descending order of the maximum value or minimum value of the thickness of the parallel plate 11 permitted by each rank. And the integer value A is 0 for the rank to which the parallel plate 11 whose thickness falls within the predetermined range from the reference value, is a rank outside the rank, and the thickness is less than the reference value. If the rank to which the larger parallel flat plate 11a belongs and the rank to which the parallel flat plate 11b whose thickness is smaller than the reference value have integer values opposite to each other, the alignment corresponding to each parallel flat plate 11 to be laminated. The accumulated value (integer value accumulated ΣA) in the range of values that n of the numerical value A can take (for example, an integral value accumulated ΣA) is within a predetermined range (for example, within a range of −2 to 2). ) Whole way, are laminated by selecting the parallel plate 11 from each rank.

  By adopting such a laminating method of specific example 1 or 2, in the cutting step of S6, the portion to be cut of each laminated divided body 12 is reliably brought close to each cutting position of a predetermined pitch. Each laminated | stacked division body 12 can be cut | disconnected. As a result, the positional deviation of the bonding surface 21 can be reliably suppressed for each optical element obtained by cutting at a predetermined pitch, and the beam shift accuracy can be easily and reliably ensured for each obtained optical element. .

  Moreover, in the specific example 1 mentioned above, since the moving average of the integer value A changes within a predetermined range (for example, within a range of −1 or more and 1 or less), the lamination method of the specific example 1 is as follows. It can also be expressed as That is, each rank is associated with an integer value A that monotonously decreases or increases in descending order of the maximum value or the minimum value of the thickness of the parallel plate 11 allowed by each rank, and the integer value A However, when the rank to which the parallel flat plate 11a whose thickness is greater than or equal to the reference value belongs and the rank to which the parallel plate 11b whose thickness is less than the reference value or less than the reference value belong have opposite signs. In the stacking step of S2, the average value of the integer values A corresponding to the respective ranks to which the predetermined number (for example, three) of the parallel flat plates 11 that are continuously stacked belongs is the predetermined number of parallel flat plates that are continuously stacked. The parallel plates 11 are selected and stacked from each rank so that the different combinations in the range of values that n of 11 can take (for example, 1 to 10) fall within a predetermined range.

  In the above-described specific example 2, the moving average of the integer value A changes within a predetermined range (for example, within a range of −1 or more and 1 or less). Therefore, the lamination method of the specific example 2 is as follows. It can also be expressed as That is, each rank is associated with an integer value A that monotonously decreases or increases in descending order of the maximum value or the minimum value of the thickness of the parallel plate 11 allowed by each rank, and the integer value A However, the rank to which the parallel plate 11 whose thickness falls within the predetermined range from the reference value is 0, and is a rank outside the rank, to which the parallel plate 11a having a thickness greater than the reference value belongs. And the rank to which the parallel flat plate 11b having a thickness smaller than the reference value is an integer value of opposite signs, in the stacking step of S2, a predetermined number (for example, three) stacked continuously For each different combination in the range (for example, 1 to 10) in which the average value of the integer value A corresponding to each rank to which the parallel plate 11 belongs can be taken by n of a predetermined number of parallel plates 11 stacked successively So as to fall within a predetermined range, laminating select the parallel plate 11 from each rank.

  Even in such a stacking method of the specific example 1 or 2, each of the stacked divided bodies 12 is in a state in which the portion to be cut of each stacked divided body 12 is surely brought close to each cutting position of a predetermined pitch. Therefore, it can be said that the positional shift of the joint surface 21 can be reliably suppressed, and the beam shift accuracy can be easily and reliably ensured.

  Further, in the first specific example, among the two ranks to which the parallel flat plate 11a having a thickness equal to or larger than the reference value belongs, the rank with the smallest allowable minimum value of the parallel flat plate 11a (the difference between the thickness and the reference value). Is a rank of 0.00 or more and less than +0.01) is associated with a first integer value (for example, 1). The other ranks (ranks where the difference between the thickness and the reference value is +0.01 or more and less than +0.02) have the same sign as the first integer value, and the absolute value is the absolute value of the first integer value. It is associated with an integer value (for example, 2) increased from the value by a predetermined value (for example, 1). When the number of divisions is 4 or more and there are a plurality of other ranks, the other ranks are the first integer values in order from the smallest minimum value of the parallel plate 11a allowed by each rank. As long as the absolute value of the first integer value is increased by a predetermined value from the absolute value of the first integer value.

  Similarly, of the two ranks to which the parallel plate 11b having a thickness smaller than the reference value belongs, the rank having the largest minimum value of the allowable thickness of the parallel plate 11b (the difference between the thickness and the reference value is −0). (Rank of .01 or more and less than +0.00) is associated with a second integer value (for example, -1) having a different absolute value from the first integer value. The other rank (the difference between the thickness and the reference value is −0.02 or more and less than −0.01) has the same sign as the second integer value, and the absolute value is the second integer value. Is associated with an integer value (for example, -2) that is increased from the absolute value by a predetermined value (for example, 1). When the number of divisions is 4 or more and there are a plurality of other ranks, the other ranks are the second integer values in order from the largest minimum value of the parallel plate 11b allowed by each rank. As long as the absolute value of the second integer value is increased by a predetermined value from the absolute value of the second integer value.

  Moreover, in the specific example 2, the allowable thickness of the parallel flat plate 11a among the plurality of ranks to which the parallel flat plate 11a having a thickness larger than the reference value belongs is ranked outside the rank of the integer value 0. The rank with the smallest minimum value (the rank where the difference between the thickness and the reference value is +0.004 or more and less than +0.012) is associated with the first integer value (for example, 1), and other ranks (thicknesses) And the reference value have a difference of +0.012 or more and less than +0.02 and have the same sign as the first integer value, and the absolute value is a predetermined value (for example, 1) from the absolute value of the first integer value. ) Is increased by an integer value (for example, 2). When the number of divisions is 5 or more and there are a plurality of other ranks, the other ranks are the first integer values in order from the smallest minimum thickness of the parallel plate 11a allowed by each rank. As long as the absolute value of the first integer value is increased by a predetermined value from the absolute value of the first integer value.

  Similarly, the minimum value of the allowable thickness of the parallel plate 11b among the plurality of ranks to which the parallel plate 11b having a thickness smaller than the reference value belongs is ranked outside the rank of the integer value 0. The largest rank (the difference between the thickness and the reference value is −0.012 or more and less than −0.004) is associated with the second integer value (for example, −1) having a different absolute value from the first integer value. The other ranks (ranks where the difference between the thickness and the reference value is −0.02 or more and less than −0.012) have the same sign as the second integer value, and the absolute value is the second value. It is associated with an integer value (for example, -2) increased by a predetermined value (for example, 1) from the absolute value of the integer value. When the number of divisions is 5 or more and there are a plurality of other ranks, the other ranks are the second integer values in order from the largest minimum value of the parallel plate 11b allowed by each rank. As long as the absolute value of the second integer value is increased by a predetermined value from the absolute value of the second integer value.

  By associating an integer value with each rank in this way, it is possible to easily grasp the stacked state of the parallel plates 11a and 11b simply by looking at the order in which the integer values are arranged, the sign of the integer value, and the absolute value. It becomes possible. In addition, by associating the integer values as described above, it is possible to easily perform the above-described positional deviation evaluation of the joint surface 21 based on the accumulated integer value ΣA and the moving average of the integer values.

  Moreover, in the specific example 1, since the rank to which the parallel plate 11a belongs is associated with a positive integer value, and the rank to which the parallel plate 11b belongs is associated with a negative integer value, there is a positive / negative integer value. It is possible to immediately and easily determine whether or not the parallel plates 11a and 11b are laminated. On the other hand, in the second specific example, when an optical element is obtained using, for example, an odd number of parallel plates 11, one of the odd numbers is set as the parallel plate 11 belonging to the rank of the integer value 0, thereby obtaining an integer value. By aligning the number of parallel plates 11 belonging to the positive rank and the number of parallel plates 11 belonging to the negative rank of the integer value, the positional deviation of the joining surface 21 can be suppressed in a balanced manner.

  In Examples 1, 2, 5, 7, 8, and 10, parallel flat plates 11 belonging to ranks having different signs of integer values are alternately stacked. Since such a stacking method works in a direction to cancel the positional deviation of the bonding surface 21, the positional deviation of the bonding surface 21 can be reliably suppressed for each optical element finally obtained. In particular, as in the first, second, seventh, and eighth embodiments, the effect can be reliably obtained by alternately stacking the parallel flat plates 11 belonging to the same rank of the absolute value of the integer value.

  In the second and eighth embodiments, the number of the parallel plates 11 stacked is an even number of ten, the parallel plates 11 whose integer value sign belongs to a positive rank, and the parallel plates 11 whose integer value sign belongs to a negative rank. The same number of flat plates 11 are stacked. Thereby, when the number of parallel flat plates 11 to be stacked is an even number, the average positional deviation of the bonding surfaces 21 can be made as small as possible. Further, for example, if the parallel flat plates 11 whose integer value sign belongs to the positive rank and the parallel flat plates 11 whose integer value sign belongs to the negative rank are alternately laminated, the optical elements finally obtained are bonded. The positional deviation of the surface 21 can be suppressed with a good balance.

  In the eleventh and twelfth embodiments, the number of parallel plates 11 stacked is an odd number of eleven, and parallel plates whose integer value sign belongs to a positive rank and parallel plates whose integer value sign belongs to a negative rank. Are stacked in the same number, and one parallel plate belonging to a rank having an integer value of 0 is stacked. Thereby, when the number of parallel flat plates 11 to be stacked is an odd number, the average of the positional deviations of the bonding surfaces 21 can be made as small as possible. In addition, by setting one of the odd-numbered sheets as the parallel plate 11 belonging to the rank of the integer value 0, the number of the parallel plates 11 belonging to the rank of the integer value and the sign of the integer value is negative. The number of parallel flat plates 11 belonging to the rank can be made uniform. For example, by alternately laminating them, it is possible to suppress the positional deviation of the bonding surface 21 with a good balance. When the number of parallel plates 11 stacked is an odd number, the number of parallel plates 11 belonging to a rank having an integer value of 0 is not limited to one, and may be an odd number of three or more.

(4. Other examples of cutting reference positions)
In each of the above Examples 1 to 13, in the cutting step of S6, the cutting reference position when cutting the plurality of laminated division bodies 12 at a predetermined pitch is set in each of the laminated division bodies 12 in the plurality of cutting positions. The position at the extreme end in the direction of taking the cutting pitch (perpendicular to the cut surface) is used, but may be, for example, the second position from the end or the middle position.

  In particular, when the number m of the parallel flat plates 11 to be stacked is an odd number, the center position in the above direction is desirable, and when the number m of the parallel flat plates 11 to be stacked is an even number, the center position in the above direction is preferable. It is desirable to be. If the length in the above direction in each layered divided body 12 is constant, when the cutting reference position is taken at the central position or the substantially central position in the above direction, the cutting reference position is taken at one end in the above direction. In comparison, the accumulation of errors at the end farthest from the cutting reference position (position displacement amount of the joining surface 21) can be suppressed to half (when the number m is an odd number) or approximately half (when the number m is an even number). it can. Thereby, the effect which can suppress the position shift of the joint surface 21 for every optical element finally obtained becomes high.

(5. Other examples of classification)
By the way, when the plurality of parallel flat plates 11 are divided into a plurality of sections according to their thicknesses, a value of ± 1/5 of the customer required standard of the positional deviation amount of the joint surface 21 and the beam shift accuracy is set as the thickness and the reference. A maximum value (upper limit value) or a minimum value (lower limit value) of a difference from the value may be set, and the range between the maximum value and the minimum value may be divided into a plurality of sections.

  For example, FIG. 19 shows a case where the customer request standard is 0.05 mm and the number of sections is four, and FIG. 20 shows a plurality of parallel plates when the customer request standard is 0.05 mm and the number of sections is five. An example of 11 divisions is shown. In FIG. 19 and FIG. 20, since the customer requirement standard is 0.05 mm, ± 0.010 (mm), which is ± 1/5, is used as the upper limit value and lower limit value of the difference between the thickness and the reference value. In between, it is divided into a plurality of categories.

  21 shows a case where the customer requirement standard is 0.10 mm and the number of sections is four, and FIG. 22 shows a plurality of parallel flat plates when the customer requirement standard is 0.10 mm and the number of divisions is five. An example of 11 divisions is shown. In FIGS. 21 and 22, since the customer requirement standard is 0.10 mm, ± 0.020 (mm), which is ± 1/5, is used as the upper limit value and the lower limit value of the difference between the thickness and the reference value. In between, it is divided into a plurality of categories.

  By selecting and laminating the parallel flat plates 11 from the ranks thus divided, the positional deviation amount of the joining surface 21 with respect to the cutting position of the wire saw falls within about 60% of the customer requirement standard or less, and the customer It was found from simulations similar to those in Examples 1 to 13 that the requirement standard is extremely reduced.

  Note that the number of sections is suitably about 4 to 6 sections, but is not particularly limited. The number of sections may be odd or plural, but the odd number is convenient in that it can be easily handled when the number of stacked parallel plates 11 is odd or plural. That is, when the number of sections is an odd number, the number of stacked parallel plates 11 depends on whether one parallel plate 11 of the middle section (rank of integer value 0) is inserted or not as shown in the specific example 2 above. When the number is an odd number, it is possible to flexibly cope with both when there are a plurality of sheets.

DESCRIPTION OF SYMBOLS 1 Laminated body 11 Parallel plate 12 Laminated division body 14 Prism (optical element)

Claims (15)

A first step of forming a laminate by laminating parallel plates made of a plurality of transparent media via an adhesive;
A second step of forming a plurality of laminated division bodies by cutting the laminated body at a predetermined inclination angle and at a predetermined pitch;
A plurality of the laminated divided bodies are stacked and cut at a predetermined pitch in a direction perpendicular to the cut surface in the second step and in the direction in which the parallel flat plates are laminated, thereby forming a plurality of laminated transparent media. And a third step of forming the optical element.
In the third step, a predetermined reference is made on the basis of a cutting reference position determined so that any one of the cut surfaces in the third step satisfies a predetermined positional relationship with respect to the joint surface which is a boundary of the parallel plates. Cutting the laminated segment at a pitch of
After the previous step of dividing the plurality of parallel plates into a plurality of ranks according to the difference between the thickness of the parallel plates and a reference value of the thickness,
In the first step, the thickness is greater than or equal to the reference value or the reference value so that the positional deviation of the joint surface with respect to each cut surface in the third step is within a predetermined range at any cutting position. A method of manufacturing an optical element, comprising: selecting and laminating the parallel flat plates belonging to a rank larger than the value and the parallel flat plates belonging to a rank having a thickness less than the reference value or less than the reference value. Each rank is associated with an integer value that monotonously decreases or increases in descending order of the maximum value or minimum value of the thickness of the parallel plate allowed by each rank, and the integer value is The rank to which the parallel plate whose thickness is greater than or equal to the reference value belongs and the rank to which the parallel plate whose thickness is less than the reference value or less than the reference value are opposite to each other. In case,
In the first step, from the plurality of ranks, the parallel values of the integer values corresponding to the parallel plates to be stacked are within the predetermined range so that the cumulative value up to any of the parallel plates is within a predetermined range. The method for manufacturing an optical element according to claim 1, wherein a flat plate is selected and laminated. The plurality of ranks are respectively associated with integer values that monotonously decrease or increase in order from the largest of the maximum or minimum values of the parallel plate thickness allowed by each rank, and The integer value of the rank to which the parallel plate whose thickness is greater than or equal to the reference value belongs and the rank to which the parallel plate whose thickness is less than or less than the reference value belong are opposite to each other. In case,
In the first step, an average value of integer values corresponding to each rank to which a predetermined number of the parallel plates to be successively stacked belongs is different for each different combination of the predetermined number of the parallel plates to be continuously stacked. 2. The method of manufacturing an optical element according to claim 1, wherein the parallel flat plates are selected and stacked from the plurality of ranks so as to be within a predetermined range. Among a plurality of ranks to which the parallel plate to which the thickness is equal to or greater than the reference value or larger than the reference value belongs, the rank with the smallest minimum thickness of the parallel plate allowed is the first integer value. On the other hand, each of the other ranks has the same sign as the first integer value in order from the smallest of the minimum thicknesses of the parallel plates allowed by each rank, and its absolute value is Associated with an integer value that increases by a predetermined value from the absolute value of the first integer value,
Among a plurality of ranks to which the parallel plates whose thickness is less than the reference value or less than the reference value belong, the rank with the largest allowable minimum value of the parallel plate is the absolute value of the first integer value. While the values are associated with different second integer values, the other ranks have the same value as the second integer value in descending order of the minimum thickness of the parallel plates allowed by each rank. The optical element manufacturing method according to claim 2 or 3, characterized in that it is associated with an integer value that is a sign and whose absolute value increases by a predetermined value from the absolute value of the second integer value. Method. Each of the plurality of ranks is associated with an integer value that monotonously decreases or increases in descending order of the maximum value or minimum value of the parallel plate thickness allowed by each rank. The numerical value is 0 for the rank to which the parallel plate whose thickness falls within a predetermined range from the reference value, and the rank other than this rank, to which the parallel plate having a thickness greater than the reference value belongs. In the case where the parallel plate with a thickness smaller than a reference value is an integer value with opposite signs to the rank to which the parallel plate belongs,
In the first step, from the plurality of ranks, the parallel values are accumulated so that an accumulated value of integer values corresponding to each of the parallel plates to be stacked is within a predetermined range regardless of the accumulated value up to any of the parallel plates. The method for manufacturing an optical element according to claim 1, wherein a flat plate is selected and laminated. Each of the plurality of ranks is associated with an integer value that monotonously decreases or increases in descending order of the maximum value or minimum value of the parallel plate thickness allowed by each rank. The numerical value is 0 for the rank to which the parallel plate whose thickness falls within a predetermined range from the reference value, and the rank other than this rank, to which the parallel plate having a thickness greater than the reference value belongs. In the case where the parallel plate with a thickness smaller than a reference value is an integer value with opposite signs to the rank to which the parallel plate belongs,
In the first step, an average value of integer values corresponding to respective ranks to which a predetermined number of the parallel plates continuously stacked belongs is different for each different combination of the predetermined number of parallel plates stacked continuously. 2. The method of manufacturing an optical element according to claim 1, wherein the parallel plates are selected and stacked from the plurality of ranks so as to fall within a predetermined range. Among ranks other than ranks of integer value 0 and having parallel plates having a thickness greater than the reference value, the rank having the smallest minimum value of the allowed parallel plate thickness is: While being associated with the first integer value, each of the other ranks has the same sign as the first integer value in order from the smallest of the minimum values of the parallel plate thickness allowed by each rank, And the absolute value is associated with an integer value that increases by a predetermined value from the absolute value of the first integer value,
Among ranks other than the rank of the integer value 0 and having the parallel plate whose thickness is smaller than the reference value, the rank having the largest allowable minimum value of the parallel plate thickness is: While each of the other ranks is associated with a second integer value that is different in absolute value from the first integer value, each rank is in descending order of the minimum value of the thickness of the parallel plate, 7. The second integer value has the same sign and is associated with an integer value whose absolute value increases by a predetermined value from the absolute value of the second integer value. The manufacturing method of the optical element of description.   8. The method of manufacturing an optical element according to claim 1, wherein in the first step, the parallel plates belonging to ranks having different signs of integer values are alternately stacked.   9. The method of manufacturing an optical element according to claim 8, wherein in the first step, the parallel plates belonging to the same rank of absolute values of integer values are alternately stacked. When the number of parallel flat plates to be laminated in the first step is an even number,
The said 1st process WHEREIN: The said parallel plate which the sign of an integer value belongs to a positive rank, and the said parallel plate which the sign of an integer value belongs to a negative rank are laminated | stacked by the same number. The manufacturing method of the optical element of any one of 1-9. When the number of parallel plates laminated in the first step is an odd number,
In the first step, the same number of the parallel plates belonging to the positive rank of the integer value and the parallel plates belonging to the negative rank of the integer value are stacked, and the rank of the integer value is 0. 8. The method of manufacturing an optical element according to claim 5, wherein an odd number of the parallel plates belonging to 1 is stacked.   The cutting reference position when cutting the plurality of laminated division bodies at a predetermined pitch in the third step is a center or a substantially central position in the direction of taking the cutting pitch in each of the laminated division bodies. The method for manufacturing an optical element according to claim 1, wherein the optical element is manufactured. A fourth step of forming an optical thin film on the surface of each parallel plate;
The method of manufacturing an optical element according to claim 1, wherein the fourth step is performed before the first step. Each of the laminated division bodies further includes a fifth step of polishing at least one of the surfaces that have become cut surfaces at the time of cutting in the second step,
The method for manufacturing an optical element according to claim 1, wherein the fifth step is performed before the third step.   15. In the third step, the plurality of stacked division bodies are stacked and cut at a predetermined pitch, and then at least one of the surfaces that became the cut surface at the time of cutting is polished. The manufacturing method of the optical element of any one of these.

Images