Method for determining a recipe for assembling a combination of n optical elements, and resulting optical module

Abstract

The invention relates to a method for determining an assembly recipe for optical elements, such as optical lenses, intended to equip an optical module, said module having a stacking axis (z) as well as a reference plane (x, y) perpendicular to the stacking axis, and comprising a combination of N types of optical lenses (Lk), k = [1, ... n] provided with two interfaces f and mounted in a barrel (10), the method aiming to determine, for a subset of n lenses, the angle Θk of rotation with respect to a reference angular position Θ0, corresponding to the closest spacing of the 2n apexes (COk, f) of said subset, projected in the reference plane (x, y).

Description

Description Title: Method for determining an assembly recipe for a combination of N optical elements, and resulting optical module

[0001] The invention relates to a method for determining an assembly recipe for a combination of N optical elements forming an optical module, and more particularly for a combination of N optical lenses from molds themselves composed of a plurality of cavities. State of the art

[0002] Optical lenses, such as those used in smartphone cameras and video cameras, typically consist of an assembly of optical components, primarily optical lenses, which are stacked in a holder such as a barrel. These optical modules are then assembled with CMOS or CCD sensors to form the camera module.

[0003] Due to their small size, the production of these lenses involves uncertainties in relation to the design, and requires special monitoring of molding parameters such as molding time, temperature and pressure profile.

[0004] During the molding stages, measuring tools based on atomic force microscopy (marketed under the name UA3P) are used to control the shape, thickness, compare them with theoretical values ​​and ensure that the measurements remain within the prescribed tolerances.

[0005] Despite these controls, the apex of each interface is very generally slightly offset from the support center defined by the lens attachment areas, among other things by the effects of thermal shrinkage during molding.

[0006] In order to compensate for these shifts, it is known to empirically carry out a large number of rotations on the different lenses in order to reduce these shifts and enhance the optical quality of the objectives.

[0007] However, these experimental designs consist of sequentially carrying out a considerable number of rotations in order to explore the range of possible combinations quite widely. Description of the invention

[0008] This is why the invention proposes a method for identifying, following a minimized number of steps and rotation tests, the optimal rotation to be carried out to best approach the alignment along the optical axis of the apices of each interface of these lenses in the assembly barrel.

[0009] The invention relates in particular to a method for determining an assembly recipe for a combination of N optical elements forming an optical lens when said optical elements are stacked in a barrel in a stacking direction substantially equal to the optical axis, at least one of the optical elements comprising two optical interfaces each comprising an apex, said method comprising for each of the optical elements of a subset of n optical elements of said combination of N optical elements, the following steps: - determining a trajectory of the apex for each optical interface of said optical element, in the plane (x, y) perpendicular to said stacking direction (z), from at least two angular orientations of said optical element around said stacking direction, - determining the coordinates of the apex for each optical interface of said optical element, in the plane (x, y) perpendicular to said stacking direction (z), from at least two angular orientations of said optical element around said stacking direction,y) perpendicular to said stacking direction (z), verifying a reduced distance compared to the trajectories of the other optical elements of the subset of n optical elements, and, - determining the angular orientation of said optical element in which the apices of the two interfaces of said optical element are closest to the trajectories of the other optical elements of the subset of n optical elements.

[0010] This requires a step, called global minimization, in which we globally determine by calculation a set of rotations of angle Θk, the rotations of the n lenses, such that we obtain the positioning of the apices of each interface as close as possible to each other.

[0011] We can use a global minimization algorithm to obtain this angle set Θk, such as a traveling salesman minimization type algorithm – which minimizes a sum of distances to be covered by the traveler, here distances between apexes – modified to be adapted to the problem to be solved.

[0012] Various embodiments of the invention are provided, integrating, according to all of their possible combinations, the various optional characteristics set out below.

[0013] According to a first preferred aspect of the invention, the optical elements are optical lenses having an axisymmetric profile of the aspherical type and more particularly a substantially flat and circular peripheral zone defining a support center and intended to be supported during the assembly of said lenses in the barrel. Each lens contributes its optical function to the complete optical function of the lens and comprises a central zone for each of its two interfaces, the position of the apex of which, i.e. the apparent center of each interface along the optical axis of the lens, is the subject of said optimization.

[0014] The method is characterized in that it comprises a set of steps aimed at determining the optimal angular orientation of a subset of n types of optical lenses, chosen from among the N types of optical lenses, namely: - a step in which the N types of optical lenses are mounted in the barrel in a reference angular position Θ0, and the position of the apex of each interface is measured, - a step in which the trajectory of the apexes is determined for the subset of n types of lenses chosen from the N types of lenses, with rotation of the n types of lenses around the optical axis passing through their center of support, for some assemblies of different lenses. From the analysis of the measurements obtained, the position of the projected trajectories in the reference plane is calculated, trajectories which represent the displacement of each apex of each interface during the rotation of the lens around its optical axis passing through its center of support,

[0015] We then use an algorithm that reveals a point of reduced distance DR with a step of searching for the point DR and a step of minimizing the distances to this point DR, i.e.: - a step in which we determine by calculation the coordinates of a point of reduced distance DR to the set of trajectories, projected in the reference plane, - a step in which we calculate for each pair of trajectories (associated with one of the two interfaces) the coordinates of the two points that are closest to the point of reduced distance DR.

[0016] From these two points associated respectively with one of the two interfaces for each of the n types of lenses, we calculate the angle Θk of rotation relative to the reference angular position Θ0, which allows the positioning of the apices of each interface at the point globally closest to the point of reduced distance.

[0017] Preferably, after determining the rotation angles, a step is carried out in which a characteristic quantity of the quality of the objective is measured, obtained with the combination of the N types of lenses including the n types of lenses oriented according to their own angle Θk.

[0018] Advantageously, a step is carried out of comparing the measurement of the characteristic quantity of the quality of the objective obtained with the combination of the N types of lenses including the subset of n types of lenses oriented according to their own angle Θk, with a first predefined threshold.

[0019] This first threshold makes it possible to define whether the assembly thus carried out is admissible or not in terms of quality.

[0020] Preferably, the characteristic quantity of the quality of the objective is the value of the Modulation Transfer Function (MTF).

[0021] According to a second preferred aspect of the invention, the step of determining the trajectories representing the displacement of the apices comprises: - a sub-step in which the coordinates of the apex projected in the reference plane and in the reference angular position Θ0 are measured for each of the n types of lenses, - a sub-step, if the trajectories sought are circles, in which a rotation of 180 degrees around the optical axis passing through its support center is carried out for each of the n types of lenses so as to place them in the angular position Θ180, for sub-parts of the n lenses of the objective, which leads to achieving several objectives, for example two.if the trajectories sought are not circles, more rotations are carried out, for example four non-cocyclic (different by a step of 90°), or five preferably (by steps of 72°) if the trajectories are ellipses, - a sub-step in which the equations of the trajectory projected in the reference plane for each apex are calculated for each of the n types of lenses.

[0022] The assembly barrel alone can determine the rotational guidance of the lenses. In this case, the equations of the trajectories of each lens obtained by the rotation of a lens Lk are independent of the rotations of the neighboring lenses.

[0023] In more complex designs, the lenses themselves may have assembly shoulders that come into contact with neighboring lenses. If these shoulders are well centered relative to the barrel, they do not significantly change the rotational guidance of the barrel. Otherwise, the position of the trajectory of each lens Lk can slightly depend on the rotation of the lens Lk-1 and / or the lens Lk+1. We can then complete the rotation step of a series of lenses Lk (for example k even) and measurement with some rotation of the neighboring lenses (for example k odd) in support in order to add and determine quantification coefficients of the effects of coupling of rotations of neighboring lenses, which at first order amount to parameterizing a rotation displacement of the lens Lk from the rotation of the neighboring lenses: in the case of circular trajectories, the diameter of the circle of each interface f of the lens Lk can be substantially constant, but the position of its center is modulated by the rotation of the lens or of the two lenses in contact (Lk-1 and / or Lk+1) and possibly of more distant lenses Lk-2 and Lk+2.

[0024] In this case, the coupling coefficients of, for example, the motion of the center of the trajectory circles are determined from an analysis of additional experiments of rotations of adjacent lenses, with the rotations of these adjacent lenses.

[0025] Similarly, in the case of angles (tip and tilt) between the support faces of a lens Lk, we may have to consider these additional coupling coefficients.

[0026] The rest of the presentation is carried out with the hypothesis that the trajectories of the apparent centers of each interface are circles, but other types of trajectories are possible and the method is adaptable for example by increasing the number of values ​​of the test angles beyond two values ​​(Θ0 and Θ180) per lens, in order to determine the set of parameters of these trajectories. With two points which are known to be one of the diameters of a circle, we determine both its radius and its center. We can take 3 points Θ0, Θ120, Θ-120 also for more redundancy and resistance to noise in the determination of the circular trajectory. With four (non-cocyclic) or preferably five points, we determine both the major and minor axes and center of an ellipse.

[0027] Advantageously, the measurement of the coordinates of the apexes is carried out using low coherence interferometry techniques, as described in document WO2023 / 057539A1.

[0028] Also advantageously, in addition to the apex coordinate measurements for several angles (for example Θ0 and Θ180), the thickness of the n types of lenses at the level of said apices is measured. This measurement makes it possible to validate whether or not the thickness is within the prescribed tolerances. It also makes it possible to modify these thicknesses if necessary.

[0029] Also advantageously, in addition to the apex coordinate measurements for several angles (e.g. Θ0 and Θ180), the angle deviations of the planes of the n types of lenses at said apices can be measured. These planes can be defined as mean planes, or planes tangent to the lens interfaces, or planes perpendicular to the normal rotational symmetry vector of each lens interface.

[0030] In fact, the plane of each lens can be slightly inclined along the (y) axis, which generates an angle called "tilt" with the optical axis (z), or inclined along the (x) axis, which generates an angle called "tip" with the optical axis (z).

[0031] According to a third preferred aspect of the invention, when the optical module comprises one or more spacers of variable thickness, intended to be interposed between some of the N types of lenses, all of the steps aimed at determining the optimal angular orientation of each of the n types of optical lenses are carried out, by varying the thickness of the spacers, then the thicknesses of the spacers are adjusted. This step can take place during the angle optimization phase or when the characteristic quantity of the quality of the lens obtained with a combination of N types of lenses including the subset of n types of lenses oriented according to their own angle Θk, is greater than the first predefined threshold.

[0032] Thus, for a combination of N types of lenses acceptable in terms of quality, the assembly can be optimized by modifying the thickness of the spacers.

[0033] According to a fourth preferred aspect of the invention, according to which each of the N types of lenses comes from a manufacturing mold comprising M cavities, all of the steps are carried out to determine the optimal angular orientation for the subset of n types of lenses chosen from the N types of lenses and preferably in parallel for the M cavities, that is to say that M assemblies of the N types of lenses are produced simultaneously, with the exposed angle variations, and determination of the trajectories for each lens for each of the series of objectives from each cavity.

[0034] On the one hand, this allows the optimization of M assemblies in parallel, or on a subset of m lenses chosen from the M (for example in the case of unexploitable cavities which make m smaller than M), which is the aim of producing the molding of M cavities for each type of lens.

[0035] But on the other hand, according to a fifth aspect of the invention, this allows during the calculation step, to compare the trajectories obtained by calculated permutations between the assemblies, in order to find a better solution than without permutation. Permutation is understood to mean the calculated virtual realization of several objectives, each objective being produced from several different cavities, for example lens L1 (k=1) from cavity 1, lens L2 from cavity 5, lens L3 from cavity 12, etc. (Such a permutation by calculation is quite close to what would be obtained by its realization and real measurement of the trajectories of the centers of the lenses, as long as for example, the lenses are fairly well guided during their rotation by support in the assembly barrel, or even that the quality of centering of the support zones is fairly good.)

[0036] Optionally in addition, even if this increases the number of physical tests to be carried out, it is possible to carry out some physical permutations in the sense of the second aspect to provide the calculation step of the fifth aspect with data of trajectories in real support with the permuted lenses, in order to choose different lens combinations which provide better alignment for all the objectives produced.

[0037] Thus, permutations are made on the cavities, to increase the characteristic magnitude of the quality of the produced lens, these permutations being carried out in the calculation stages or during the assembly of the lenses in the barrel.

[0038] Thus, preferably, we determine the combination of N types of optical lenses including the subset of n types of optical lenses optimally oriented according to their own angles Θk,j, and coming from a subset of m cavities chosen from among the M cavities, which globally offers the best characteristic quantity of the quality of each of the m objectives.

[0039] Preferably, the set of parameters to be controlled at the output of the manufacturing mold includes the shape, the thickness, the offsets, the angular deviations from the plane (x,y), the optical index.

[0040] The invention also relates to an optical module obtained from the implementation of an embodiment of the method according to the invention, the optical module having a Modulation Transfer Function (MTF) greater than a first predefined threshold.

[0041] According to a sixth preferred aspect of the invention, it is possible to optimize all the angles found at the end of one of the preceding steps to further improve the optical quality obtained, by small variations of the angles (Θk). To do this, a series of measurements of the characteristic magnitude of the quality of the lens is carried out for angle values ​​close to the values ​​of the natural angles Θk, under the control of an increase in the characteristic magnitude of the quality of the lens. For example, a Newton gradient type method can be used.

[0042] The lenses produced according to the present invention can in particular be used to produce camera modules for smartphones, digital tablets, computers, webcams, etc.

[0043] The lenses produced according to the present invention can also be used in the mass production of camera modules for use in automobiles, for example as driver assistance or to power autonomous piloting systems, or for taking images or piloting flying systems such as airplanes, drones, helicopters, etc.

[0044] The lenses produced according to the present invention can also be used to equip camera modules used in medicine, such as those used in endoscopes, or even camera modules for video surveillance.

[0045] Of course, these application examples are in no way limiting. List of figures

[0046] Other features and advantages of the invention will emerge from the detailed description of a non-limiting embodiment, and from the appended drawings in which: - [Fig. 1] is a schematic view along a longitudinal section of an optical module according to an embodiment of the invention. - [Fig. 2] is a schematic top view of a lens of an optical module according to an embodiment of the invention. - [Fig. 3] is a schematic view along a longitudinal section of an optical component of the lens type according to an embodiment of the invention. - [Fig. 4] is an explanatory diagram relating to an embodiment of the invention. Description

[0047] Figure 1 represents a schematic view along a longitudinal section of an optical module 1 intended to equip an optical lens.

[0048] Said module has an optical axis (z) as well as a reference plane (x, y) perpendicular to the optical axis, and comprises a combination of N types of optical lenses Lk, k = [1, … N] mounted in a barrel 10.

[0049] the optical module may also include one or more spacers Si, of variable thickness ei, i = [1, … E], intended to be interposed between some of the N types of lenses.

[0050] As shown in Figure 2, each of the interfaces of the lenses Lk has an axisymmetric profile of the aspherical type and more particularly a substantially flat peripheral circular zone MECk,f defining a support center CMk,f and intended to be in support when said lenses are mounted in the barrel, as well as a central zone OPTk,f, linked to the optical function of each lens Lk and whose apparent center of each interface defines what we call an apex COk,f, where f takes the values ​​1 or 2 for each interface. Let us recall that the optimization in question in the method aims to align these apices COk,f as best as possible on the same axis which is also the optical axis of the objective.

[0051] The method according to the invention comprises a set of steps aimed at determining the optimal angular orientation of a subset of n types of optical lenses Lk, chosen from the N types of optical lenses.

[0052] Said method firstly comprises a step in which the N types of optical lenses Lk are mounted in the barrel in a reference angular position Θ0. This reference position may consist of the juxtaposition of a visual mark made on the barrel with a visual mark made on the optical lenses Lk.

[0053] As shown in Figure 4, the method then comprises a step in which n types of lenses are selected from the N types of lenses, and for the subset of n types of lenses, by rotating them on themselves around the optical axis (z) passing through their respective center of support CMk,f, the equation of the n circles C-Shiftk,f projected in the reference plane (x, y), representing the displacement of their apex COk,f around the optical axis (z) passing through their respective center of support CMk,f is determined. Indeed, the apex COk,f is generally slightly offset from the center of support CMk,f, among other things by the effects of thermal shrinkage during molding.

[0054] The subset of n types of lenses among the N types of lenses can take the form for example of the subset for which k is even, or odd, or even equal to 1 or N, or even of types of lenses whose shape makes any decentring more sensitive on the global optical function than for others. Of course, taking into account the number reduced number of assemblies to be tested, it is entirely possible to take n = N to obtain the best possible quality at the end of the process, which does not make the process any further to carry out physically.

[0055] According to the principle of determining a point of reduced distance DR, the method then comprises a step in which the coordinates (xDR, yDR) of the point of reduced distance of the trajectories of the set of 2n circles C-Shiftk,f, projected into the reference plane (x, y) are determined by calculation.

[0056] The coordinates (xDR, yDR) of the reduced distance point (DR) can for example be determined by the Leibniz vector function associated with a list of proximity or intersection points between trajectories (CIk,f, pk), k = [1, … n], pk representing the weighting of each C-Shiftk,f circle as a function of the sensitivity of the index interface k, f on the global optical function.

[0057] Then, the method includes a step in which we calculate for each circle C-Shiftk,f the coordinates of the point Mk,f of the circle C-Shiftk,f which is closest to the point at reduced distance.

[0058] The reduced distance point DR can be calculated by the following procedure:

[0059] (a) Calculate the intersection points of each pair (in the sense of the two interfaces f) of two trajectories among the n trajectories, for each set of two lenses. We obtain a set of a generally even number of points (except in the case of tangent circles), between zero and eight points. If none of these circles intersect, we will take a single point located at an equal distance between the closest circles. We obtain for example A(2,n) (binomial coefficients number of ways to take 2 elements among n) sets of 8 points.

[0060] (b) We take all pairs of 2 games from the previous A(2,n). (There are A(2, A(2,n)) games to consider). For each pair of games of up to eight points, eliminate the points that are furthest from the other points.

[0061] Refinements are possible in steps (a) and (b), for example by applying more global methods of choosing points to keep, or by repeating the choices along different paths in the lists of points.

[0062] (c) Calculation for example of the barycenter of the set of points remaining in steps (a) and (b). This barycenter is considered as a point at reduced distance DR with respect to the set of circles.

[0063] Finally, the method includes a step in which the rotation angle Θk is calculated for the n types of lenses, relative to the reference angular position Θ0, corresponding to the positioning of the apices COk,f closest to the point DR.

[0064] Thus, for each lens in the subset of n types of optical lenses Lk, we defined the rotation angle Θk to be applied to best realign the 2n apices COk,f.

[0065] This process assumes that the plane of each lens is parallel to the reference plane (x, y), which is not entirely correct since the plane of each lens can be slightly inclined along the (y) axis, which generates an angle called "tilt" with the optical axis (z) as shown in Figure 3, or inclined along the (x) axis, which generates an angle called "tip" with the optical axis (z).

[0066] Thus, when rotating the lenses, the trajectories C-Shiftk,f projected into the reference plane (x, y), representing the trajectory of the displacement of their apex COk,f when rotating around the optical axis (z) passing through their respective center of support CMk,f, are in fact imperfect.

[0067] Preferably, after the orientation of the n types of lenses according to their own angle Θk, a step is carried out in which a characteristic quantity of the quality of the objective is measured, obtained with the combination of the N types of lenses including the n types of lenses oriented according to their own angle Θk.

[0068] We can thus proceed to a step of comparison of the measurement of the characteristic magnitude of the quality of the objective obtained with the combination of the N types of lenses including the n types of oriented lenses according to their own angle Θk, with a first predefined threshold FTMmin1, in order to validate or not the mounting of the lenses on the barrel, and if necessary trigger the sixth aspect of the invention, or even search for another point of reduced distance DR, and another set of angles Θk providing a better solution.

[0069] Advantageously, the characteristic quantity of the quality of the objective is the value of the Modulation Transfer Function (MTF).

[0070] The optical transfer function or OTF of an optical system is a complex function that relates the luminance of the object space to the illumination of the image space. It allows us to model the influence of the optical system on the distribution of light energy in the image space.

[0071] The optical transfer function is often considered only in the conjugate object and image planes but is three-dimensional in the general case. This complex function can be decomposed into an amplitude called the modulation transfer function and a phase called the phase transfer function.

[0072] The modulation transfer function or MTF is therefore a function that characterizes the capacity of the optical system to restore contrast as a function of the fineness of the details of the object; in other words, its capacity to transmit the spatial frequencies of the object. It is used to evaluate the quality of the optical system, particularly in photography and cinematography.

[0073] The phase transfer function characterizes the phase shifts introduced by the optical system.

[0074] According to a preferred embodiment of the invention, the step of determining the circles C-Shiftk,f representing the displacement of the apexes COk,f comprises: - a sub-step in which the coordinates of the apex COk,f projected in the reference plane (x, y) and in the reference angular position Θ0 are measured for each of the n types of lenses, - a sub-step in which a rotation of 180 degrees around the optical axis (z) passing through its support center CMk,f is carried out for each of the n types of lenses so as to place them in the angular position Θ180,

[0075] More precisely, we can achieve objective No. 1 with an absence of rotation (called case Θ0) on the n lenses.

[0076] We can realize a second objective No. 2 with a rotation Θ180 on all the lenses Lk for k even among the n lenses, leaving at Θ0 the lenses of k odd.

[0077] We can also make a third objective No. 3 with a rotation Θ180 on all the lenses Lk for k odd among the n lenses, leaving at Θ0 the lenses of k even.

[0078] The position of the apparent centers COk,f of each lens interface of each assembly of these three objectives is measured.

[0079] In the more complex case where the centers of the trajectories can depend on the rotation of the adjacent lenses, the objectives N°2 and N°3 are themselves realized in several objectives N°2a, 2b, 2c, 2d, 3a, 3b, 3c, 3d: The lenses of the objectives N°2 a to d are rotated to Θ180 for the lenses of even k, and: - N°2a the lenses Lk of odd k are rotated by Θ0. - N°2b the lenses Lk of odd k and k ^ 1 [4] (that is to say k congruent to 1 modulo 4, that is k = 1, 5, 9 …) are rotated by Θ0, and where the lenses of k ^ 3 [4] (k = 3, 7 …) are rotated by Θ180. - N°2c the lenses Lk of odd k are rotated by Θ90. - N°2d the lenses Lk of odd k and k ^ 1 [4] (k =1, 5 …) are rotated by Θ90, and where the lenses of k ^ 3 [4] (k = 3, 7 …) are rotated by Θ270. Lenses N°3 a to d are rotated to Θ180 for the lenses of odd k, and: - N°3a the lenses Lk of even k are rotated by Θ0.- N°3b the lenses Lk of k even and k ^ 2 [4] (k = 2, 6 …) are rotated by Θ0, and where the lenses of k ^ 4 [4] (k = 4, 8 …) are rotated by Θ180. - No. 3c the lenses Lk of k even are rotated by Θ90. - No. 3d the lenses Lk of k even and k ^ 2 [4] (k = 2, 6 …) are rotated by Θ90, and where the lenses of k ^ 4 [4] (k = 4, 8 …) are rotated by Θ270. For example, a number of 9 objectives.

[0080] The more the dependencies of the centers of the trajectories depend on more distant adjacent lenses, the more we can increase the above combinations in order to estimate the coefficients of dependence of the centers of rotations on more distant lenses. - a sub-step in which we calculate for each of the n types of lenses the equations of the circle C-Shiftk,f projected in the reference plane (x, y), for example from the measurements made on the centers of the 3 lenses produced.

[0081] In more complex cases, for example, the 9 series of measurements on the tested objectives are analyzed in order to additionally determine the coupling coefficients of the centers of the trajectories to the adjacent lenses, or even the coupling coefficients determining the shapes of the trajectories to the adjacent lenses.

[0082] Preferably, the measurement of the apex coordinates COk,f is carried out using low coherence interferometry techniques.

[0083] Interferometry is a family of techniques in which waves, usually electromagnetic waves, are superimposed, causing the phenomenon of interference in order to extract information.

[0084] Interferometers are widely used in science and industry to measure small displacements, changes in refractive index, and surface irregularities. In an interferometer, light from a single source is split into two beams that travel different optical paths and then combined again to produce interference. The resulting interference fringes provide information about the difference in length between the optical paths. Interferometers can be used to measure the lengths and shapes of components. optics with nanometer precision; they are among the highest precision length measuring instruments.

[0085] Typically, laser diodes have a coherence length between tens of centimeters and meters, while for LEDs (Light Emitting Diodes) it varies between tens and hundreds of micrometers. Consequently, by illuminating an interferometer with a low coherence source, it is possible to identify the position for which the Optical Path Difference (OPD) is zero by looking for the position where the fringes have the best visibility. We can therefore use this technique to remove the ambiguity on the order of the fringes (i.e. real displacement of the fringe system) which is a recurring problem in interferometry. Thus the measurement is obtained without ambiguity.

[0086] In addition, the coordinates of the apexes COk,f for Θ0 and Θ180 are measured, as is the thickness of the n types of lenses at the level of said apices (COk,f). This measurement makes it possible to validate whether or not the thickness is within the prescribed tolerances.

[0087] Also optionally, in addition to the apex coordinate measurements COk,f for Θ0 and Θ180, the angle deviations of the planes of the n types of lenses in the peripheral optical zones of the lenses can also be measured. Indeed, the plane of each lens can be slightly inclined along the (y) axis, which generates a so-called "tilt" angle with the optical axis (z) as shown in Figure 3, or inclined along the (x) axis, which generates a so-called "tip" angle with the optical axis (z).

[0088] According to a particular embodiment, the optical module may comprise one or more spacers Si (for example E Spacers), of variable thickness ei, i = [1, … E], intended to be interposed between some of the N types of lenses.

[0089] In this case, when carrying out all the steps aimed at determining the optimal angular orientation for the subset of n types of optical lenses Lk, the thickness ei of the spacers (Si) can be varied with an increment on i, i varying from 1 to E, when the characteristic quantity of the quality of the lens obtained with a combination of N types of lenses including the subset of n types of oriented lenses according to their own angle Θk, is not satisfactory and needs to be optimized. This can be the case when the characteristic quantity of the quality of the objective obtained is greater than the first predefined threshold FTMmin1, but less than a second predefined threshold FTMmin2.

[0090] Usually, each of the N types of lenses comes from a manufacturing mold Mk comprising M cavities Mk,j, j = [1, … M].

[0091] In this case, for each lens Lk from the subset of n types of optical lenses, we carry out permutations on a subset of m cavities Mk,j chosen from the M cavities Mk,j, j = [1, … M].

[0092] We can thus determine which is the best cavity Mk,j for the lens Lk oriented according to its own angle Θk, which makes it possible to optimize the characteristic quantity of the quality of the lens.

[0093] The subset of m cavities chosen from the M cavities can take the form, for example, of a subset for which the manufacturing quality is acceptable.

[0094] According to a first approach which favors a minimum quality for the objective, we determine the first combination of N types of optical lenses Lk,j including the subset of n types of optical lenses Lk,j oriented optimally according to their own angles Θk,j, and coming from a subset of m cavities Mk,j chosen among the M cavities Mk,j, which proposes a characteristic quantity of the quality of the objective, greater than a first predefined threshold (FTMmin1). Note that each angle Θ k,j depends on the choice of cavities assembled together to form each of the m objectives.

[0095] This approach makes it possible to shorten the steps and more quickly determine an acceptable recipe (or rather m acceptable recipes) for assembling the N types of lenses from each of the m mold cavities.

[0096] According to a second approach which favors an optimal quality for the objective, we determine the combination of N types of optical lenses Lk,j including the subset of n types of optical lenses Lk,j oriented optimally according to their own angles Θk,j, and coming from a subset of m cavities Mk,j chosen among the M cavities Mk,j, which offers the best characteristic quantity of the quality for each of the m objectives.

[0097] Preferably, a parameter check should be carried out for each lens Lk,j as it leaves the manufacturing mold Mk before mounting on the barrel, in order to exclude lenses whose parameters deviate too much from the specified tolerances.

[0098] These parameters can be the shape, thickness, decentrations, angular deviations from the (x,y) plane, the optical index of the lenses.

[0099] As an explanatory example, in the case where the lens is composed of 6 lenses each from a 16-cavity mold, for a first series of tests, we perform a calculation of the optimal angle for the even lenses from cavity 1, namely L2.1, L4.1 and L6.1, then we test the quality of the lens for the assembly: L1.1, L2.2, L3.1, L4.1, L5.1, L6.1 with L2.1, L4.1 and L6.1 oriented according to their respective angle.

[0100] Then we perform permutations on the cavities of the lenses L2,j, L4,j and L6,j by calculating, each time testing the quality of the objective of the assemblies, i.e.: L1,1, L2,2, L3,1, L4,2, L5,1, L6,2 L1,1, L2,3, L3,1, L4,3, L5,1, L6,3 L1,1, L2,4, L3,1, L4,4, L5,1, L6,4 … L1,1, L2,16, L3,1, L4,16, L5,1, L6,16

[0101] For a second series of tests, we calculate the optimal angle for the odd lenses from cavity 1, namely L1.1, L3.1 and L5.1, then we test the quality of the lens for the assembly: L1.1, L2.2, L3.1, L4.1, L5.1, L6.1 with L1.1, L3.1 and L5.1 oriented according to their respective angle.

[0102] Then we carry out permutations on the cavities of the lenses L1,j, L3,j and L5,j, testing each time the quality of the objective of the assemblies, that is: L1.2, L2.1, L3.2, L4.1, L5.2, L6.1 L1.3, L2.1, L3.3, L4.1, L5.3, L6.1 L1.4, L2.1, L3.4, L4.1, L5.4, L6.1 … L1.16, L2.1, L3.16, L4.1, L5.16, L6.1

[0103] The analysis of the two series makes it possible to obtain a comparative model of the relative movement of the apices between the assembled cavities.

[0104] This gives more degrees of freedom to find a better solution.

[0105] It is also possible to construct an analytical model of the relative positions of the apices in displacement and angle between the cavities, by this method, in order to test a wide variety of possible assemblies not physically realized, but realizable virtually, in order to obtain a very efficient optimal solution for assembling for example 16 cavities of each type of lenses, for the N lenses.

[0106] This method can allow testing a relatively still limited number of permutations, in order to mathematically calculate many more permutations which would exceed a reasonable number of assemblies to be made and tested. Nomenclature 1 optical module 10 barrel Lk optical lens of rank k Si spacer COk,f apex of the optical lens of rank k, interface f CMk,f center of support of the optical lens of rank k, interface f CIk,k' point of intersection OPTk,f central zone of the optical lens of rank k, interface f MECk,f peripheral circular zone of the optical lens of rank k, interface f C-Shiftk,f trajectory of the apex of rank k, interface f Ik,f point of the trajectory C-Shiftk,f closest to the point of reduced distance, interface f DR Point of reduced distance of the set of circles C-Shiftk,f Θk rotation angle of the optical lens of rank k Θk,j rotation angle of the optical lens of rank k, in the context of m objectives made in parallel (x, y) plane perpendicular to the axis (z), called reference plane (z) stacking axis, called optical axis (Mk) mold for manufacturing the optical lens of rank k (Mk,j) cavity of rank j of the mold for manufacturing the optical lens of rank k

Claims

Claims 1. Method for determining an assembly recipe for a combination of N optical elements (Lk), k = [1, … N], such as lenses, and forming an optical objective when said optical elements are stacked in a barrel (10) in a stacking direction (z), at least one of the optical elements comprising two optical interfaces f, f = [1, 2], each comprising an apex (COk,f), said method comprising for each of the optical elements of a subset of n optical elements of said combination of N optical elements, the following steps: - determining a trajectory (C-Shiftk,f) of the apex (COk,f) for each optical interface of said optical element, in the plane (x, y) perpendicular to said stacking direction (z), from at least two angular orientations of said optical element around said stacking direction, - determining the coordinates of the apex (COk,f) for each optical interface of said element optics,in the plane (x, y) perpendicular to said stacking direction (z), verifying a reduced distance compared to the trajectories of the other optical elements of the subset of n optical elements, and - determination of an angular orientation Θk of said optical element in which the apices (CO, k,f ) of the two interfaces of said optical element are closest to the trajectories of the other optical elements of the subset of n optical elements.

2. Method for determining an assembly recipe according to claim 1, in which the optical elements are lenses having an axisymmetric profile of the aspherical type and more particularly a substantially flat peripheral circular zone (MECk,f) defining a support center (CMk,f), characterized in that said method comprises a set of steps aimed at determining the optimal angular orientation Θk of a subset of n types of optical lenses (Lk), chosen among the N types of optical lenses, namely: - a step in which the N types of optical lenses (Lk) are mounted in the barrel in a reference angular position Θ0, - a step in which the equation of the trajectories (C-Shiftk,f) projected in the plane (x, y) is determined for the subset of n types of lenses chosen from the N types of lenses and by rotation of the n types of lenses around the axis (z) passing through their center of support (CMk,f), representing the displacement of their apex (COk,f) around the axis (z) passing through their center of support (CMk,f), - a step in which the coordinates (xDR, yDR) of a point of reduced distance (DR) are determined relative to the set of trajectories (C-Shiftk,f), projected in the plane (x, y), - a step in which the coordinates (xDR, yDR) of a point of reduced distance (DR) are calculated for each trajectory (C-Shiftk,f) from point Ik,f of this trajectory (C-Shiftk,f) which is closest to the reduced distance point (DR),- a step in which the rotation angle Θk is calculated for the n types of lenses with respect to the reference angular position Θ0, corresponding to the positioning of the apex (COk,f) at the point (Ik,f).

3. Method for determining an assembly recipe according to claim 1 or 2, in which a step is carried out in which a characteristic quantity of the quality of the lens is measured, obtained with the combination of the N types of lenses including the n types of lenses oriented according to their own angle Θk.

4. Method for determining an assembly recipe according to claim 3, in which a series of measurements of the characteristic quantity of the quality of the lens is carried out for angle values ​​close to the values ​​of the own angles Θk, and the value of the own angles Θk is optimized under the control of an increase in the characteristic quantity of the quality of the lens., 5. Method for determining an assembly recipe according to claim 4 or 5, in which a step is carried out of comparing the measurement of the characteristic quantity of the quality of the lens obtained with the combination of the N types of lenses including the n types of lenses oriented according to their own angle Θk, with a first predefined threshold (FTMmin1).

6. Method for determining an assembly recipe according to one of claims 3 to 5, in which the characteristic quantity of the quality of the lens is the value of the Modulation Transfer Function (FTM). 7.Method for determining an assembly recipe according to any one of the preceding claims, in which the trajectories (C- Shiftk,f) representing the displacement of the apices (COk,f) are circles and the step of determining the trajectories comprises: - a sub-step in which the coordinates of the apex (COk,f) projected in the reference plane (x, y) and in the reference angular position Θ0 are measured for each of the n types of lenses, - a sub-step in which a rotation of 180 degrees around the optical axis (z) passing through its center of support (CMk,f) is carried out for each of the n types of lenses so as to place them in the angular position Θ180, - a sub-step in which the coordinates of the apex (CO. k,f) projected into the reference plane (x, y), - a sub-step in which the equations of the circle (C-Shiftk,f) projected into the reference plane (x, y) are calculated for each of the n types of lenses.

8. Method for determining an assembly recipe according to the preceding claim, in which the measurement of the coordinates of the apexes (COk,f) is carried out using low coherence interferometry techniques.

9. Method for determining an assembly recipe according to any one of the preceding claims, in which, in addition to the measurements of the coordinates of the apexes (COk,f) for several angles of rotation, the thickness of the n types of lenses at the level of said apices (COk,f) is measured.

10. Method for determining an assembly recipe according to the preceding claim, in which, in addition to the measurements of the coordinates of the apexes (COk,f) for several angles of rotation, the angle deviations of the planes of the n types of lenses in the peripheral optical zones of the lenses are measured. 11.Method for determining an assembly recipe according to any one of the preceding claims, together with claim 3, in which the optical module comprises one or more spacers (Si), of variable thickness (ei), i = [1, … E], intended to be interposed between some of the N types of lenses and in which the set of steps is carried out to determine the optimal angular orientation of each of the n types of optical lenses (Lk), with an increment on i, i varying from 1 to E, to vary the thickness (ei) of the spacers (Si), to increase the characteristic magnitude of the quality of the lens obtained with a combination of N types of lenses including the subset of n types of lenses oriented according to their own angle Θk. 12.Method for determining an assembly recipe according to any one of the preceding claims, in which each of the N types of lenses comes from a manufacturing mold (Mk) comprising M cavities (Mk,j), j = [1, … M], characterized in that all the steps are carried out to determine the optimal angular orientation Θk,j for the subset of n types of lenses chosen from among the N types of lenses, and coming from a subset of m cavities (Mk,j) chosen from among the M cavities (Mk,j).

13. Method for determining an assembly recipe according to the preceding claim, in which permutations are carried out on the cavities, to increase the characteristic quantity of the quality of the lens produced, these permutations being carried out in the calculation steps or during. of the mounting of the lenses in the barrel.

14. Method for determining an assembly recipe according to claim 12 or 13, in which the first combination of N types of optical lenses (Lk,j) is determined, including the subset of n types of optical lenses (Lk,j) optimally oriented according to their own angles Θk, and originating from a subset of m cavities (Mk,j) chosen from among the M cavities (Mk,j), which proposes a characteristic quantity of the quality of the lens, greater than a first predefined threshold (FTMmin1). 15.Method for determining an assembly recipe according to claim 12, in which the combination of N types of optical lenses (Lk,j) is determined, including the subset of n types of optical lenses (Lk,j) optimally oriented according to their own angles Θk,j, and originating from a subset of m cavities (Mk,j) chosen from among the M cavities (Mk,j), which offers the best characteristic quantity of the quality of the lens.

16. Optical module (1) obtained from the implementation of a method according to any one of the preceding claims, together with claim 6, the optical module having a Modulation Transfer Function (MTF) greater than a first predefined threshold (MTFmin1).

Images