IPR2020-00897
`Apple Inc. v. Corephotonics, Ltd.
`
`U.S. Patent No. 10,324,277
`
`IPR2020-00897 | SLIDE 1
`
`

`

`Ins$tuted Grounds
`
`• Ground 1: Claims 1-3 and 5-8
`Obviousness over Ogino Example 4 and Bareau
`
`• Ground 2: Claims 1-24
`Obviousness over Ogino Example 5 and Bareau
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`IPR2020-00897 | SLIDE
`
`2
`
`

`

`Overview of Argument
`• All Ins’tuted Grounds: Pe’’on Uses Improper Hindsight
`•
`Improperly uses ‘277 patent as a guide to obviousness
`• No mo8va8on to modify Ogino with Bareau
`• Overlapping lens elements ignore manufacturability
`• Ground 1:
`•
`Ignores Bareau Teaching of Rela8ve Illumina8on
`• Ground 2:
`• Hindsight analysis ignores standard industry design prac8ces
`•
`Inconsistent Characteriza8on of Ogino Example 5
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`IPR2020-00897 | SLIDE
`
`3
`
`

`

`Pe##oner Improperly Uses the
`‘277 Patent as a Guide
`
`IPR2020-00897 | SLIDE 4
`
`

`

`Dr. Sasian Designed Lenses Based on ‘277 Patent Claims
`Q: So once you found a spacing between L3
`and L4 that met the claim limita;ons, you just
`stopped; right?
`MS. SIVINSKI: Objec;on. Misstates tes;mony.
`A: I wouldn't characterize like that. I apply the
`teaching of Ogino and change the spacing and
`have probably -- probably three different
`solu;ons, and one of them was within the
`range of -- of -- in thickness as specified in
`the -- in the '277 Patent. The others, maybe
`they weren't within the range of the '277
`Patent.
`Ex. 2003, Febraury 19, 2021, Sasian Dep. Tr., 171:2-13.
`
`Dr. Jose Sasian
`Pe..oner’s Expert
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`IPR2020-00897 | SLIDE
`
`5
`
`

`

`
`
`Declaration of José Sasián, Ph.D. in Support of Petitioner Reply
`
`of keeping these spaces constant, “a POSITA would have allowed spacings
`
`between lens surfaces to vary” because “[t]his would permit better performance to
`be obtained during the design process.” Id., ¶¶99-100. However, Dr. Milster does
`A POSITA Would Use Industry Design Standards, Not the Patent
`not provide any example of how Ogino’s Example 4 lens design could have been
`
`• A POSITA Would
`Not Look at ‘277
`Patent
`
`• A POSITA Would
`Not Look at Prior
`Art Patent Claims
`
`improved by varying spacing between lens surfaces, instead relying on the bare
`
`assertion that prohibiting the lens spacings to vary “might have prevented a
`
`POSITA from finding the best performance result.” Ex. 2001, ¶100. Further, a
`
`POSITA would have known that releasing too many variables for optimization
`
`often leads to drastic changes to the lens structure and would have first made
`
`minimum changes to a lens to maintain the lens within the scope of a patent.
`
`6.
`
`Keeping certain variables constant, such as spacing between lenses,
`
`while varying other parameters, is precisely the approach a POSITA would have
`
`taken. See APPL-1017, p.168 (stating that after entering the lens design to be
`
`improved into a design computer program, “each variable is changed a small
`
`amount, called an increment, and the effect to performance is then computed”). Dr.
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`IPR2020-00897 | SLIDE
`6
`Milster testified that he took a similar gradual “step-wise process” in modifying
`
`lenses. APPL-1028, 21:6-18. This is also the same process that Patent Owner’s
`
`expert Dr. Moore described when he was deposed in earlier, related proceedings
`
`involving patents in the same family. APPL-1023, 99:6-18 (stating that variables in
`
`

`

`Grounds 1 and 2: No Mo#va#on
`to Modify Ogino with Bareau
`
`IPR2020-00897 | SLIDE 7
`
`

`

`with a concave object - side surface marked 308a and an
`thickness ratio L11 / Lle of 3 . 08 .
`image - side surface marked 308b ; and a fifth plastic lens
`While this disclosure has been described in terms of
`element 310 with negative refractive power having a nega
`certain embodiments and generally associated methods ,
`tive meniscus , with a concave object - side surface marked 15 alterations and permutations of the embodiments and meth
`310a and an image - side surface marked 310b . The optical
`ods will be apparent to those skilled in the art . The disclosure
`lens system further comprises an optional glass window 312
`is to be understood as not limited by the specific embodi
`disposed between the image - side surface 310b of fifth lens
`ments described herein , but only by the scope of the
`element 310 and an image plane 314 for image formation of
`appended claims .
`What is claimed is :
`20
`Claims 1, 11 and 18 Require Fno. < 2.9
`In embodiment 300 , all lens element surfaces are
`1 . A lens assembly , comprising : a plurality of refractive
`aspheric . Detailed optical data is given in Table 5 , and the
`lens elements arranged along an optical axis , wherein at least
`aspheric surface data is given in Table 6 , wherein the
`one surface of at least one of the plurality of lens elements
`markings and units are the same as in , respectively , Tables
`is aspheric , wherein the lens assembly has an effective focal
`1 and 2 . The equation of the aspheric surface profiles is the 25 length ( EFL ) , wherein a lens system that includes the lens
`same as for embodiments 100 and 200 .
`assembly plus a window positioned between the plurality of
`lens elements and an image plane has a total track length
`( TTL ) of 6 . 5 millimeters or less , wherein a ratio TTL / EFL
`is less than 1 . 0 , wherein the plurality of lens elements
`comprises , in order from an object side to an image side , a
`first lens element with positive refractive power , a second
`lens element with negative refractive power , and a third lens
`element , wherein a focal length f1 of the first lens element
`is smaller than TTL / 2 and wherein a lens assembly F # is
`smaller than 2 . 9 .
`2 . The lens assembly of claim 1 , wherein the third lens
`element has negative refractive power .
`3 . The lens assembly of claim 1 , wherein the plurality of
`refractive lens elements includes five lens elements .
`4 . The lens assembly of claim 1 , wherein the focal length
`fl , a focal length f2 of the second lens element and a focal
`length f3 of the third lens element fulfill the condition
`1 . 2x | f31 > | f2 > 1 . 5xf1 .
`
`1 . 5148 / 63 . 1
`1 . 63549 / 23 . 91
`1 . 5345 / 57 . 09
`1 . 63549 / 23 . 91
`1 . 5345 / 57 . 09
`1 . 5168 / 64 . 17
`
`3
`
`40
`
`IPR2020-00897 | SLIDE 8
`
`Nd / Vd
`
`Diameter
`[ mm ]
`2 . 4
`
`2 . 1
`1 . 8
`1 . 8
`1 . 7
`3 . 4
`4 . 0
`4 . 4
`3 . 0
`3 . 0
`
`

`

`to design “faster” lenses for brighter images).
`
`53. Thus, a POSITA designing lens assemblies for use in a cell phone
`
`would have been informed by Bareau to design or modify a lens system that fit
`
`within the specifications of including maintaining a short track length, an f-number
`Ground 1: Sole Motivation To Decrease Fno. < 2.9 is to Make Brighter
`of 2.8 or lower for ¼” and smaller pixel sensor formats, and relative illumination
`greater than 50 percent.
`
`3.
`
`Reasons to combine Ogino and Bareau – (I)
`
`54.
`
`It is my opinion that a POSITA would have found it obvious to
`
`modify Ogino’s Example 4 lens assembly in view of Bareau’s specifications for
`
`cell phone camera lenses with an f-number of 2.8 or less for ¼” and smaller pixel
`
`image sensors. Such a combination would have been nothing more than applying
`
`Bareau’s specification for a brighter lens system for smaller pixel image sensors,
`
`according to known lens design and modification methods (as taught in APPL-
`
`1017, p.172), to yield a predictable result of Ogino’s Example 4 lens assembly
`Ex. 1003 at ¶ 54.
`
`likewise supporting an f-number of 2.8 or lower. See id., pp.3-4.
`
`Apple v. Corephotonics
`
`28
`
`APPL-1003
`
`IPR2020-00897 | SLIDE 9
`
`

`

`Sasián Decl.
`
`Inter Partes Review of U.S. 10,324,277
`
`67.
`Included in Table 9 is “f = 5.956,” “Bf = 2.438,” and “TL = 5.171.”
`The total track length with the optical member CG for Example 5 can be calculated
`Ground 2: Sole Mo$va$on To Decrease Fno. < 2.9 is to Make Brighter
`by summing the widths D1 to D13 and is 5.273 mm.
`
` Reasons to combine Ogino and Bareau – (II)
`
`2.
`It is my opinion that a POSITA would have found it obvious to
`
`68.
`
`modify Ogino’s Example 5 lens assemblies in view of Bareau’s specifications for
`
`cell phone camera lenses with an f-number of 2.8 or less for ¼” and smaller pixel
`
`image sensors. Similar to the discussion of Ogino’s Example 4 above, such a
`
`combination would have been nothing more than applying Bareau’s specification
`
`for a brighter lens system for smaller pixel image sensors, according to known lens
`
`design and modification methods (as taught in APPL-1017, p.172), to yield a
`Ex. 1003 at ¶ 68.
`predictable result of Ogino’s Example 5 lens assemblies likewise supporting an f-
`
`number of 2.8 or lower. See id., pp.3-4. Thus, modifying Ogino’s Example 5 lens
`
`IPR2020-00897 | SLIDE 10
`
`assemblies to have an F number of 2.8, as taught in Bareau, would have been
`
`nothing more than applying Bareau’s specification of an F# of 2.8 for a ¼” image
`
`sensor format according to known lens design methods (as taught in Fischer
`
`

`

`
`
`
`
`
`
`
`
`
`
`
`% 0 | -
`
`U.S. Patent
`
`U.S. Patent
`
`Sep. 8, 2015
`
`Sep. 8, 2015
`
`Sheet 8 of 14
`
`IPR2020-00897 | SLIDE
`
`11
`
`Sheet 9 of 14
`
`US 9,128.267 B2
`
`% 0 | -
`
`Uu??
`00||O UU 77 00 | -
`
`HNIT-p–
`
`% 0 | -
`
`APPL-1005 / Page 13 of 28
`APPLE INC. v. COREPHOTONICS LTD.
`
`No Motivation to Modify Ogino Examples 4 and 5 With Bareau:
`Ogino Contains Four Examples with Fno. < 2.9
`• Ogino Example 4
`
`
`
`
`• Ogino Example 5
`
`
`
`
`• Apple Would Modify to Fno.=2.8 Based on Bareau Despite Four
`Other Examples in Ogino having an Fno.<2.8
`
`Lu 77 00||O
`UUTÍ
`OO|-
`
`APPL-1005 / Page 12 of 28
`APPLE INC. v. COREPHOTONICS LTD.
`
`APPL-1005 / Page 11 of 28
`APPLE INC. v. COREPHOTONICS LTD.
`
`APPLE INC. v. COREPHOTONICS LTD.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`

`

`No Mo$va$on to Modify Ogino Examples 4 and 5 With Bareau
`– POSITA mo*vated to have fno. = 2.8 as taught by
`Bareau would have made a small modifica*on to
`one of Ogino Examples 1-3 and 6
`
`– Would not have made the large modifica*ons to
`the fno. 3.04 in Ogino Example 4 or fno. = 3.94 in
`Ogino Example 5
`
`IPR2020-00897 | SLIDE 12
`
`

`

`Grounds 1 and 2: Pe##oner
`Ignores Manufacturability
`
`IPR2020-00897 | SLIDE13
`
`

`

`A POSITA Would Consider Manufacturability
`A POSITA “would have had experience in
`analyzing, tolerancing, adjus$ng, and
`op$mizing mul$-lens systems for
`manufacturing, and would have been
`familiar with the specifica$ons of lens
`systems and their fabrica$on.”
`
`Ex. 1003, SasianDecl. at ¶ 19.
`
`Dr. Jose Sasian
`Petitioner’s Expert
`
`IPR2020-00897 | SLIDE 14
`
`

`

`Board Has Already Spoken About Manufacturability Considera$ons:
`We disagree that a person having ordinary skill in op3cal lens
`design at the 3me of the ’568 patent would not consider ‘the
`limits of fabrica3on’ such as those discussed in Beich,
`par3cularly in light of Beich’s disclosure that “it is important
`that the designer has a basic understanding of the
`manufacturing process and of the limits of size and tolerances
`that might be expected of the finished op3cs.”
`
`IPR2019-00030, Paper No. 32, Final WriZen Decision, at
`44 (quo;ng Ex. 1020 (Ex. 1007 of the present IPR) at 7)
`
`IPR2020-00897 | SLIDE 15
`
`

`

`Beich (Ex. 1007) Teaches Against Lenses Effec$vely Touching
`Polymer Optics: A manufacturer’s perspective on the factors that
`contribute to successful programs
`
`• Manufacturing
`tolerances would
`not allow lenses to
`be posi’oned
`closer than 40
`microns (0.020 mm
`x 2)
`
`Ex. 1007 at 7.
`
`accumulates at the end of the screw it is injected at an appropriate speed and pressure into the mold. This causes the
`material to flow into the mold to fill the cavities. The molding machine provides complete control over this process,
`governing the size of the shot, injection speed, injection pressure, backpressure, cushion, and other critical variables that
`will determine the final outcome of the optic. After an appropriate cooling time, the moveable platen moves away from
`the fixed platen, and the mold opens. This allows the optics (still attached to the runner system) to be removed. After
`the shot is removed, the cycle starts over again.
`
`Other equipment is often found along side the molding machine. For parts that require a large amount of material, auto
`loading hoppers are used to feed material into the machine. Also, the thermoplastics must be dried before being fed into
`the injection unit. It is common to see desiccating equipment located near the press for this purpose. Once the molding
`cycle is completed it is desirable to promptly remove the shot so that the entire molding process may be repeated with
`regularity. To aide in this, a robotic arm is frequently used to ensure that the removal is done on time. This enables the
`entire process to go into a steady state. Depending on the nature of the program, additional automation or end of arm
`tooling may be required to remove of the parts from the press, degate them from the runner, and package them into trays
`for final shipment. Degating is the process whereby the optical elements themselves are removed from the runner
`system.
`
`3. TYING IT ALL TOGETHER
`
`As noted above, it is important that the designer has a basic understanding of the manufacturing process and of the
`limits of size and tolerances that might be expected of the finished optics. In general terms, overall shape and
`tolerances of the optic will drive cost and manufacturability. There are some general guidelines: thicker parts take
`longer to mold than thinner parts. Optics with extremely thick centers and thin edges are very challenging to mold.
`William S. Beich*a, Nicholas Turnera
`Negative optics (thin centers with heavy edges) are difficult to mold. Optics with very tight tolerances may not be
`aG-S Plastic Optics, 408 St. Paul Street, Rochester, NY 14605
`manufacturable at all in a one cavity mold, much less in a mold with more than one cavity. There are some other
`general tolerances that can describe the limits of fabrication in an ideally designed optic.
`
`
`
`
`
`
`
`Attribute
`Radius of Curvature
`EFL
`Center Thickness
`Diameter
`Wedge (TIR) in the Element
`S1 to S2 Displacement (across the parting line)
`Surface Figure Error
`Surface Irregularity
`Scratch-Dig Specification
`Surface Roughness (RMS)
`Diameter to Center Thickness Ratio
`Center Thickness to Edge Thickness Ratio
`Part to Part Repeatability (in a one cavity mold)
`
`Table 2. Rules of thumb.
`
`Rules of Thumb Tolerances
`± 0.50%
`± 1.0%
`± 0.020mm
`± 0.020mm
`< 0.010mm
`< 0.020mm
`(cid:1) 2 fringes per 25.4mm (2 fringes = 1 wave @ 632nm)
`(cid:1) 1 fringes per 25.4mm (2 fringes = 1 wave @ 632nm)
`40-20
`(cid:1) 100 Å
`< 4:1
`< 3:1
`< 0.50%
`
`IPR2020-00897 | SLIDE 16
`
`Copyright 2010 Society of Photo-Optical Instrumentation Engineers. This paper was published in Polymer Optics
`Design, Fabrication, and Materials, edited by David H. Krevor, William S. Beich, Proceedings of SPIE Vol. 7788, and
`is made available as an electronic reprint with permission of SPIE. One print or electronic copy may be made for
`personal use only. Systematic or multiple reproduction, distribution to multiple locations via electronic or other means,
`Proc. of SPIE Vol. 7788 778805-6
`duplication of any material in this paper for a fee or for commercial purposes, or modification of the content of the
`paper are prohibited.
`
`APPL-1007 / Page 7 of 10
`
`

`

`greater than the maximum tolerance, which would be 0.006 mm added to
`
`0.004 mm, or an absolute minimum of 0.010 mm, notwithstanding other pos-
`
`sible tolerances associated with wedge, changes in lens size due to tempera-
`
`ture variations and other factors. A POSITA would understand that a lens
`
`assembly configuration with lenses that are overlapping or touching in the
`Beich Teaches Tolerances of ± 20 microns
`area that optical rays pass through would never be acceptable for an optical
`
`design, let alone for manufacturing, tolerances or desensitization. Such a lens
`
`assembly design would have been rejected immediately.
`
`123. As a further example of tolerancing issues, the Beich paper cited by Dr.
`
`Sasián provides “rules of thumb” for part tolerances of “± 0.020 mm” for cen-
`
`ter thickness and diameter, “< 0.020 mm” for the “S1 to S2 Displacement,”
`
`i.e., the displacement between the halves of the mold forming the front and
`
`back surfaces of the lens, and “< 0.010 mm” for “wedge,” i.e., the difference
`
`Ex. 2001, ¶ 95.
`
`71
`
`Dr. Tom Milster
`Patent Owner’s Expert
`
`APPLE V. COREPHOTONICS
`IPR2020-00896
`Exhibit 2001
`Page 74
`
`IPR2020-00897 | SLIDE 17
`
`

`

`McGuire (Ex. 2006) Teaches Against Lenses Effectively Touching
`PROCEEDINGS OF SPIE
`
`
`
`SPIEDigitalLibrary.org/conference-proceedings-of-spie
`
`
`Approaching direct optimization of
`as-built lens performance
`
`McGuire, James, Kuper, Thomas
`
`
`
`Table 4. Typical mobile phone camera specifications
`Parameter
`Value
`Wavelengths (nm)
`650, 610, 555, 510, 470 with
`2:4:8:2:1 weights
`4.24
`2.4
`5.72
`68
`1.4
`none
`> 44
`< 1
`< 29 (to match sensor)
`< 4.9
`
`Focal length (mm)
`F-number
`Sensor (mm, diagonal)
`Full field (deg, diagonal)
`Pixel size (um)
`Vignetting
`Relative illumination (%)
`Distortion (%)
`Telecentricity (deg)
`Total track (mm)
`
`Table 5. Plastic injection molding tolerances
`Parameter
`Value
`Power (fringes)
`1.5
`Thickness (um)
`+/-3
`± 5 x 10-4
`Index
`Irregularity (fringes)
`0.5
`Wedge (um, TIR)
`1
`Decenter (um)
`+/-2
`
`James P. McGuire Jr., Thomas G. Kuper, "Approaching direct optimization of
`as-built lens performance," Proc. SPIE 8487, Novel Optical Systems Design
`and Optimization XV, 84870D (19 October 2012); doi: 10.1117/12.930568
`Event: SPIE Optical Engineering + Applications, 2012, San Diego, California,
`United States
`APPLE V. COREPHOTONICS
`IPR2020-00896
`Exhibit 2006
`Page 1
`Downloaded From: https://www.spiedigitallibrary.org/conference-proceedings-of-spie on 11 Mar 2021 Terms of Use: https://www.spiedigitallibrary.org/terms-of-use
`
`The second step was to construct an error function. While mobile phone camera MTF is typically specified and
`measured on each camera before it is shipped, we elected to use SAB and the RMS WFE rather than an MTF-based
`nominal error function due to computational efficiency. Since the area under the MTF curve is related to WFE, this
`technique works quite well for improving as-built MTF. The combination of SAB and MTF optimization is most useful
`at the very end of the design process, if one needs to improve the MTF at a particular frequency or focus position at the
`expense of the MTF at other frequencies or focus positions.
`Ex. 2006 at 13.
`
`In order to efficiently explore the design space, we used Global Synthesis with SAB, a Gaussian quadrature aperture
`sampling (7 rings and 7 spokes), and twenty points to sample the field. CODE V generated approximately 50 solutions
`over the period of one day on a 2009 vintage 8-core PC. Figure 7b shows the best design after local optimization and
`IPR2020-00897 | SLIDE 18
`substitution of real plastics for the fictitious plastics (the field weights were adjusted during the optimization to give
`more weight to the as-built performance at the edge of the field). As the aspheric orders of surfaces were progressively
`reduced by the SAB component of the merit function, we removed the higher aspheric orders from the surfaces. This is
`in contrast to Ref. 30, where the aspheric orders were explicitly targeted (through weighted constraints) in order to
`reduce the sensitivity. SAB allowed the aspheric orders to be reduced more naturally and with less intervention by the
`designer (i.e. the design progressed more quickly and a better solution was found). This process allowed us to rather
`quickly find an improved design with a negative second element. The plastics in Figure 7b are, in order, A5514,
`OKP4HT, A5514, and Zeonex 330R. Both the nominal and as-built MTF of the SAB optimized lens are significantly
`improved. The as-built plots in (e) and (f) have the meaning that when the lenses are built to the tolerances in Table 5,
`
`

`

`Case Nos. IPR2020-00897
`U.S. Patent No. 10,324,277
`
`Inter Partes Review of U.S. 10,324,277
`
`tolerances for thickness (+/- 0.003 mm) and decenter (+/- 0.002 mm) in mo-
`No Mo$va$on to Combine Because Pe$$oner’s Proposed
`bile phone cameras plastic injection molding indicate that lenses should be
`Sasián Decl.
`Case Nos. IPR2020-00897
`Combina$ons Are Not Manufacturable
`U.S. Patent No. 10,324,277
`B. Ogino Example 4 modified for F#=2.8 using Zemax (v. 02/14/2011)
`spaced greater than the maximum tolerance, which would be 0.006 mm added
`
`Fig. 2A – Ray Trace Diagram
`
`to 0.004 mm, or an absolute minimum of 0.010 mm, notwithstanding other
`
`1.
`
`possible tolerances associated with wedge, changes in lens size due to tem-
`
`perature variations and other factors. A POSITA would understand that a lens
`
`assembly configuration with lens that are overlapping or touching would
`
`never be acceptable for an optical design, let alone for manufacturing, toler-
`
`ances or desensitization. Such a lens assembly design would have been re-
`
`jected immediately.
`
`95. As a further example of tolerancing issues, the Beich paper cited by Dr.
`Ex. 2001, ¶ 94.
`Sasián provides “rules of thumb” for part tolerances of “± 0.020 mm” for cen-
`
`
`
`
`
`
`
`
`L5
`
`
`
`
`
`
`
`
`
`
`
`
`L2
`
`L1
`
`L3
`
`L4
`
`i.e., the displacement between the halves of the mold forming the front and
`
`(Ex. 1003, Sasián Decl. at 115.)
`Steps for modification:
`ter thickness and diameter, “< 0.020 mm” for the “S1 to S2 Displacement,”
`1) Open the aperture to support f-number at 2.8; FOV=+/- 32.5°;
`94. However, a POSITA would never use Modified Example 4 as it results
`2) Allow vignetting as in Ogino’s original design;
`3) Maintain original EFL, TTL, f1, f2, f3, f4, and f5;
`in two lens elements touching each other in the region of the optical rays.
`IPR2020-00896 | SLIDE 19
`4) Optimize for image quality;
`back surfaces of the lens, and “< 0.010 mm” for “wedge,” i.e., the difference
`5) Optimize radii and aspheric coefficients.
`Specifically, lens elements L4 and L5 are touching each other, as shown in
`EFL=4.543 mm, TTL=4.362, EPD=1.6228 mm, F#=ISFN=2.8; thickness and
`in edge thickness from one side of the lens to the other. (Ex. 1007, Beich at
`annotated Fig. 2A above with red boxes. According to [McGuire Jr, J. P., &
`spacing of L1-L5 remain unchanged; f1=1.824 mm; f2=-2.491 mm; f3=-72.932 mm;
`f4=3.838 mm; f5=-3.0498 mm (data calculated for standard wavelength of 587 nm).
`Kuper, T. G. (2012, October). Approaching direct optimization of as-built lens
`7.) Tolerances for glass molding are similar. (Ex. 2008, Symmons at 95.) As
`
`the Field Guide notes, “high repeatability from component to component” is
`
`performance. In Novel Optical Systems Design and Optimization XV (Vol.
`
`
`
`

`

`Case Nos. IPR2020-00896
`U.S. Patent No. 10,317,647
`
`alone for manufacturing. According to [McGuire Jr, J. P., & Kuper, T. G.
`McGuire Teaches Tolerances of at least 10 microns
`(2012, October). Approaching direct optimization of as-built lens perfor-
`mance. In Novel Optical Systems Design and Optimization XV (Vol. 8487,
`
`p. 84870D). International Society for Optics and Photonics], (Ex. 2006), tol-
`
`erances for thickness (+/- 0.003 mm) and decenter (+/- 0.002 mm) in mobile
`
`phone cameras plastic injection molding indicate that lenses should be spaced
`
`greater than the maximum tolerance, which would be 0.006 mm added to
`
`0.004 mm, or an absolute minimum of 0.010 mm, notwithstanding other pos-
`
`sible tolerances associated with wedge, changes in lens size due to tempera-
`
`ture variations and other factors. A POSITA would understand that a lens
`
`assembly configuration with lenses that are overlapping or touching in the
`Ex. 2001, ¶ 94.
`
`area that optical rays pass through would never be acceptable for an optical
`
`design, let alone for manufacturing, tolerances or desensitization. Such a lens
`
`assembly design would have been rejected immediately.
`
`123. As a further example of tolerancing issues, the Beich paper cited by Dr.
`
`Dr. Tom Milster
`Patent Owner’s Expert
`
`IPR2020-00897 | SLIDE 20
`
`

`

`Petitioner’s Reply
` IPR2020-00897 (Patent No. 10,324,277)
`
`the L4 and L5 lens elements of the modified Example 4 do not touch or overlap as
`
`
`
`
`
`
`
`shown by the zoomed in ray trace of these lens elements below. APPL-1037, ¶14
`Ground 1 – Ogino Example 4 in view of Bareau
`Q: In looking at that blowup of the ray
`trace there in Figure 3B, what is the
`distance between lenses L4 and L5 at
`that closest point?
`A: Well, I don't have that number with
`me, but it's bigger than 0.
`Q: Okay. Is it bigger -- I mean, is it 1
`millimeter?
`A: … And so the radii where the lenses
`get closer may be in the order of several
`micrometers.
`
`Ex. 2012, July 16, 2021, Sasian Dep. Tr., 25:11-26:10
`Reply at p. 9.
`(omiMng discussion of calcula.ons).
`APPL-1037, ¶14. Thus, Patent Owner’s statements regarding overlapping lenses
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`IPR2020-00897 | SLIDE
`
`are untrue.
`
`
`
`21
`
`C. Manufacturing considerations are not required by the claims nor
`can they be imported to avoid unpatentability.
`1.
`
`Patent Owner seeks to import manufacturing requirements
`
`

`

`
`
`
`
`Petitioner’s Reply
`
` IPR2020-00897 (Patent No. 10,324,277)
`Ground 2 (Claims 11-17) – Ogino Example 5 in view of Bareau
`Q: And in particular, I want you to look at
`lens elements L2 and L3. What is the
`distance between lens L2 and L3 at the
`closest point in the blown-up figure?
`A: I didn't calculate it, but I will give the
`same answer as before for the
`other case. It may be in the order of
`several micrometers.
`
`
`
`Ex. 2012, July 16, 2021, Sasian Dep. Tr., 60:20-61:2.
`
`APPL-1037, ¶41.
`
`Reply at p. 22.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`IPR2020-00897 | SLIDE
`The zoomed-in ray trace of lens elements L2 and L3 above clearly shows
`
`22
`
`space between these lens elements. APPL-1037, ¶42. Therefore, the allegation that
`
`the first modified Example 5 has overlapping lenses is without merit.
`
`
`
`

`

`in lenses with relative illumination that were not identical. (Ex. 2003, Feb. 19,
`
`2021 Dep. Tr. at 152:2-13.) Thus, lens design process that Dr. Sasián de-
`
`scribes requires a POSITA to begin at different starting points. However, he
`
`does not explain why a POSITA would select one over the other for this par-
`Ground 2 (Claims 11-17) – Ogino Example 5 in view of Bareau
`ticular design.
`115. The First Modified Example 5 would also be unacceptable to a POSITA
`
`because of the large center-to-edge thickness ratio. As shown below, a
`
`reproduction of Dr. Sasián’s First Modified Example 5 indicates an edge
`
`thickness of about 0.06444 mm. Since the center thickness of this lens is about
`
`1.12444 mm, the center-to-edge thickness ratio is about 17.42, which is 4.355
`
`times larger than the maximum center-to-edge thickness ratio as specified by
`
`Beich. As explained in the Field Guide to Molded Optics (Ex. 2009), this
`Case Nos. IPR2020-00897
`U.S. Patent No. 10,324,277
`large center-to-edge thickness ratio would increase likelyhood of flow-line
`defects, which are voids in the central portion of the molded part that are not
`
`filled
`with
`Ex. 2001, ¶ 115.
`
`the
`60
`
`lens
`
`material.
`APPLE V. COREPHOTONICS
`IPR2020-00897
`Exhibit 2001
`Page 63
`
`Dr. Tom Milster
`Patent Owner’s Expert
`
`IPR2020-00897 | SLIDE 23
`
`

`

`Ground 2 (Claims 11-17) – Ogino Example 5 in view of Bareau
`
`
`
`Case Nos. IPR2020-00897
`U.S. Patent No. 10,324,277
`
`116. A further problem with Dr. Sasián’s First Modified Example 5 is that
`
`the first lens shape leaves no room to oversize or to have rounded or cham-
`
`fered corners. Given the 0.06455 mm edge thickness and large slope, the di-
`defects, which are voids in the central portion of the molded part that are not
`ameter could be increased by less than 0.037 mm before the edge thickness
`material.
`reached zero, i.e., less than 4%. And that assumes that an edge thickness of
`
`the
`
`lens
`
`filled
`
`with
`
`zero were possible, which it is not.
`
`117. Thus, the lens assemblies that Dr. Sasián describes in the First Modified
`
`Example 5 is not usable and is at best an incomplete design that would have
`
`been further modified to address these other issues. Dr. Sasián has not pro-
`
`vided any evidence as to what that ultimate final design, taking into account
`Ex. 2001, p. 61.
`
`
`116. A further problem with Dr. Sasián’s First Modified Example 5 is that
`
`the first lens shape leaves no room to oversize or to have rounded or cham-
`
`fered corners. Given the 0.06455 mm edge thickness and large slope, the di-
`
`ameter could be increased by less than 0.037 mm before the edge thickness
`
`reached zero, i.e., less than 4%. And that assumes that an edge thickness of
`
`61
`
`IPR2020-00897 | SLIDE 24
`APPLE V. COREPHOTONICS
`IPR2020-00897
`Exhibit 2001
`Page 64
`
`

`

`1.
`
`Fig. 1A – Ray Trace Diagram
`designing with the further objective for such manufacturing would have had the
`
`requisite skill to do so. For example, besides the modified Ogino Example 5
`
`presented in the Petition (“alternative 1”), Dr. Sasián provided a further modified
`Pe$$oner’s New Designs on Manufacturability Are Improper New
`design (“alternative 2) that meets Dr. Milster’s “manufacturing” requirements, as
`Evidence and Must Be Rejected
`shown below for comparison:
`2.
`Fig. 1B – Marginal Ray and Mounting
`
`
`
`
`
`
`
`
`
`
`
`Modified Ogino Example 5,
`(alternative 1) APPL-1003, p.122.
`Ex. 1037 at 46.
`As with the other modifications including alternative 2 offered above, Dr.
`
`Modified Ogino Example 5,
`(alternative 2) APPL-1037, ¶49.
`
`
`
`Sasián process included taking gradual steps within the level of skill of a POSITA.
`IPR2020-00487 | SLIDE 25
`
` Apple v. Corephotonics
`
`APPL-1037, ¶50. Specifically, he started with the modified alternative 1 Ogino
`46
`APPL-1037 / IPR2020-00897
`Example 5 lens assembly (see APPL-1003, Appendix, Figs. 4A-4D) and
`
`maintained all radii of curvature from the original Ogino Example 5 lens to keep
`
`

`

`Ground 1: Ignores Bareau
`Teaching of Relative Illumination
`
`IPR2020-00897 | SLIDE26
`
`

`

`$10 (est.)
`
`$0.50 (est.)
`
`$1 (est.)
`
`Cost:
`
`If we were able to simply scale the 35 mm lens design by 1/10x, we would encounter a few issues:
`
`1) Smaller entrance pupil: Depth of field will be much greater, but diffraction will limit performance sooner than with
`larger formats.
`
`2) Surface figure tolerances: Figure tolerances (fringes of irregularity, for example) will be somewhat tighter, because
`spatial frequencies of interest are higher, but because the surfaces are smaller, they will be easier to achieve in practice.
`3) Geometric tolerances: Scaling the system’s size requires linear tolerances to scale as well. So center thickness
`curves. The effect of mismatch is a drop in light collection efficiency or decreased relative illumination at the image, or
`tolerances and surface and element decenter tolerances will be tighter by a factor of ten. This proves to be the greatest
`cross-talk between microlenses and adjacent pixels, resulting in false coloration.
`challenge of producing these lenses.
`
`4) Angular tolerances: Lens tilt tolerances do not scale down, but small defects on flanges or mounting surfaces will
`Rela$ve Illumina$on Results Violate Bareau
`Today, maximum CRA specifications for different sensor formats are readily available in the <12 degree to <26 degree
`
`have a larger effect on tilt.
`range, with the larger CRA allowances corresponding to smaller VGA formats (2.2um, 3.6um). The demand for shorter
`5) Stray light considerations: An aperture or baffle feature that has an acceptably small dimension at the large scale
`TTL’s is putting pressure on sensor manufacturers to increase their maximum allowable CRA values. Added constraints
`should be scaled down by 1/10. However, some parts cannot be made thin enough, or they may become translucent, so
`and fewer elements are lessening the lens designer’s ability to deliver good image quality performance and low CRA’s.
`they will cause a larger fraction of the light to scatter from their edges, resulting in flare or veiling glare.
`The Optics of Miniature Digital Camera Modules
`
`6) Scratch/Dig and Contamination: The smaller system is much more sensitive to defects and contamination causing
`Relative Illumination – The relative illumination is the level of light energy incident at the image plane for a given field
`
`shadowing on the image. Acceptable defect dimensions scale with the format size, and the situation is often worse in
`point relative to that at the center of the image.
`Jane Bareau and Peter P. Clark
`practice, because the back focal distance is very short and defects close to the image