_;f/§\\,,c:=E/\"r'Ivr'="'
`Q a:s.==‘L._/xv =_=v=_=5“r’Er\/I55
`
`
`
`PBS Feasibility Study
`
`Optical Design Concepts
`
`Submitted to: RealD
`
`Submitted by:
`
`Creative Display Systems, LLC
`
`5909 Sea Lion Place, Suite A
`
`Carlsbad, California 92010
`
`(760) 476-0339
`
`(760)476-0620 FAX
`
`This data is furnished to AFRL and shall not be duplicated, used, or disclosed in whole or in part for any purpose
`other than for evaluation. This restriction does not limit AFRL’s right to use information contained in such data if it
`is obtained from another source.
`
`REALD INC.
`Exhibit 2104-1
`
`MASTERIMAGE 3D, et al. v REALD INC.
`IPR2015-00035
`
`

`
`6/‘CREATIVE
`zusr-'=L../xv a—‘:5v=_-=.*rEM5
`
`Table of Contents
`
`Table of Contents .......................................................................................................................... .. i
`
`CDS Technical Proposal 06-1001
`
`Use or disclosure ofwhite paper data is subject to the restrictions on the title page ofthis white paper
`
`i
`
`REALD INC.
`Exhibit 2104-2
`
`MASTERIMAGE 3D, et al. V REALD INC.
`IPR2015—00035
`
`

`
`$6/‘CREATIVE
`:0 ETI3l'3:F=’t.z rm» °..‘£‘~v":‘S"!‘?::t\.«'lE-
`
`PBS Feasibility Study
`Optical Design Concepts
`
`6-14-2005
`
`Described in the following sections are various preliminary optical design concepts for the PBS project.
`A summary is included that shows all options are feasible.
`
`The next step is for a meeting to review these results and collectively look at preliminary options for a
`PBS mechanical concept that will work with most projectors. After that CDS will prepare a final report
`with further optical analyses, a mechanical concept and information required by The Statement of
`Workse.
`
`1.
`
`Use of a Wire-Grid Polarizer (WGP)
`
`In this configuration, a WGP is utilized to divide the unpolarized beam into P and S-polarized beams in
`the vertical direction. As shown in Figure 1, the screen is set at a distance about 45.5m away from the
`projector so that the screen width corresponding to the projector field-of-view of $150 is 25m,. As shown
`in Figure 2,
`the P-polarized transmitted beam will go through a Z-Screen, which includes a linear
`polarizer for cleanup. The Z-Screen is a at-cell that can convert the S-polarized light to a circularly
`polarized light. A 3D view of the configuration is given in Figure 3 that provides a more clear view of
`the vertical fold.
`
`For the reflected path, similarly, the WGP reflects the unpolarized incident beam and converts it to an S-
`polarized light.
`It then goes through a second Z-Screen which converts the linearly S-polarized light to a
`circularly polarized light.
`It is conceived that a HWP is not needed, but the Z-Screen will have its linear
`polarizer oriented so that it passes S-polarized light. The Z-Screen will also need to be driven so that its
`handedness of the circularly polarized light is the same as that of the above transmitted light after
`reflected by the mirror. The reflected beam has a slight toe-in angle of 0.3060 so that its beam center
`will coincide with that of the straight through beam.
`
`For a projected image of 25x13.5m that corresponds to a screen aspect ratio of 1.85:1, and a pixel
`count of 1920x1080, the pixel pitch, P, is
`
`_ 25m
`p ____
`1920
`
`= 13mm
`
`The toe-in angle results in a tilt of the image plane from the reflected path. The largest depth difference
`at the edge of the field-of-view (FOV) between the two paths is about 35mm, which is expected to be
`less than the depth of field of the projection lens. As such, the two images should appear in good focus.
`The two images also has a lateral shift of about 35mm, which is about 3 pixels wide. Both the lateral
`shift and the depth difference between the two images from the two paths are shown in Figure 4. The
`geometric relationship of the angles and distances is shown in Figure 5.
`
`The WGP is available with the maximum size of 180mm round in diameter. The layout given in Figure
`2 has placed the WGP as close to the projection lens front surface as 3mm clearance without vignetting.
`The beam footprint is shown in Figure 6 for a WGP of 160(H)x180(V) dimension. One can see that the
`WGP barely covers the entire beam bundle. As such, there is no margin for any errors that would
`increase its footprint.
`
`The high transmission model of the WGP has an average transmittance and reflectance of about 88%.
`The published data is given in Figure 7. As such, the expected throughput efficiency, together with an
`assumed 94% reflectance of a mirror, for polarized light output with 100% unpolarized light input is 86%,
`
`REALD INC.
`Exhibit 2104-3
`
`MASTERIMAGE 3D, et al. v REALD INC.
`IPR2015-00035
`
`

`
`ED!£F”L../\"r’ F.5‘w‘E3“l"E!\/IE:
`
`which has a gain of almost 100% when compared to the use of a typical linear polarizer made of
`stretched plastics, which has about 45% transmittance for unpolarized input light.
`
`Table 1: Estimated sizes and costs of components for a Moxtek wire—grid
`polarizer configuration.
`
`Part Descri tion
`a 45° WGP reflective olarizer
`160X180mm
`~$1-2K and available
`1 fold mirror M1 for S-ol. ath
`300x300mm
`~$400
`2 Z-Screens
`
`210x170mm
`
`Figure 1: Configuration 1 with a Moxtek 45° wire—grid polarizer
`and 1 fold mirror.
`
`25x13.5m
`
`Toe-in angle =
`a 0.306"
`
`- ¥
`“‘ *~
`N \ 1
`g \_ \\ xx‘ \
`_\_
`‘ i‘ \\\\\\—_ at
`
`4
`
`50””
`
`/L/Ti:
`K\\
`Z-Screen for S-pol.
`210(H)x170(V)mm Z. ‘j
`
`\
`
`1
`
`,
`;
`Lens front aperture ep 1
`100mm diam.
`‘ T’ “
`M0XtekP°1ariZ°I / it
`
`16O(H)X180(V)mm
`
`"
`
`T
`
`1
`
`e
`
`'1
`’
`
`i
`
`1
`
`~
`
`1
`
`1
`
`g
`
`~
`~_~—_
`
`6——e—»»_e_\t1>-‘pint. Path straight— *
`'~ \\_\_g\
`through
`' I " \\'
`
`,\ _ g
`3"‘
`:’::‘:___ ___
`
`*
`
`—e
`
`210(H) x170(V) mm
`
`3
`
`REALD INC.
`Exhibit 2104-4
`
`MASTERIMAGE 3D, et al. v REALD INC.
`IPR2015-00035
`
`

`
`15$ CREATIVE
`:>r<=_-»:==zM/\v =5va~r'sr\/is
`
`Figure 3 3D view of the attachment module
`
`
`
`
`Ill'}l It'll I‘
`Il‘..‘l'-l”'
`‘
`_
`I11;I
`
`Figure 4: Lateral shift and image plane difference between the transmitted and
`reflected paths are both 40mm.
`
`l
`
`The distances of 35mm in longitudinal shift between the toe-in path and the straight path can be
`calculated from the toe-in angle of 0.3060 and the distance of 6.5m from the screen center to the edge
`of the screen in height as below.
`
`d = Ltan(t9) = 6500 * tan(O.306) = 34.7mm
`
`REALD INC.
`Exhibit 2104-5
`
`MASTERIMAGE 3D, et al. v REALD INC.
`IPR2015-00035
`
`

`
`9?/\\4CRE/\1'|\/E
`lg Ds5F=:_/xv =;—3vs‘raM<5
`
`Figure 5: Geometric relationship between toe—in angle and longitudinal shift in
`focal plane, d.
`
`6
`
`
`
`Figure 6: Footprint diagram showing the beam bundle on a 160(H)x180(V)mm
`WGP.
`
`
`
`
`
`SEHLE1;UD.DGUDMILLIHETERS
`
`HPERTUEE FULL I WIDTH 1
`lHPEPTUPE FULL V HEIEH“r
`
`1DO.EUW
`
`4 P9iS THROUGH =
`
`UHFIT
`
`u __,
`_
`SFH TUH mi
`J;!L_'l’:
`SLlF:FF1[‘E H‘
`T \Fll—lSl'l
`PFH 4'. MIN I —e‘3i
`PFH’
`l
`i'lIN :
`—:3‘T
`[VWRV P.FlDILlE:
`ET
`
`‘
`
`
`
`:
`:
`:
`
`':iF~.11ELl
`E¢E.Ll'c«i_H
`Cl.ԤEl:H:l
`
`T
`CDS—llCl,'TEl< ::i*1lZl~CJ:l:l ZH
`C*:il._ll:Ii:;lJP\FlTI"ll:
`;Cll_l_ 2
`
`Figure 7: Published performance of the WGP with an averaged transmittance and
`reflectance of about 88%.
`
`REALD INC.
`Exhibit 2104-6
`
`MASTERIMAGE 3D, et al. v REALD INC.
`IPR2015-00035
`
`

`
`g\/‘CREATIVE
`gJJ $l=.SF"‘-L..,/”\‘~r‘
`‘.-5%/§’T"::."'5k»‘\E:
`
`Typical Polarizer Performance:
`
`45° Angle of Incidence
`
`450nm 550nm 650nm
`
`84.6% 38.4% 89.0%
`
`0.33% 0.19% 0.13%
`
`88.8% 88.8% 86.8%
`
`4.50% 3.38% 2.97%
`
`Tp
`
`Ts
`
`Rs
`
`Rp
`
`
`
`2.
`
`Use of Cube Polarizing Beamsplitter (PBS) with Either Dielectric Thin-Film Coating or Embedded
`WGP
`
`in this configuration, a cube PBS is utilized to divide the unpolarized beam into P and S-polarized
`beams, as shown in Figure 8. The transmitted beam will go through a P-polarized WGP to clean up the
`polarization, followed by going through a HWP to rotate it to S-polarized light before entering the Z-
`Screen. The Z-Screen is a r:—ce|l that can convert the S-polarized light to a circularly polarized light.
`
`For the reflected path, similarly, the PBS reflects the unpolarized incident beam and becomes an S-
`polarized light.
`It then goes through an S-polarized WGP to clean up its polarization, followed by going
`through a second Z-Screen which will convert the linearly S-polarized light to a circularly polarized light.
`The reflected beam has a slight toe-in angle of 0.290 so that its beam center will coincide with that of the
`straight through beam as indicated in Figure 8.
`
`The toe-in angle results in a tilt of the image plane from the reflected path. As indicated in Figure 9, the
`largest depth difference at the edge of the field-of—view (FOV) between the two paths is about 30mm, which
`is expected to be less than the depth of field of the projection lens. As such, the two images should appear
`in good focus. The largest lateral shift is 33mm, which is about 2.5 pixel pitch. A list of components and
`their sizes are provided in Table 2. Some of their estimated costs are given as well. A 3D model is shown
`in Figure 10.
`
`REALD INC.
`Exhibit 2104-7
`
`MASTERIMAGE 3D, et al. V REALD INC.
`IPR2015-00035
`
`

`
`Q/\\ CREATIVE
`fl :3:-s:L./xv §\/E>“T‘Ef\/Vt‘:
`
`IIIIII I
`
`II III
`
`Z-Screen for S-pol.
`220(H)x180(V)mm
`
`'
`
`Lens front aperture
`100mm diam.
`
`—‘>
`
`Figure 8 Configuration with a cube PBS, either with a dielectric thin-film coating
`or an embedded WGP
`
`M1
`:4____
`300x300mm
`I
`
`
`
`Toe-in angle = -
`o.29°
`
`\
`
`I
`
`450mm
`
`\\ __\
`\ _\
`
`I
`
`160(H)x16O(D)x20/
`O(W)mm
`
`U\I:’;~:.~=_{\_ T _ _ :";\_’*;:;___: ,7: _____ _ _
`Z-Screen for P-po1.
`\“
`K
`220(H)x180(V)mm
`
`7
`
`REALD INC.
`Exhibit 2104-8
`
`MASTERIMAGE 3D, et al. v REALD INC.
`IPR2015-00035
`
`

`
`E2}iEr-"“-“L,_/\“~/ 5~«=_s"rEMs:a
`
`Figure 9: Lateral shift and focal plane shift.
`
`Figure 10: 3D model for the cube PBS configuration.
`
`REALD INC.
`Exhibit 2104-9
`
`MASTERIMAGE 3D, et al. v REALD INC.
`IPR2015—00035
`
`

`
`caa/\'rIvr=.-.-
`elf :.>:§s=::,.,.»‘-we E-.~‘3'~/‘E'5“§‘>‘r".:.-'.l‘~/;,‘L~”E«
`
`Table 2 Estimated sizes and costs of components for a PBS configuration
`
`
`
`
`
`E
`
`
`2 Z-Screens
`
`220x180mm
`
`An example of PBS coating performance in T]: and T5 is shown in Figure 11, which is topographical
`view of T]: and T3 vs. angle of incidence (AOI) and wavelengths. As one can see, this example shows
`performance up to about $120. Tp can be as high as 97-99% over a substantial area on the map while
`Ts is less than 0.02% over half of the map, which makes the extinction ratio above 5000 over the
`overlapped region. The weighted averaged curves over for Tp and Rs are given in Figure 12.
`Comparing to those in Figure 7 for the WGP, cube PBS appears to have higher performance. The
`tradeoff would be the weight, thermal loading, availability and cost issues. As opposed to WGP being a
`commercially available part that one can purchase at a reasonable price and delivery schedule, using a
`cube PBS would involve a whole development project in order to realize the advantages.
`
`Figure 11 Design performance, Tp and T3, of a PBS with dielectric thin-film
`coating on glass prism.
`% Transmittance P
`
`:: v—— —~—w———~~—~—¢‘
`
`‘:‘
`
`..;.
`
`
`
`Angleoflll\.'l(|L‘I1CCixlcgl
`
`2}:2
`:1'4
`2-)
`'32
`
`u:
`
`r,,
`;
`
`REALD INC.
`Exhibit 2104-10
`
`MASTERIMAGE 3D, et al. v REALD INC.
`IPR2015-00035
`
`

`
`(“CREATIVE
`mus.’-=t._/\-r ‘_=3‘1’S>"'l’&‘l'\/l5
`
`Figure 12 Weighted averages of TP and RS for a cube PBS with dielectric thin-
`film coating.
`
`100
`
`95
`
`.4--T
`
`90
`
`—Tp —Fb ——Rs ———Ts
`
`-
`r
`75‘
`4-2043044045046047048049050051052053054055056057056059060061062063064065066067068O
`
`3.
`
`Embedded WGP
`
`It is feasible to embed the WGP in between two right angle glass prisms. Due to refraction of the edge
`rays, the prism would reduce the beam size at the diagonal plane, as shown in Figure 13, so that more
`edge clearance is available to allow for any errors.
`Due to the fragility of the wire grid coating,
`special care needs to be taken in assembling the WGP in between the prisms.
`
`Figure 13 Size of WGP, shown in ray trace footprint, needed as embedded
`between 2 glass prisms. The circle is 180mm diameter.
`
`T
`
`l wl
`CL’LU>-
`£
`F31
`J
`4 lE .
`
`9 ,
`
`‘3
`C)
`F?
`
`1
`
`in
`, J
`“F
`’
`
`l ‘
`
`FlFEF.Tl.<F‘E DIFlliETEF'
`.
`IUH ll.
`EL’.
`3ClLii
`EL‘
`FEE ‘ii FES
`P.Fla'
`. HI‘!
`
`FF1
`ll-Cl.
`
`
`/
`
`r
`
`\
`
`X
`
`~
`
`v
`
`
`
`\
`
`\
`
`l
`
`\,
`
`/
`
`'
`
`l5cw:»—3r:-«:‘
`FDDTFFIHT DIFU:-FFll
`
`Ears Tl-lF.ClUE-H :
`DPFIT
`
`to»:w.o~:~.
`
`“Fl
`I
`=
`.
`'3 H=
`
`REALD INC.
`Exhibit 2104-11
`
`MASTERIMAGE 3D, et al. v REALD INC.
`IPR2015-00035
`
`

`
` :?§AT|VE
`A if ::2l=_-=1.-==e.Av <-;s~ws*rEr~AL=5
`
`4.
`
`Summary
`
`A summary of the above 3 cases is given in Table 3. Some compared parameters are clarified in the
`notes. At this time, all three approaches look feasible in providing desired desirable geometry, weight,
`and efficiency.
`Further study will be needed to analyze the other performance factors, such as
`brightness and color uniformity.
`
`Table 3 Comparison summary of 3 cases.
`
`coatin -
`
`embedded WGP
`
`
`
`
`
`Focal plane shift
`
`
`
`
`
`Notes:
`
`(1) The volume includes the space needed for the attachment of the beamsplitting device (WGP or
`PBS), the fold mirror, and the 2 Z-Screens.
`(2) Cost includes the estimated material costs of beamsplitting device (WGP or PBS) and the fold
`mirror. Development cost of a thin—film PBS coating is not included. Cost of Z—Screen is not included.
`(3) Weight includes the beamsplitting device (WGP or PBS) and the fold mirror. Z—Screen weight is not
`included.
`
`(4) Efficiency includes the beamsplitting device (WGP or PBS) and the fold mirror (~94%).
`(5) Risk is mostly associated the available beamsplitting device. For WGP,
`it is related to the small
`clearance in the analyzed case leading to the potential risk in the assumptions made in the configuration
`setup. Also of concern is the bonding of the WGP to a large substrate while maintaining flatness. For
`the cube PBS, it is the risk in successfully completing a thin-film coating over a large area with desired
`performance.
`
`11
`
`REALD INC.
`Exhibit 2104-12
`
`MASTERIMAGE 3D, et al. V REALD INC.
`IPR2015-00035