`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`Paper No. ____
`Filed: ____
`
`
`
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`FOX Factory, Inc.
`
`Petitioner
`
`v.
`
`SRAM, LLC
`
`Patent Owner
`
`____________
`
`
`Case IPR2017-01440
`Patent 9,291,250
`
`____________
`
`
`PETITIONER’S REPLY
`
`
`
`
`

`

`
`
`Contents
`INTRODUCTION ........................................................................................... 1
`REPLY ARGUMENTS ................................................................................... 2
`A.
`SRAM’s arguments against FOX’s prima facie case of
`obviousness in Ground I do not withstand scrutiny. ............................. 2
`1.
`SRAM does not dispute that a POSITA would achieve
`the claimed invention through routine experimentation
`and optimization. ......................................................................... 2
`SRAM’s arguments fail to persuasively rebut the merits
`of Ground I .................................................................................. 5
`a.
`FOX is not required to prove “additive benefits” or
`physical combinability of the references. ......................... 5
`Hattan and JIS do not discourage high outer link
`axial fill. ............................................................................ 7
`JP-Shimano does not limit its wide teeth to the
`thickness of the inner spaces. ........................................... 8
`SRAM misconstrues the claims. ....................................... 8
`d.
`SRAM’s Secondary Considerations Evidence Does Not
`Demonstrate Non-Obviousness. .......................................................... 11
`1.
`SRAM should not receive a presumption of nexus
`between the asserted objective evidence and the claims of
`the ’250 patent. .......................................................................... 12
`SRAM’s Assertions of Unexpected Results, Skepticism,
`Licensing, Long-Felt But Unmet Need, and Copying Do
`Not Provide Persuasive Evidence of Non-Obviousness. .......... 14
`a.
`SRAM’s asserted evidence is factually and legally
`deficient. ......................................................................... 14
`i.
`The claimed invention did not produce
`“unexpected” results. ............................................ 15
`i
`
`b.
`
`c.
`
`B.
`
`Case No. IPR2017-01440
`
` United States Patent No. 9,291,250
`
`
`
`
`
`I.
`II.
`
`
`
`2.
`
`2.
`
`

`

`3.
`
`
`
`
`
`
`
`b.
`
`ii.
`
`Case No. IPR2017-01440
`
`
` United States Patent No. 9,291,250
`
`SRAM has not shown skepticism about the
`X-Sync technology before the alleged
`invention. .............................................................. 18
`The claimed invention did not solve any
`long-felt, unmet need. ........................................... 19
`SRAM’s has not proved copying. ........................ 22
`SRAM’s licensing program is not driven by
`the ’250 patent claims. ......................................... 23
`The asserted evidence applies only to the chainring
`claims (if at all). .............................................................. 25
`No weight should be given to commercial success of the
`drivetrains or praise for the X-Sync chainring because the
`purported presumption of nexus has been rebutted. ................. 26
`a.
`Commercial Success ....................................................... 27
`i.
`SRAM’s commercial success evidence
`applies only to the drivetrains claims. .................. 27
`Because the commercial success of SRAM’s
`drivetrain groupsets is driven by unclaimed
`and non-novel features, the presumption of
`nexus is rebutted. .................................................. 29
`(a) Any commercial success was largely
`driven by the wide-range rear
`cassette, which SRAM asserts is in
`the prior art. ................................................ 30
`(b) The X-Sync includes unclaimed
`features for improving chain
`retention, which are required for
`SRAM’s alleged commercial success. ....... 31
`The unclaimed Type 2 rear derailleur
`made XX1 possible and was largely
`
`ii.
`
`iii.
`
`iv.
`v.
`
`(c)
`
`ii
`
`

`

`4.
`
`
`
`
`
`
`
`
`
`b.
`
`Case No. IPR2017-01440
`
`
` United States Patent No. 9,291,250
`
`responsible for the success of
`SRAM’s drivetrains. .................................. 33
`Praise ............................................................................... 34
`i.
`SRAM’s evidence of praise applies only to
`the chainring claims. ............................................ 34
`Praise for the X-Sync chainring is directed
`to non-novel or unclaimed features of the X-
`Sync chainring. ..................................................... 34
`SRAM’s asserted secondary considerations do not
`outweigh FOX’s strong showing of prima facie
`obviousness and rebuttal of nexus. ........................................... 35
`
`ii.
`
`iii
`
`

`

`
`
`
`
`I.
`
`
`
`
`Case No. IPR2017-01440
`
` United States Patent No. 9,291,250
`
`INTRODUCTION
`This case involves technology that SRAM’s expert concedes is “simple,
`
`intuitive, and readily understandable, even to a person who does not have an
`
`advanced degree in physics or engineering,” and that “mechanical inventions at the
`
`human observable scale are generally more intuitive and easier to understand than
`
`inventions in other fields.” Ex. 1030 at 22:1-23:3. SRAM admits that “the
`
`technology of bicycle chain drives is well over 100 years old and would be well-
`
`understood by a POSITA,” “the art in question is a mechanical art pertaining to
`
`bicycles, recognized to be a fairly predictable field,” and arranging wide teeth to fit
`
`into the outer link spaces of a chain “requires little more than a basic grasp of
`
`elementary geometry.” Ex. 1028 at 27, 30. SRAM has failed to refute FOX’s
`
`strong showing of prima facie obviousness in this predictable and simple
`
`mechanical art.
`
`SRAM’s alleged secondary considerations evidence is also insufficient to
`
`render SRAM’s claims non-obvious. Much of SRAM’s purported secondary
`
`considerations evidence is legally and factually deficient. Although there is some
`
`evidence of commercial success of SRAM’s drivetrains and praise of the X-Sync
`
`chainring, FOX’s counterevidence rebuts any presumption that the success or
`
`praise was driven by the claimed and novel features of the drivetrains and
`
`chainrings. As a result, the challenged claims are unpatentable.
`1
`
`
`
`

`

`
`
`
`Case No. IPR2017-01440
`
` United States Patent No. 9,291,250
`
`
`
`
`II. REPLY ARGUMENTS
`A.
`SRAM’s arguments against FOX’s prima facie case of obviousness
`in Ground I do not withstand scrutiny.
`1.
`SRAM does not dispute that a POSITA would achieve the
`claimed invention through routine experimentation and
`optimization.
`
`SRAM’s Substitute Patent Owner Response (Paper 31, “Response”)
`
`conflates FOX’s Grounds I and II. While FOX’s Ground II (Hattan in view of JP-
`
`Shimano) involves modifying the Hattan chainring to have the narrow-wide shaped
`
`teeth of JP-Shimano (Ex. 1023 ¶ 103), Ground I simply requires optimizing the
`
`thickness of JP-Shimano’s teeth to provide at least 80% or 85% axial fill (as shown
`
`in Hattan and teaching reference JIS) of the outer link spaces to improve chain
`
`retention (id. ¶ 30).
`
`FOX relies on JP-Shimano’s teachings of narrow and wide teeth and its
`
`explicit disclosure that its wide teeth “can be set to conform to the spaces between
`
`the outer link plates (32), so that the chain does not drop when the outer link plates
`
`(32) engage with the teeth (22), even if the chain line is displaced.” Petition at 29
`
`(quoting Ex. 1006 at 16, ll. 108-115 (emphasis added)). Although JP-Shimano does
`
`not explicitly teach a numerical axial fill, because it teaches that the wide teeth are
`
`set to “conform” to the outer link spaces, a POSITA would have known that JP-
`
`Shimano’s wide teeth could provide an axial fill of at least 85% at the halfway
`
`point in the outer link spaces, as confirmed by a CAD analysis by FOX’s expert.
`2
`
`
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`Ex. 1023, ¶¶ 54, 62 (Dr. Neptune explaining that axial fill of at least 85% at the
`
`halfway point can be achieved by “designing the slope of the sides of the teeth [of
`
`JP-Shimano] such that the tapering of the teeth still results in axial fill of 85% at
`
`the point halfway between the root circle and top land.”); Ex. 1031 ¶¶ 17-20
`
`(explaining CAD analysis); Ex. 1034.
`
`Against the backdrop of JP-Shimano’s narrow-wide teeth teachings, FOX
`
`further articulates how and why a POSITA would have looked to Hattan, which
`
`addresses how to prevent chain detachment (just like JP-Shimano), and JIS, an
`
`industrial standard, for their teachings of axial fill ratios at the point halfway
`
`between the top land and the root circle (Petition at 12, 21, 29-36), and would have
`
`optimized the axial fills of JP-Shimano’s teeth as “result-effective variables” to
`
`arrive at the claimed axial fill ranges (id. at 34-35). As described in the Petition,
`
`“[a] recognition in the prior art that a property is affected by the variable is
`
`sufficient to find the variable result-effective.” Petition at 34 n. 7 (quoting In re
`
`Applied Materials, Inc., 692 F.3d 1289, 1297 (Fed. Cir. 2012)). SRAM does not
`
`dispute that axial fill ratios in the inner and outer link spaces are art-recognized
`
`“result-effective variables” that effect chain retention on the chainring. Id. And the
`
`law is clear that the determination of the optimum ranges of a result-effective
`
`variable is within the grasp of a POSITA. Id. at 34-35 (citing In re Boesch, 617
`
`F.2d 272, 276 (C.C.P.A. 1980) (“[D]iscovery of an optimum value of a result
`
`
`
`3
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`effective variable...is ordinarily within the skill of the art.”)). Because Hattan
`
`discloses an overlapping range (of about 74.6% to about 96% axial fill) for the
`
`inner link spaces and JIS further discloses 88% axial fill for the inner link spaces,
`
`and because axial fill is a “result-effective variable,” further optimization to arrive
`
`at the “at least 85 percent” axial fill for the outer link spaces is a matter of routine
`
`experimentation, not innovation. Petition at 35 (citing In re Aller, 220 F.2d 454,
`
`456 (C.C.P.A. 1955)).
`
`SRAM implies that a POSITA may have been skeptical about optimizing the
`
`axial fill of JP-Shimano because of chain-suck concerns. Response at 39. But
`
`avoidance of chain-suck is not an “unexpected benefit” of the claimed invention.
`
`Infra, Section.II.B.2.a.i; Ex. 1031 ¶¶ 21-27. And a POSITA would not have been
`
`skeptical or dissuaded from optimizing the prior art narrow-wide teeth of JP-
`
`Shimano to arrive at the axial fill ratios claimed in the ’250 patent. Ex. 1031 ¶¶ 24-
`
`25. This is because prior art references such as JP-Shimano already teach that
`
`narrow-wide teeth, and particularly wider teeth in the wide-link chain link spaces,
`
`improve chain retention. Id. ¶ 24; Ex. 1006 at 15-16, ll. 78-81, 108-113; Ex. 1023 ¶
`
`54. And a POSITA would have known and reasonably expected that the
`
`thicknesses of JP-Shimano’s teeth could be further optimized through standard
`
`experimentation and optimization processes to arrive at axial fill ratios that
`
`maximize chain retention while minimizing any purported chain-suck issues.
`
`
`
`4
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`Petition at 34-35; Ex. 1023 ¶¶ 37, 53-55, 59-60, 63-64, 69 (Dr. Neptune explaining
`
`that a POSITA would have optimized the teeth of JP-Shimano to improve chain
`
`retention); Ex. 1031 ¶ 25. As Dr. Neptune explained, a POSITA would have
`
`known that Hattan’s and JIS’s axial fill teachings apply at the halfway point
`
`between the root circle and the top land (Ex. 1023 ¶¶ 60, 61), and a POSITA would
`
`have reasonably selected that halfway point when determining an optimum axial
`
`fill to minimize sideways movement of the chain and prevent chain disengagement
`
`(id. at ¶ 59). Mr. Ganaja confirms this. Ex. 1031 ¶ 18 (“It would have been obvious
`
`to set the axial thickness at about halfway between the root circle and the top land
`
`of the chainring teeth to be at least 85% of the axial space between the outer link
`
`plates in order to ‘conform’ the thickness of teeth to the outer link spaces of the
`
`chain, and thereby improve retention of the chain.”).
`
`2.
`
`SRAM’s arguments fail to persuasively rebut the merits of
`Ground I
`
`a.
`
`FOX is not required to prove “additive benefits” or
`physical combinability of the references.
`
`SRAM overstates what is required to establish prima facie obviousness,
`
`apparently contending that the references must be physically combinable and that
`
`FOX must provide quantitative, experimental data showing that the combinations
`
`would additively improve chain retention relative to the individual references.
`
`Response at 18-19, 22-23. SRAM is wrong on both counts.
`
`
`
`5
`
`

`

`
`
`
`
`Case No. IPR2017-01440
`
`
` United States Patent No. 9,291,250
`
`Whether the references can be physically combined is “basically irrelevant,
`
`the criterion being not whether the references could be physically combined but
`
`whether the claimed inventions are rendered obvious by the teachings of the prior
`
`art as a whole.” In re Etter, 756 F.2d 852, 859 (Fed. Cir. 1985) (en banc).
`
`Moreover, Ground I does not require a bodily incorporation of Hattan’s device into
`
`JP-Shimano’s device, nor does it require bodily incorporation of the JIS device into
`
`JP-Shimano’s device (JIS is cited merely for its teachings regarding standard tooth
`
`thickness/axial fill). Ex. 1023 ¶¶ 29-30. JP-Shimano alone teaches all claim
`
`elements, lacking only an explicit teaching of the axial fill at the halfway point,
`
`which a POSITA would have optimized through routine experimentation in view
`
`of the axial fill teachings of Hattan and JIS, as explained in Section II.A.1 above.
`
`Indeed, “[t]he normal desire of artisans to improve upon what is already generally
`
`known can provide the motivation to optimize variables,” In re Ethicon, Inc., 844
`
`F.3d 1344, 1351 (Fed. Cir. 2017), such as axial fill. Petition at 34-35.
`
`SRAM’s argument that Dr. Neptune only said what a POSITA could do, not
`
`what a POSITA would do, is also unfounded. Response at 25. For example, Dr.
`
`Neptune explained why it “would” have been obvious to a POSITA to optimize the
`
`axial fill ranges to improve chain retention. Ex. 1023 at ¶¶ 31, 64, 65. SRAM’s
`
`assertion that Dr. Neptune failed to argue that a POSITA would view the benefits
`
`of these references as combinable (or “additive”) is also unfounded, as is SRAM’s
`
`
`
`6
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`argument that FOX’s obviousness position lacks supporting evidence—both Dr.
`
`Neptune and Mr. Ganaja have testified that achieving the claimed axial fill ranges
`
`would have been a matter of routine experimentation and would have yielded
`
`predictable results. Ex. 1023 ¶¶ 22, 37, 64, 65, 69; Ex. 1031 ¶¶ 18-19.
`
`b. Hattan and JIS do not discourage high outer link
`axial fill.
`
`SRAM’s argument that Hattan and teaching reference JIS discourage high
`
`outer link axial fill because they have implied outer link axial fill ratios that are
`
`lower than JP-Shimano, see Response at 14-16, fails for three reasons.
`
`First, SRAM’s argument is nonsensical because JP-Shimano, the POSITA’s
`
`starting point in Ground I, expressly teaches using wide teeth that are wider than
`
`standard/narrow teeth, which by definition will have higher axial fill in the outer
`
`links than the narrow teeth do. Second, SRAM’s argument is essentially that the
`
`low outer link axial fill ratios taught by Hattan and JIS teach away from high outer
`
`link fill ratios. But neither reference discourages or teaches away from increasing
`
`the outer link axial fill ratio. In re Ethicon, Inc., 844 F.3d 1344, 1351 (Fed. Cir.
`
`2017). Third, JIS’s alleged teaching of zero tolerance for the upper limit of the
`
`tooth width (Response at 36) is directed to the teeth fitment and resulting axial fill
`
`of the narrow link spaces, and it thus fails to discourage applying its axial fill
`
`teachings to the wide teeth of JP-Shimano. A POSITA would instead use the axial
`
`
`
`7
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`fill teachings of JIS as a reference or starting point for optimizing the axial fill of
`
`the outer link spaces. Ex. 1031¶ 25; Ex. 1023 ¶ 55.
`
`c.
`
`JP-Shimano does not limit its wide teeth to the
`thickness of the inner spaces.
`
`SRAM argues that JP-Shimano teaches only that its wide teeth “should be”
`
`as wide as the inner chain link spaces (see, e.g., Response at 21), despite all three
`
`JP-Shimano translations (including SRAM’s) unambiguously stating that the wide
`
`teeth have a width greater than the inner chain link spaces. Ex. 1006 at 16, ll. 108-
`
`110; Ex. 1010 at 255; Ex. 2112 at 18.
`
`d.
`
`SRAM misconstrues the claims.
`
`Contrary to SRAM’s assertion that the claims require the maximum teeth
`
`thickness to be at the halfway point between the root circle and the top land (e.g.,
`
`Response at 33-34), independent claims 1 and 14 instead recite “wherein a
`
`maximum axial width about halfway between a root circle and a top land of the
`
`first group of teeth fills at least 80 percent of an axial distance defined by the outer
`
`link spaces.” SRAM’s construction of this limitation is contradicted by both the
`
`plain language of the claims and the specification.
`
`The plain language of this limitation simply recites that “a maximum width
`
`about halfway between a root circle and a top land” provides an axial fill of at least
`
`80 percent of the outer link spaces. Ex. 1031 ¶ 16. The claims recite only the width
`
`at that recited halfway point, and leave open the possibility of the thickest part of
`8
`
`
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`the teeth being at a location other than that halfway point. Id. This is consistent
`
`with the specification, which never describes the maximum axial width of the
`
`entire tooth as being at that halfway point. Id. ¶ 20.
`
`
`
`Moreover, the specification describes certain amounts of axial fill for the
`
`wide teeth (Ex. 1001 at 4:32-42) in a cross-sectional view “taken through the
`
`tooth,” described as being at the halfway point between the root circle and the
`
`“outside circle.” Id. at 4:15-19. In doing so, the specification contrasts the “overall
`
`width[s]” WO2, WO3, and WO4 of different embodiments of wide teeth—i.e., “a
`
`maximum width about halfway between a root circle and a top land”—with the
`
`“overall width WO1” of the narrow teeth. Id. at 4:20-5:6, Figs. 12-15. The
`
`specification thus describes wide teeth having both a maximum/“overall” width
`
`(e.g., WO4 of Figure 15) and a minimum width (e.g., WO4 – W1 – W1 of Figure 15)
`
`at the halfway point of the teeth, and defines a certain axial fill ratio at that
`
`particular maximum/“overall” width. See, e.g., Ex. 1001 at 4:20-5:6, Fig. 15
`
`(demonstrated below).
`
`
`
`9
`
`

`

`
`
`
`
`
`
`
`Case No. IPR2017-01440
`
` United States Patent No. 9,291,250
`
`
`
`
`
`Contrary to SRAM’s assertions, the plain language of this limitation is met
`
`by tapered chainring teeth (such as those of JP-Shimano) as long as a maximum
`
`axial width of the wide teeth at the halfway point provides at least 80% axial fill of
`
`the outer link spaces. Ex. 1031 ¶ 17. As demonstrated in Ex. 1034, “a maximum
`
`width about halfway between a root circle and a top land” of the wide teeth of JP-
`
`Shimano can provide an axial fill of at least 80 percent of the outer link spaces. Ex.
`
`1031 ¶ 18 (citing Exs. 1032, 1033 and 1034 (demonstrated below)).
`
`
`
`
`
`10
`
`

`

`
`
`
`
`Case No. IPR2017-01440
`
`
` United States Patent No. 9,291,250
`
` SRAM’s attempt to distinguish JP-Shimano by stressing a claim
`
`interpretation contrary to the plain language of the claim should be rejected. If
`
`SRAM had truly desired the different claim scope it now asserts, it would have
`
`claimed, e.g., “wherein the maximum axial width of the first group of teeth is
`
`about halfway between a root circle and a top land, and the maximum axial width
`
`fills at least 80 percent of an axial distance defined by the outer link spaces.” It did
`
`not.
`
`B.
`
`SRAM’s Secondary Considerations Evidence Does Not
`Demonstrate Non-Obviousness.
`SRAM’s asserted evidence of unexpected results, skepticism, long-felt but
`
`unmet need, licensing, and copying is legally and factually deficient and should be
`
`given no probative weight. Although SRAM has some evidence of commercial
`
`success for its drivetrains and praise for its X-Sync chainring, the evidence does
`
`not support non-obviousness because SRAM is not entitled to a presumption of
`
`nexus between the claimed invention and the X-Sync chainring (or the cranksets
`
`and drivetrains that include the X-Sync chainring).
`
`Moreover, FOX’s rebuttal evidence defeats any presumption, if awarded, or
`
`at least significantly reduces its weight because the alleged commercial success
`
`and praise are driven by either unclaimed features (e.g., the rear derailleur of the
`
`drivetrain, the asymmetric teeth of the X-Sync chainring) or features that were
`
`already in the prior art (e.g., narrow-wide teeth, the wide-range rear cassette).
`11
`
`
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`Thus, in evaluating the Graham factors, any evidence of non-obviousness fails to
`
`outweigh FOX’s strong showing of prima facie obviousness. Graham v. John
`
`Deere Co., 383 U.S. 1 (1966).
`
`1.
`
`SRAM should not receive a presumption of nexus between
`the asserted objective evidence and the claims of the ’250
`patent.
`
`Secondary considerations evidence can only be accorded substantial weight
`
`if the patent owner meets its burden to prove a nexus between that evidence and
`
`the merits of the claimed invention. Wyers v. Master Lock Co., 616 F.3d 1231,
`
`1246 (Fed. Cir. 2010). As an alternative to proving nexus directly, the patent owner
`
`may receive a presumption of nexus by showing that “the asserted objective
`
`evidence is tied to a specific product and that product ‘embodies the claimed
`
`features, and is coextensive with them.’” Polaris Indus., Inc. v. Artic Cat, Inc., 882
`
`F.3d 1056, 1072 (Fed. Cir. 2018) (internal citation omitted).
`
`SRAM has not shown direct nexus but asserts entitlement to presumed nexus
`
`because its products are covered by the ’250 patent. Response at 49. A product,
`
`however, is not entitled to presumed nexus where several patents cover it.
`
`Therasense, Inc. v. Becton, Dickinson & Co., 593 F.3d 1289, 1299 (Fed. Cir. 2010)
`
`(no presumption of nexus where patent owner conceded that two different patents
`
`covered the product); Douglas Dynamics, L.L.C. v. Meyer Prods. LLC, No.
`
`IPR2015-01839, Paper No. 29, at 7 (PTAB July 26, 2016) (same).
`
`
`
`12
`
`

`

`
`
`
`
`Case No. IPR2017-01440
`
`
` United States Patent No. 9,291,250
`
`FOX recognizes that in the IPRs for SRAM’s related ’027 patent, the Board
`
`accorded SRAM a presumption of nexus. FOX Factory, Inc. v. SRAM, LLC,
`
`IPR2017-00118, Paper 59, at 21-22 (PTAB April 2, 2018). Respectfully, however,
`
`that decision was based on an error of law. The Board erred in holding that
`
`Therasense applies only if the non-challenged patent is prior art to the challenged
`
`patent. Id. at 20-21. The Federal Circuit’s guidance in Therasense does not support
`
`that proposition. 593 F.3d at 1299. Rather, Therasense broadly instructs that the
`
`presumption is appropriate only “where the success of a product can be attributed
`
`to a single patent . . . .” Id. While the court noted later in the decision that the
`
`second patent was prior art, that observation was relevant only to the direct nexus
`
`analysis and appears nowhere in the Therasense court’s analysis of whether the
`
`presumption was warranted. Whether the other patent is or is not prior art has little
`
`to do with whether the secondary considerations were a direct result of the merits
`
`of the invention claimed in the challenged patent.
`
`SRAM’s asserted product—the X-Sync chainring—is covered by at least
`
`SRAM’s ’027, ’250, and ’211 patents. SRAM concedes the ’027 patent covers the
`
`X-Sync narrow-wide teeth and “offset” features. IPR2017-00118, Paper 13, at 7.
`
`The ’250 patent covers the X-Sync narrow-wide teeth and “gap-filling” features.
`
`Response at 6; Ex. 2141 ¶ 17. And the ’211 patent covers the X-Sync asymmetric
`
`teeth, protruding tip portions, and hooks on rear flanks of the teeth, all of which
`
`
`
`13
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`SRAM touts in the specification of the ’250 patent and all preferred embodiments,
`
`and which FOX’s expert, Mr. Ganaja, has opined are important features that
`
`promote chain retention. Ex. 1031 ¶¶ 46-49. In fact, SRAM is applying largely the
`
`same evidence of secondary considerations to different patents—the ’027 patent
`
`and the ’250 patent—with different claim features, the only commonality at the
`
`heart of these claims being a feature known in the art (narrow-wide teeth). SRAM
`
`is not entitled to a presumption of nexus.
`
`2.
`
`SRAM’s Assertions of Unexpected Results, Skepticism,
`Licensing, Long-Felt But Unmet Need, and Copying Do Not
`Provide Persuasive Evidence of Non-Obviousness.
`
`a.
`
`SRAM’s asserted evidence is factually and legally
`deficient.
`
`Regardless of whether SRAM receives a presumption of nexus, SRAM’s
`
`asserted evidence of unexpected results, skepticism, licensing, long-felt but unmet
`
`need, and copying fails on the merits because there was no skepticism before
`
`SRAM’s claimed invention, nor did the claimed invention generate unexpected
`
`results or satisfy any unmet need. There has also been no copying of SRAM’s X-
`
`Sync chainrings. And although the “X-Sync technology” has been licensed, the
`
`licenses were not driven by the ’250 patent.
`
`
`
`14
`
`

`

`
`
`
`
`Case No. IPR2017-01440
`
`
` United States Patent No. 9,291,250
`
`The claimed invention did not produce
`“unexpected” results.
`SRAM urges that the claimed invention achieved “unexpected results”
`
`i.
`
`because a POSITA allegedly would have expected chain-suck problems at above
`
`80% axial fill. Response at 38-44. SRAM’s arguments fail, however, because
`
`SRAM provides no prior art or references that mention chain-suck, or indicate that
`
`a POSITA would have expected chain-suck at the claimed axial fill range. Kao
`
`Corp. v. Unilever U.S., Inc., 441 F.3d 963, 970 (Fed. Cir. 2006) (“[W]hen
`
`unexpected results are used as evidence of nonobviousness, the results must be
`
`shown to be unexpected compared with the closest prior art.”) (citation omitted;
`
`emphasis added); In re Swain, 33 C.C.P.A. 1266, 1268, 156 F.2d 246, 247-48
`
`(1946) (requiring unexpected result “over the disclosure of the prior art”); Ex. 1031
`
`¶ 21. Because SRAM has presented no prior art-based evidence that axial fill
`
`above 80% would have been expected to cause chain-suck, or that avoiding chain-
`
`suck was an unexpected result over the closest prior art, such as JP-Shimano,
`
`SRAM’s asserted unexpected results should be accorded no weight. See Kao, 441
`
`F.3d at 970; Ex. 1031 ¶ 21. SRAM also has not provided any comparative testing
`
`with prior art chainrings to demonstrate the asserted “unexpected result” (Ex. 1031
`
`¶ 21), so SRAM’s asserted “unexpected result” fails for that reason as well. In re
`
`Hoch, 428 F.2d 1341, 1343-44 (C.C.P.A. 1970).
`
`
`
`15
`
`

`

`
`
`
`
`Case No. IPR2017-01440
`
`
` United States Patent No. 9,291,250
`
`Moreover, SRAM’s “showing of unexpected results must be commensurate
`
`in scope with the claimed range.” In re Peterson, 315 F.3d 1325, 1330 (Fed. Cir.
`
`2003). SRAM’s asserted unexpected result fails because, although SRAM claimed
`
`an axial fill range of at least 80% (i.e., between 80-100%), SRAM has not provided
`
`any data showing unexpected results throughout the claimed range. Ex. 1031 ¶ 22.
`
`In fact, chain-suck cannot be avoided at 100% axial fill, even though SRAM’s
`
`claims cover axial fill of 100%. Id. SRAM’s unexpected results assertion is not
`
`commensurate with the scope of the claimed range and fails for that reason as
`
`well. In re Peterson, 315 F.3d at 1330.
`
`As discussed previously, the claimed axial fill is simply the optimization of a
`
`“result-effective variable” “that flows from the normal desire of scientists or
`
`artisans to improve upon what is already generally known.” Pfizer, Inc. v. Apotex,
`
`Inc., 480 F.3d 1348, 1368 (Fed. Cir. 2007). A POSITA would not have been
`
`dissuaded from experimenting with (and optimizing) wide teeth having axial fill in
`
`the claimed range (i.e., above 80% or above 85%) to maximize chain retention and
`
`minimize any chain-suck issues. Ex. 1031 ¶ 24. Instead, because JP-Shimano
`
`teaches that higher axial fill improves chain retention (Ex. 1006 at 15-16, ll. 78-81,
`
`108-113), a common-sense optimization method would have been to start at about
`
`88% or 96% axial fill (the highest axial fill taught by the prior art) and iteratively
`
`lower the axial fill to arrive at optimum axial fill percentages for the inner and
`
`
`
`16
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`outer links where chain-suck is not a concern. Ex. 1004 at 7:52-66 (teaching 96%
`
`axial fill of the inner link spaces); Ex. 1025 at 6 (Fig. 4) (teaching 88% axial fill of
`
`the inner link spaces); Ex. 1031 ¶¶ 24-25. This would have been only routine
`
`experimentation, not innovation. Ex. 1023 ¶¶ 28, 64; Ex. 1031 ¶ 25. At an
`
`optimum axial fill, a POSITA would have reasonably expected improved chain
`
`retention and minimal chain-suck. Ex. 1031 ¶ 26. Moreover, this iterative
`
`optimization process would have comprised routine mechanical processes easily
`
`executed by a POSITA without requiring excessive experimentation. Ex. 1023 at
`
`¶¶ 37, 121; Ex. 1031 ¶ 25. Indeed, even SRAM concedes that its inventor, Mr.
`
`Braedt, determined that chain suck was not a problem after only “some testing,”
`
`and a POSITA would have found the same. Response at 38 (emphasis added); Ex.
`
`1031 ¶ 27.
`
`Regardless, the primary cause of chain-suck is not lateral friction with the
`
`link plates due to high axial fill, as SRAM contends, but worn or bent teeth and
`
`mud/grit accumulation between the chain and the teeth, which prevent
`
`disengagement of the chain from the bottom teeth of the chainring. Ex. 1031 ¶ 23.
`
`Because a POSITA would have understood that high axial fill is not a major cause
`
`of chain-suck, he/she would not have been concerned with increasing the axial fill
`
`to 80% or more. Id. And even if a POSITA had been concerned about chain-suck
`
`
`
`17
`
`

`

`
`Case No. IPR2017-01440
`
`
`
` United States Patent No. 9,291,250
`
`
`at very high axial fills, he/she would have known that the axial fill could be readily
`
`optimized without undue experimentation. Id. ¶ 24.
`
`ii.
`
`SRAM has not shown skepticism about the X-
`Sync technology before the alleged invention.
`For evidence of skepticism to be given any weight, it must be expressed by
`
`industry experts or “skilled artisans before the invention.” In re Rouffet, 149 F.3d
`
`1350, 1355 (Fed. Cir. 1998) (citation omitted; emphasis added); Brown &
`
`Williamson Tobacco Corp. v. Philip Morris, 229 F.3d 1120, 1129 (Fed. Cir. 2000)
`
`(“skepticism or disbelief before the invention [is] an indicator of
`
`nonobviousness.”) (citation omitted; emphasis added); United States v. Adams, 383
`
`U.S. 39, 52 (1966).
`
`SRAM’s purported evidence of skepticism constitutes statements in
`
`magazine articles about how SRAM’s chainring “could actually maintain the chain
`
`on the chainring, without the assistance from other well-known add-ons, such as
`
`chain guides.” Response at 52-53, 56. These statements, however, were made after
`
`SRAM’s alleged invention, when the X-Sync was being tested by magazine
`
`authors/reviewers. Such post-invention skepticism is not an indicator of non-
`
`obviousness because it does not show that a POSITA would have been dissuaded
`
`at the time of the invention from combining the prior art features to arrive at the
`
`claimed invention.
`
`
`
`18
`
`

`

`
`
`
`
`Case No. IPR2017-01440
`
`
` United States Patent No. 9,291,250
`
`Moreover, there is no evidence that the magazine authors are POSITAs or
`
`were aware that JP-Shimano already taught that narrow-wide chainrings improve
`
`chain retention. Ex. 1006 at 15-16, ll. 78-81, 108-113. These expressions of
`
`skepticism by lay persons unaware of JP-Shimano should be given no weight in
`
`determin