Zircon Corp. v. Stanley Black & Decker, Inc. , 452 F. App'x 966 ( 2011 )


Menu:
  •        NOTE: This disposition is nonprecedential.
    United States Court of Appeals
    for the Federal Circuit
    __________________________
    ZIRCON CORPORATION
    Plaintiff-Appellant,
    v.
    STANLEY BLACK & DECKER, INC. (FORMERLY
    THE STANLEY WORKS),
    Defendant-Appellee.
    __________________________
    2010-1405
    __________________________
    Appeal from the United States District Court for the
    Northern District of California in Case No. 09-CV-0042,
    Magistrate Judge Elizabeth D. Laporte
    _________________________
    Decided: October 5, 2011
    _________________________
    CLARK S. STONE, Haynes and Boone, LLP, of San Jose,
    California, argued for plaintiff-appellant. With him on
    the brief were DEBRA J. MCCOMAS and JOHN R. EMERSON,
    of Dallas, Texas.
    BRYAN P. COLLINS, Pillsbury Winthrop, Shaw
    Pittman, LLP, of McLean, Virginia, argued for defendant-
    appellee. With him on the brief were JACK S. BARUFKA
    ZIRCON   v. STANLEY                                       2
    and BENJAMIN L. KIERSZ; and DIANNE L. SWEENEY, of Palo
    Alto, California.
    __________________________
    Before LOURIE, MAYER, and O’MALLEY, Circuit Judges.
    O’MALLEY, Circuit Judge.
    Zircon Corporation (“Zircon”) alleges that Stanley
    Black & Decker, Inc. (formerly The Stanley Works)
    (“Stanley”) infringes United States Patent No. 7,116,091
    (“’091 patent”), which discloses a device for ratiometric
    stud sensing. Zircon appeals the judgment of the United
    States District Court for the Northern District of Califor-
    nia, granting Stanley’s Motion for Summary Judgment of
    Non-Infringement. The district court granted summary
    judgment because it construed the term “ratio” to include
    only division, and it is undisputed that Stanley’s device
    does not utilize division. Because we find that the district
    court correctly construed the term “ratio,” determined
    that there was no literal infringement, and that the
    disclosure-dedication rule barred infringement under the
    doctrine of equivalents, we affirm the judgment of the
    district court.
    BACKGROUND
    A. Factual Background
    This dispute involves the alleged infringement of a
    patent disclosing a product commonly known as a “stud
    finder.” As the name suggests, these electronic devices
    are generally used to locate a stud (i.e., a structural
    member of a building to which an interior wall surface is
    affixed) behind the surface of a wall. ’091 patent col.1
    ll.5–20. Zircon, the owner of the ’091 patent, alleges that
    Stanley’s Stud Sensor 200 and FatMax® Stud Sensor 400
    infringe both the patented device and method disclosed in
    3                                           ZIRCON   v. STANLEY
    the patent. The key issue on appeal centers on whether
    the ’091 patent claims a method of and an apparatus for
    sensing studs using subtraction, or only by using division.
    1. The ’091 Patent
    Zircon’s ’091 patent discloses an “electronic stud-
    sensor used to detect centerlines and edges of wall studs,
    floor joists, and the like.” 1 
    Id. at col.1
    ll.10–12. To detect
    studs, the invention employs a radiometric capacitive
    sensor that uses capacitive measurements from multiple
    conductive plates to determine the presence of a stud
    behind the surface of a wall. 2 
    Id. at col.4
    ll.30–35. The
    presence of a stud behind a wall changes the dielectric
    constants measured by the conductive plates. 
    Id. at col.1
    ll.32–33. Measuring these changes enables detection of
    studs.
    All of the claims of the ’091 patent disclose either a
    method or a device using at least two conductive plates to
    measure fluctuations in dielectric constants. To locate a
    stud using two conductive plates, “[e]ach conductive plate
    acts as part of a separate capacitor.” 
    Id. at col.4
    ll.35–36.
    Circuitry coupled to each plate measures the effective
    change in capacity of the separate capacitors, “while the
    sensor is moved along the wall surface.” 
    Id. at col.1
    ll.33–
    34. The capacitance of a wall covering an underlying stud
    is larger than that of a wall not covering an underlying
    stud. 
    Id. at col.
    4 ll.39–42. In light of this fact, as the
    1   While it is clear that the patent discusses detec-
    tion of objects other than studs, for clarity, this opinion
    will only refer to studs as the objects being located. This
    decision should not be read as limiting the patent to
    detecting only studs, an issue that is not before this court
    and one on which we express no opinion.
    2   Capacitance” is the ability to store electrical
    charge in a dielectric or insulating material. Joint Ap-
    pendix (“J.A.”) 599.
    ZIRCON   v. STANLEY                                       4
    sensor is moved along the wall, “[t]he capacitance meas-
    urement from one plate may then be compared to a ca-
    pacitance measurement of another plate to determine
    boundaries and features of the materials in the vicinity of
    the plates.” 
    Id. at col.4
    ll. 42–46.
    The patent discloses various methods for comparing
    the capacitances of the two plates to determine whether
    the sensor is over a stud’s centerline or edge. For exam-
    ple, describing figure 5A, the specification explains that:
    In some embodiments, capacitance measurements
    are used to calculate a ratio. A first capacitance
    measurement represents the change in capaci-
    tance from a minimum value experienced on a
    first plate 301. A second capacitance measure-
    ment represents the change in capacitance from a
    minimum value experienced on a second plate
    302. A ratio between the first and second capaci-
    tance measurements may be computed. If the ra-
    tio is approximately equal to a predetermined
    value, it may be determined that a centerline 304
    of the sensor 300 is centered over an edge 102 of a
    stud 100. If the capacitance measurements are
    equal or the ratio is approximately equal to unity,
    both plates may be centered over the stud’s edge
    102 and the centerline 304 of the sensor 300 may
    be centered over the centerline 101 of the stud
    100.
    
    Id. col.7 ll.20–34.
    Furthermore, “in accordance with the
    present invention[,] [t]his ratio may be computed as the
    smaller capacitance divided by the larger capacitance,
    thereby resulting in a ratio that is equal to or less than
    one.” 
    Id. at col.
    7 ll.50–52.
    In contrast to this method employing division, the
    patent’s specification also discloses a method that utilizes
    5                                         ZIRCON   v. STANLEY
    subtraction to determine whether the sensor is positioned
    over the centerline of a stud. Describing Figure 12, the
    specification explains that:
    The comparison circuit 414 may determine
    whether the capacitive measurements are within
    a predetermined value of each other. . . . For ex-
    ample, [the] comparison circuit 414 may deter-
    mine that the sensor 300 is centered over a stud
    100 by detecting that the capacitance measure-
    ments are equal to each other and also above a
    floor threshold.
    
    Id. at col.1
    4 ll.48–63. With respect to the ’091 patent,
    “[c]apacitance measurements may be considered equal
    when they are within a predetermined percentage value
    or absolute value from each other.” 
    Id. at col.1
    4 ll.64–66
    (emphasis added). This reference to “an absolute value
    from each other” indicates subtraction. These alternate
    methods of comparing the measured capacitances are
    crucial to this appeal because every independent claim of
    the ’091 patent claims either “computing” or “generating”
    a “ratio of the first and second capacitances.” E.g., 
    id. at claim
    10.
    In addition to the specification, the prosecution his-
    tory of the ’091 patent also discusses both the division and
    subtraction methods of comparing capacitances. The ’091
    patent issued from U.S. Patent Application No.
    10/794,356 (“’356 application”). J.A. 493–558. Original
    independent claim 21 of the ’356 application recited
    “comparing the first and second capacitances.” J.A. 219.
    The remaining independent claims of the ’356 application
    all recited either computing or generating “a ratio of the
    first and second capacitances.” E.g., J.A. 217, 220, 222.
    The ’356 application, therefore, distinguished between the
    ZIRCON   v. STANLEY                                      6
    general function of “comparing” and the specific “ratio”
    function.
    Similarly, claims 22 and 23 of the ’356 application,
    which were dependent upon claim 21, clarified that “com-
    paring” denoted both the ratio and the subtraction meth-
    ods.    Specifically, claim 22 stated that claim 21’s
    comparing function was performed by “computing a ratio
    between the first and second capacitances,” — i.e., by
    using division. J.A. 219. Conversely, claim 23 performed
    claim 21’s comparison by “determining whether the first
    and second capacitances differ by less than a threshold,”
    — i.e., by using subtraction. J.A. 219–20. Thus, the
    original claims of the ’356 application made clear that the
    generic term “comparing” denoted both the division
    method and the subtraction method, while “computing a
    ratio” only referred to the former.
    In Zircon’s August 24, 2005 preliminary amendment,
    however, claim 21 was altered to be dependent upon claim
    1 of the ’356 application. J.A. 248–59. Claim 1 requires
    the computation of a “ratio” on the basis of the first and
    second capacitances, J.A. 250, while amended claim 21
    recites that claim 1’s “computing a ratio” is performed by
    comparing the first and second capacitances. J.A. 253.
    While this preliminary amendment was not filed in
    response to an examiner’s rejection, and it was voluntary,
    the fact remains that, because of the amendment, none of
    the independent claims of ’356 application continued to
    recite the general “comparing” limitation.
    After the preliminary amendment was filed, the ex-
    aminer issued an Office Action rejecting a number of
    claims, including amended claims 1 and 21. J.A. 260–73.
    The examiner indicated, however, that certain claims,
    including claim 8, would be allowable if rewritten into
    independent form. J.A. 271–72. Claim 8 was originally
    7                                          ZIRCON   v. STANLEY
    dependent upon claim 1; thus it recited “computing a
    ratio.” In response to this Office Action, Zircon filed an
    amendment on May 5, 2006, altering claim 8 to be inde-
    pendent. In addition, Zircon altered claim 21 so that it
    depended upon claim 8 instead of claim 1. Eventually the
    ’356 application’s amended claim 8 was allowed as ’091
    patent claim 10. J.A. 67; J.A. 280. This claim recites
    “computing a ratio of the first and second capacitances.”
    ’091 patent col.21 l.56. In addition, amended claim 21
    issued as claim 19 of the ’091 patent. 3 J.A. 67; 280.
    Significantly, every independent claim of the ’091 patent
    that issued from the ’356 application includes the ratio
    limitation and none use the term “comparing.”
    2. Stanley’s Products
    Zircon alleges that Stanley’s Stud Sensor 200 and
    FatMax® Stud Sensor 400 infringe the ’091 patent. As
    explained by Stanley’s expert:
    Stanley’s Stud Finders compare the two capaci-
    tances (via the voltages Va and Vb) by subtraction,
    not division. Stanley’s Stud Finders never gener-
    ate or compute a ratio of the capacitances through
    plates A and B or a ratio of the voltages Va and Vb
    for determining the presence or detection of a stud
    or other object behind a wall.
    J.A. 315. While Stanley’s stud finders utilize a method
    that is very similar to that claimed in the ’091 patent,
    Zircon admits that Stanley’s products utilize the subtrac-
    tion method of comparing capacitances. Appellant’s Br.
    3   Claims 21 and 22 of the ’091 patent issued from
    amended claims 23 and 24 of the ’356 application, respec-
    tively. J.A. 67; J.A. 281. As in the ’091 patent, in the ’356
    application, amended claims 23 and 24 where dependent
    upon amended claim 21, which became claim 19 in the
    ’091 patent. 
    Id. ZIRCON v.
    STANLEY                                      8
    25 (“The subtraction approach used by the accused prod-
    ucts performs substantially the same function in substan-
    tially the same way to obtain the same result as the
    division approach.”).
    B. Procedural History
    After Zircon commenced this suit, Stanley moved for
    summary judgment of non-infringement.            J.A. 112.
    Specifically, Stanley argued that, under the proper con-
    struction of “ratio,” which it asserted should be inter-
    preted as the quotient of dividing two values, its products
    do not literally infringe because they employ subtraction,
    not division, to compare capacitances. J.A. 126–27.
    Stanley argued, moreover, that its products do not in-
    fringe under the doctrine of equivalents because Zircon is
    barred from claiming that the subtraction method is an
    equivalent of the division method under either: (1) the
    disclosure-dedication rule; or (2) prosecution history
    estoppel. J.A. 127.
    In opposition to Stanley’s motion, Zircon did not dis-
    pute that Stanley’s devices use subtraction to compare
    capacitances. Instead, Zircon argued that the term “ra-
    tio,” as properly construed, means the “relation in degree
    or number between two similar things.” J.A. 438. Under
    this construction, Zircon argued that Stanley’s products
    literally infringe the ’091 patent. J.A. 444–48. In addi-
    tion, Zircon argued that, even if the court adopted
    Stanley’s proposed claim construction, Stanley’s products
    still infringed under the doctrine of equivalents because
    the doctrine of prosecution history estoppel did not pre-
    vent Zircon from claiming subtraction as an equivalent.
    J.A. 448–49. Finally, Zircon argued that the disclosure-
    dedication rule did not apply under the facts of this case
    because Zircon both “disclosed and claimed the subtrac-
    tion method for ‘computing a ratio’ . . . .” J.A. 449.
    9                                          ZIRCON   v. STANLEY
    With respect to summary judgment of literal in-
    fringement, the district court noted that the parties were
    in agreement that resolution of the motion turned on the
    proper construction of the term “ratio.” Zircon Corp. v.
    Stanley Works, 
    713 F. Supp. 2d 881
    , 889 (N.D. Cal. 2010).
    After reviewing the claims, the specification, the prosecu-
    tion history, and relevant dictionary definitions, the
    district court construed the term “ratio” to mean the
    “result of dividing two values.” 
    Id. at 894.
    In light of this
    construction, the district court concluded that Stanley’s
    products did not literally infringe the ’091 patent because
    there was no dispute that its products used subtraction
    and not division to compare capacitances. 
    Id. at 895.
         Turning to infringement under the doctrine of equiva-
    lents, the district court found that Zircon could not assert
    that subtraction was an equivalent of the division method
    because both the disclosure-dedication rule and prosecu-
    tion history estoppel barred Zircon from claiming that the
    two methods were equivalents. 
    Id. at 900–01.
    The dis-
    trict court, therefore, granted Stanley’s Motion for Sum-
    mary Judgment of Non-Infringement.
    Following the district court’s entry of judgment of
    non-infringement, Zircon timely appealed.       We have
    jurisdiction pursuant to 28 U.S.C. § 1295(a)(1). As dis-
    cussed below, because we agree with the district court’s
    construction of the term “ratio” and find that the disclo-
    sure-dedication rule prevents Zircon from claiming sub-
    traction as an equivalent of the division method, we
    affirm.
    DISCUSSION
    Claim construction is an issue of law that we review
    de novo. Cybor Corp. v. FAS Techs., Inc., 
    138 F.3d 1448
    ,
    1456 (Fed. Cir. 1998) (en banc). We review the grant of
    summary judgment without deference, drawing all rea-
    ZIRCON   v. STANLEY                                       10
    sonable factual inferences in favor of the non-moving
    party. Anderson v. Liberty Lobby, Inc., 
    477 U.S. 242
    , 255
    (1986); Johns Hopkins Univ. v. Cellpro, Inc., 
    152 F.3d 1353
    (Fed. Cir. 1998). Summary judgment is appropriate
    when the moving party demonstrates that “there is no
    genuine issue as to any material fact and that the moving
    party is entitled to a judgment as a matter of law.” Fed.
    R. Civ. P. 56(c); Celotex Corp. v. Catrett, 
    477 U.S. 317
    ,
    322–23 (1986). While we generally review infringement,
    either literal or by equivalents, as a question of fact,
    where the parties do not dispute any relevant facts re-
    garding the accused product, the question of literal in-
    fringement collapses into claim construction, and thus,
    becomes a question of law that we review de novo. 4 See
    General Mills, Inc. v. Hunt-Wesson, Inc., 
    103 F.3d 978
    ,
    983 (Fed. Cir. 1997) (“Where the parties do not dispute
    any relevant facts regarding the accused product, . . . but
    disagree over possible claim interpretations, the question
    of literal infringement collapses into claim construction
    and is amenable to summary judgment.”).
    With respect to infringement under the doctrine of
    equivalents, prosecution history estoppel, as a limit on
    this doctrine, presents a question of law that we review de
    novo. Wang Labs., Inc. v. Mitsubishi Elecs. Am., Inc., 
    103 F.3d 1571
    , 1578 (Fed. Cir. 1997). Similarly, application of
    the disclosure-dedication rule is a question of law subject
    to de novo review. Pfizer, Inc. v. Teva Pharms., USA, Inc.,
    
    429 F.3d 1364
    , 1378 (2005).
    I.
    The focus of the parties’ dispute is the district court’s
    construction of the term “ratio.” The district court con-
    4  Here, because the parties concede that there is no
    genuine issue of material fact, Appellant’s Br. 11, the case
    presents pure questions of law.
    11                                         ZIRCON   v. STANLEY
    strued the term to mean the result of dividing two values.
    
    Zircon, 713 F. Supp. 2d at 894
    . As previously discussed,
    the crux of this disagreement turns on whether “ratio,” as
    used in the ’091 patent, encompasses only division, or,
    instead, encompasses either division or subtraction. In
    challenging the district court’s construction of the term
    “ratio,” Zircon argues that the term should be construed
    to mean simply a “relation in degree or number between
    two similar things,” regardless of how that relationship is
    calculated. Appellant’s Br. 8–9. Specifically, Zircon
    asserts that its proposed construction is correct in light of
    the plain meaning of the term “ratio,” the claim language,
    the specification, and the prosecution history of the ’091
    patent. On the basis of this same evidence, Stanley
    argues that the district court’s construction of “ratio” is
    correct.
    Generally, claim terms should be given their ordinary
    and customary meaning to a person having ordinary skill
    in the art at the time of the effective date of the patent
    application. Phillips v. AWH Corp., 
    415 F.3d 1303
    , 1312–
    13 (Fed. Cir. 2005) (en banc); see also Innova/Pure Water
    Inc. v. Safari Water Filtration Sys., Inc., 
    381 F.3d 1111
    ,
    1116 (Fed. Cir. 2004) (“A court construing a patent claim
    seeks to accord a claim the meaning it would have to a
    person of ordinary skill in the art at the time of the inven-
    tion.”). To determine the scope and meaning of disputed
    claim terms, we look to the words of the claims them-
    selves, the written description, the prosecution history,
    and, finally, any relevant extrinsic evidence. 
    Phillips, 415 F.3d at 1312
    –19.
    While a court can examine this evidence in any or-
    der, 5 generally, our examination begins with intrinsic
    5 Nor is the court . . . required to analyze sources in
    any specific sequence, as long as those sources are not
    ZIRCON   v. STANLEY                                           12
    evidence, namely the language of the claims. Vitronics
    Corp v. Conceptronic, Inc., 
    90 F.3d 1576
    , 1582 (Fed. Cir.
    1996). We have noted that “the context in which a term is
    used in the asserted claim can be highly instructive.”
    
    Phillips, 415 F.3d at 1314
    . Next, the court should inter-
    pret the claims in light of the specification. 
    Id. at 1315.
    While it is improper to read limitations from the specifica-
    tion into the claims, the claims “must be read in view of
    specification, of which they are a part.” 
    Id. at 1315
    (cita-
    tion omitted); see also Merck & Co. v. Teva Pharms. USA,
    Inc., 
    347 F.3d 1367
    , 1370 (Fed. Cir. 2003) (“[C]laims must
    be construed so as to be consistent with the specification .
    . . .”). In fact, the specification is the single best guide to
    the meaning of a claim term; it is, usually, dispositive.
    
    Phillips, 415 F.3d at 1318
    (“[T]he specification ‘is always
    highly relevant to the claim construction analysis. Usu-
    ally, it is dispositive . . . .’ ”) (internal citations omitted).
    As the final step in analyzing the intrinsic evidence,
    the court should “consider the patent’s prosecution his-
    tory, if it is in evidence.” 
    Id. at 1317.
    The court should be
    aware, however, that the prosecution history “often lacks
    the clarity of the specification” and thus, can be of limited
    use in claim construction. 
    Id. While our
    case law emphasizes the importance of in-
    trinsic evidence, as noted, extrinsic evidence is also rele-
    vant to claim construction. 
    Phillips, 415 F.3d at 1317
    .
    Such evidence consists of all evidence extrinsic to the
    patent and its prosecution history, including “expert and
    inventor testimony, dictionaries, and learned treatises.”
    
    Id. (internal quotation
    omitted).       While authorizing
    used to contradict claim meaning that is unambiguous in
    light of the intrinsic evidence.” Phillips, 
    415 F.3d 1324
    .
    In this regard, it is never error to consider extrinsic
    evidence, it is only error to give it undue weight.
    13                                          ZIRCON   v. STANLEY
    examination of extrinsic evidence, we have warned that,
    while it “can shed useful light on the relevant art,” it is
    “less significant than the intrinsic record in determining
    the legally operative meaning of claim language.” 
    Id. (internal quotation
    omitted).
    With these standards in mind, we turn to the lan-
    guage of the ’091 patent’s claims. Representative inde-
    pendent method claim 10 recites:
    A method of finding a feature behind a surface us-
    ing a sensor having first and second plates, the
    method comprising the acts of: moving the sensor
    and surface adjacent one another; measuring a
    first capacitance of a first capacitor including the
    first plate; measuring a second capacitance of a
    second capacitor including the second plate; com-
    puting a ratio of the first and second capacitances;
    determining whether the ratio is within a prede-
    termined range; and wherein the predetermined
    range has fixed boundaries.
    ’091 patent col.21 ll.48–60 (emphasis added). Representa-
    tive apparatus claim 23, similarly, recites the ratio limita-
    tion, it claims:
    A sensor for finding a feature of a structure, com-
    prising: a first plate having a first capacitance and
    adapted for forming a first capacitor with the
    structure; a second plate having a second capaci-
    tance and adapted for forming a second capacitor
    with the structure; a first measurement circuit
    coupled to the first plate, the first measurement
    circuit measuring a first capacitance value of the
    first capacitor; a second measurement circuit cou-
    pled to the second plate, the second measurement
    circuit measuring a second capacitance value of
    the second capacitor; and a comparison circuit
    ZIRCON   v. STANLEY                                        14
    coupled to the first and second measurement cir-
    cuits, the comparison circuit generating a ratio of
    the first and second capacitance values; an indica-
    tor coupled to the comparison circuit thereby to
    provide an indication of the ratio of the capacitan-
    ces; wherein if the indication is that the ratio is
    approximately one, thereby locating a centerline of
    the structure.
    
    Id. at col.
    22 ll.49–66 (emphases added). The emphasized
    language of claim 23 identifies that “ratio,” as used in this
    claim, denotes division because the patented sensor
    locates the centerline of a structure when the ratio is
    approximately one; the quotient of two equal numbers is
    one, also known as “unity.” 
    Id. at col.
    7 ll.30–33 (“If the
    capacitance measurements are equal or the ratio is ap-
    proximately equal to unity, both plates may be centered
    over the stud’s edge 102 and the centerline 304 of the
    sensor 300 may be centered over the centerline 101 of the
    stud 100.”). 6
    While the language of claim 23 itself makes clear that
    the “ratio” is to be calculated using division, it is unclear
    from reference only to the language of claim 10 whether
    the “ratio” referenced there also could be determined by
    use of subtraction. The claim states that a “ratio” is
    computed to determine “whether the ratio is within a
    predetermined range.” 
    Id. at col.
    21 ll.56–58. Because
    either division or subtraction could produce a result that
    is “within a predetermined range,” this language is con-
    ceivably broad enough to encompass either methodology.
    Zircon asserts that any question regarding the scope
    of claim 10 can be clarified using a number of claim
    6   When two equal values are divided, the result is
    one, or “unity.” When two equal values are subtracted,
    however, the result is zero.
    15                                        ZIRCON   v. STANLEY
    construction tools. First, citing the American Heritage
    Dictionary, Zircon contends that the plain and ordinary
    meaning of the term “ratio” is the “relation in degree or
    number between two similar things.” Appellant’s Br. 19–
    20 (citing Am. Heritage Dictionary 1028 (2d. College Ed.
    1991)). This definition lends scant support to Zircon’s
    proposed construction, however. There is an alternative
    definition for ratio in that same dictionary that defines it
    as “the relative size of two quantities expressed as the
    quotient of one divided by the other” — i.e., the result of
    division. Am. Heritage Dictionary 1028. That definition
    is classified, moreover, as the mathematical definition of
    the term which, in the context of this patent, we find to be
    the more pertinent definition to apply. Thus, to the
    extent that reference to dictionary definitions of this
    disputed claim term is helpful, we find such reference
    supports Stanley’s construction, not Zircon’s.
    Zircon next asks us to focus on the language of the
    representative claims as well as those which depend
    therefrom. Zircon argues that application of basic princi-
    ples regarding the primary importance of the claim lan-
    guage, as well as the doctrine of claim differentiation,
    virtually mandate the broad construction for “computing a
    ratio” which it urges. J.A. 14–19. Pointedly, Zircon
    argues that the district court’s construction of “computing
    a ratio” renders dependent claims 21 and 22 superfluous
    over claim 10 from which they depend, and renders them
    illogical. Claim 21 recites that the “computing a ratio”
    step of claim 10 is performed by “comparing the first and
    second capacitances” and “determining whether the first
    and second capacitances differ by less than a threshold.”
    
    Id. at col.
    22 ll.35–36. Stanley admits that this language
    denotes subtraction. J.A. 129–130 (“Clearly, to determine
    whether measurements ‘differ by less than a threshold,’
    are ‘within a predetermined value of each other,’ or ‘are
    ZIRCON   v. STANLEY                                       16
    within a[n] . . . absolute value from each other,’ the values
    are being subtracted from one another to determine the
    difference between them.”). On the other hand, claim 22
    states that “computing a ratio” between the two capaci-
    tances is done by “determining whether the ratio is within
    a range of inclusively 0.9 to 1.1.” ’091 patent col.22 ll.42–
    45. If the ratio is within this range, the sensor has de-
    tected the centerline of the stud. 
    Id. at col.
    22 ll.45–48.
    This claim denotes division. On the basis of these de-
    pendent claims, Zircon asserts that the term “ratio” as
    used in claim 10 must encompass values derived by way
    of either division or subtraction.
    Zircon’s argument has some superficial appeal. In
    Phillips, we made clear that the claims of a patent are not
    to be unduly limited where their plain meaning does not
    allow for such 
    limitation. 415 F.3d at 1324
    –27. And, in
    that case, we repeatedly referenced the doctrine of claim
    differentiation to avoid importing limitations from de-
    pendent claims of the patents in suit into the term “baffle”
    in the asserted independent claim. 
    Id. Indeed, the
    pre-
    sumption arising from claim differentiation is a strong
    one when the very limitation one seeks to import into an
    independent claim appears in a claim dependent there-
    from. Liebel-Flarsheim Co. v. Medrad, Inc., 
    358 F.3d 898
    ,
    910 (Fed. Cir. 2004).
    Despite this facial appeal, Zircon’s argument ulti-
    mately fails, however, because the claims are simply not
    drafted as neatly as Zircon would have this court believe.
    Indeed, when read in the context of the other claims, the
    dependent claims are virtually incomprehensible on their
    face and certainly do not follow the pattern we examined
    in Phillips. 7
    7   We note, moreover, that the doctrine of claim dif-
    ferentiation, while a useful construction tool, does not
    17                                        ZIRCON   v. STANLEY
    Zircon focuses on claims 21 and 22 and implies that
    they depend directly from claim 10. Thus, Zircon argues,
    claim 10 claims “computing a ratio” via either subtraction
    or division, while claim 21 claims “computing a ratio” only
    via subtraction and claim 22 claims doing so only via
    division. What Zircon first ignores is that claims 21 and
    22 do not depend from claim 10; they depend from claim
    19 which, in turn, depends from claim 10. Claim 19
    reads:
    The method of claim 10, wherein the first and a
    second plates are of approximately equal areas,
    the computing a ratio comprising: comparing the
    first and second capacitances; and repeating the
    acts of measuring and comparing.
    ’091 patent col.22 ll.22–25.
    To be valid under 35 U.S.C. § 112, ¶ 4, claim 19 must
    be read to claim all of the steps of claim 10 and contain a
    further limitation thereof. “To establish whether a claim
    is dependent upon another, this court examines if the new
    claim both refers to an earlier claim and further limits
    that referent.” Monsanto Co. v. Syngenta Seeds, Inc., 
    503 F.3d 1352
    , 1357 (Fed. Cir. 2007) (citing § 112, ¶ 4) (“[A]
    claim in dependent form shall contain a reference to a
    claim previously set forth and then specify a further
    limitation of the subject matter claimed. A claim in de-
    pendent form shall be construed to incorporate by refer-
    ence all the limitations of the claim to which it refers.”).
    And, claims 21 and 22 must further limit claim 19.
    otherwise trump the clear import of the intrinsic evidence.
    Laitram Corp. v. Rexnord, Inc., 
    939 F.2d 1533
    , 1538 (Fed.
    Cir. 1991) (quoting Autogiro Co. of Am. v. United States,
    
    384 F.2d 391
    , 404 (Ct. Cl. 1967) (“Claim differentiation is
    a guide, not a rigid rule. If a claim will bear only one
    interpretation, similarity will have to be tolerated.”)).
    ZIRCON   v. STANLEY                                      18
    Because claims 21 and 22 purport to describe the
    manner in which the “comparing” step of claim 19 is
    undertaken, it is that “comparing” language which must
    constitute the further limitation from claim 10 under §
    112, ¶ 4. Because claim 22 defines “computing a ratio” as
    a sub-set of “comparing” the capacitances, moreover,
    claim 19’s “comparing” limitation is both broader than
    and includes “computing a ratio.” Thus, the “comparing”
    step of claim 19 cannot be the same as the “computing a
    ratio” step in claim 10, as Zircon seems to urge. And, if it
    is indeed an added limitation — presumably requiring a
    second calculation — it would only be that second calcula-
    tion to which claims 21 and 22 refer, and impose further
    limitations. 8
    While Zircon argues that this reading of its claims is
    illogical, any other reading of the claims would render
    them invalid under § 112, ¶ 4. See Michilin Prosperity Co.
    v. Fellows Mfg. Co., 
    450 F. Supp. 2d 35
    , 40 (D.D.C. 2006)
    (holding that, because of § 112, ¶ 4, dependent claim 4
    must have three switches because independent claim 1,
    from which it depends, “recites two switches, and claim 4
    recites an additional ‘single switch.’ Claim 4’s single
    switch, then, must be construed as a switch in addition to
    the paper touch switch and disc touch switch.”). It ap-
    pears that claim 19 was meant to be an independent
    claim from which claims 21 and 22 would flow, allowing
    Zircon to capture a broad “comparing” methodology, which
    8    It seems that Zircon believes that a dependent
    claim can provide a new limitation that acts as a substi-
    tute for a limitation contained in the independent claim.
    The law does not support that position. See Pfizer, Inc. v.
    Ranbaxy Labs. Ltd., 
    457 F.3d 1284
    , 1291 (Fed. Cir. 2006)
    (holding that dependent claim 6 cannot cover salts of
    atorvastatin acid because the claim it depends upon only
    claims atorvastatin acid and not the salts of atorvastatin
    acid).
    19                                          ZIRCON   v. STANLEY
    could employ alternate methods of calculating the com-
    parisons. Because claim 19 must include all the steps of
    claim 10, however, the comparing methodology it claims
    must be in addition to the “computing a ratio” in claim 10.
    The fact that the claims do not appear in the patent as,
    perhaps, they were originally intended is not a problem
    this court has the authority to remedy. Chef Am. Inc. v.
    Lamb-Weston Inc., 
    358 F.3d 1371
    , 1374 (Fed. Cir. 2004)
    (“This court, however, repeatedly and consistently has
    recognized that courts may not redraft claims, whether to
    make them operable or to sustain their validity.”).
    As the district court concluded, moreover, adopting
    Zircon’s construction would introduce even more inconsis-
    tency into other claims:
    [T]o accept [Zircon’s] argument would be to create
    much more inconsistency between the claims and
    to only selectively apply applicable patent law re-
    garding dependent claims. Patent claims 1, 17,
    22, 23, 31 and 32 expressly contemplate the in-
    vention finding a centerline when it detects a “ra-
    tio” of “approximately one” or some derivative
    thereof (i.e., 0.9-1.1). This claim language only
    makes sense if the invention is computing a ratio
    by dividing equal or almost-equal values to get a
    result of approximately one, and not if the values
    were instead being subtracted, because in the lat-
    ter case the result of subtracting two equal values
    would be zero. If “ratio” were interpreted to in-
    clude subtraction in each of those claims specify-
    ing that the result of the ratio should be close to
    one (i.e., in the range of 0.9 to 1.1) to determine a
    center-line, then the device would not function
    properly to indicate that it was centered over the
    stud in the very instance where it is most cen-
    tered.
    ZIRCON   v. STANLEY                                      20
    
    Zircon, 713 F. Supp. 2d at 892
    (internal citations omit-
    ted).
    For these reasons, the language of the asserted
    claims, even when considered in light of those that depend
    therefrom, does not compel the construction Zircon urges.
    We, thus, turn to the specification and prosecution history
    for guidance.
    Zircon asserts that the specification of the ’091 patent
    confirms that its proposed construction of “ratio” is cor-
    rect. Specifically, Zircon highlights the fact that the
    specification discloses embodiments of the ’091 patent
    that employ either division or subtraction to compare
    capacitances. Appellant’s Br. 21. Stanley disagrees. It
    argues that, rather than establish that the term “ratio”
    encompasses both division and subtraction, the specifica-
    tion clearly refers to division when it discusses the term
    “ratio” and uses other nomenclature to refer to subtrac-
    tion. In other words, Stanley argues that the specification
    does not interchangeably use the term “ratio” to describe
    the results of both the division and subtraction opera-
    tions. Appellee’s Br. 23. We agree with Stanley.
    The specification repeatedly and consistently uses the
    word “ratio” to describe division. For example, the brief
    summary of the invention explains that “a ratio of ap-
    proximately one may indicate a centerline of a stud or
    joist or similar member.” ’091 patent col.2 ll.48–49.
    Describing figure 5A,
    21                                         ZIRCON   v. STANLEY
    ’091 patent fig.5A, the specification indicates that, in
    some embodiments, capacitance measurements are used
    to calculate a ratio: “If the capacitance measurements are
    equal or the ratio is approximately equal to unity, both
    plates may be centered over the stud’s edge 102 and the
    centerline 304 of the sensor 300 may be centered over the
    centerline 101 of the stud 100.” 
    Id. at col.
    7 ll.30–33.
    Significantly, the specification even explains that a ratio
    is computed by dividing one capacitance measurement by
    the other. E.g., 
    id. at col.12
    ll.29–38; col.16 ll.39–42. For
    example, the ratio “may be computed as the smaller
    capacitance divided by the larger capacitance, thereby
    resulting in a ratio that is equal to or less than one.” 
    Id. at col.
    7 ll. 51–53. As these examples illustrate, the speci-
    fication consistently refers to “ratio” to denote a product
    arrived at by way of division.
    In contrast to the consistent use of the term to de-
    scribe division, the specification does not use the term
    “ratio” to describe subtraction. Instead, the specification
    uses terms such as “are within a predetermined value of
    each other,” or “absolute value from each other” to denote
    when subtraction is required. 
    Id. col.14 ll.51–52,
    65.
    ZIRCON   v. STANLEY                                       22
    Describing the comparison circuit that the ’091 patent
    uses to compare capacitances, the specification states:
    [C]omparison circuit 414 may determine that the
    sensor is centered over a stud 100 by detecting
    that the capacitance measurements are equal to
    each other and also above a floor threshold. Ca-
    pacitance measurements may be considered equal
    when they are within a predetermined percentage
    value or absolute value from each other.
    
    Id. at col.1
    4 ll.60–66. This example posits two different
    methods the comparison circuit can employ to compare
    the capacitances: (1) calculating a ratio, i.e., division; or
    (2) determining whether the two values are within a
    predetermined value of each other, i.e., subtraction. The
    specification confirms that the comparison circuit oper-
    ates utilizing two distinct methods, stating, “[f]or exam-
    ple, the comparison circuit may compute a ratio between
    the capacitive measurements. The comparison circuit
    may determine whether the capacitive measurements are
    within a predetermined value of each other.” 
    Id. at col.1
    4
    ll.48–52. The specification, therefore, consistently uses
    the term “ratio” to denote division and other terms to
    refer to subtraction.
    Despite the specification’s explicit distinction, Zircon
    argues that its construction is correct because the specifi-
    cation discloses embodiments that utilize division or
    subtraction.     Zircon’s relies on Boehringer Ingelheim
    Vetmedica, Inc. v. Schering-Plough Corp., 
    320 F.3d 1339
    (Fed. Cir. 2003), in support of this argument. In Boe-
    hringer, the specification described the term “isolating” as
    defining a process that occurred in two different stages of
    growing a virus. 
    Id. at 1347.
    In light of this fact, the
    court held that isolating could not be construed narrowly
    to cover “isolating” at only one of these stages. 
    Id. 23 ZIRCON
      v. STANLEY
    Zircon’s argument misses the point. Unlike in Boe-
    hringer, the ’091 patent does not use the term “ratio” to
    denote subtraction. In this case, the question is: What
    does “ratio” mean in the context of the ’091 patent? The
    fact that the specification discloses both division and
    subtraction methodologies does not necessarily mean that
    “ratio” means either division or subtraction. The key to
    answering this question is determining how the ’091
    patent uses the term “ratio.” As discussed above, the ’091
    patent only uses the term “ratio” when it means division.
    The specification never uses the term “ratio” to describe
    subtraction. “Ratio” cannot, therefore, be construed to
    encompass both division and subtraction, regardless of
    whether both methodologies are otherwise disclosed in the
    specification.
    The prosecution history also supports this conclusion.
    Prosecution history “cannot be used to limit the scope of a
    claim unless the applicant took a position before the PTO
    that would lead a competitor to believe that the applicant
    had disavowed coverage of the relevant subject matter.”
    Schwing Gmbh v. Putzmeister Aktiengesellschaft, 
    305 F.3d 1318
    , 1324 (Fed. Cir. 2002). As discussed in the
    background section, during the prosecution of the ’091
    patent, original claim 32 broadly claimed comparing the
    capacitances. This claim was amended, however, to
    depend upon a claim that more narrowly required “calcu-
    lating a ratio.” While Zircon is correct that the record is
    unclear about why this amendment was made, it is clear
    that Zircon knew how to broadly claim division or sub-
    traction as computing methods, yet it amended all of its
    independent claims to require “calculating” or “computing
    a ratio.” One of ordinary skill in the art reading the file
    wrapper of this patent would believe that Zircon dis-
    avowed subtraction from the scope of its claims. This
    prosecution history, therefore, supports the construction
    ZIRCON   v. STANLEY                                        24
    of “ratio” to encompass only division because it establishes
    that compare meant either division or subtraction, while
    “ratio” means only division.
    In light of the foregoing, we agree with the district
    court that the proper construction of “ratio” is the “result
    of dividing two values.”
    II.
    We review de novo the district court’s grant of sum-
    mary judgment of no literal infringement. On appeal,
    Zircon does not argue that Stanley’s accused devices
    infringe the ’091 patent under the district court’s con-
    struction of “ratio.” It is undisputed, moreover, that
    “Stanley’s Accused Stud Finders compare capacitances by
    subtraction.” Appellant’s Br. 25. Because we conclude
    that the district court correctly construed “ratio” to en-
    compass only division, and it is undisputed that Stanley’s
    products use subtraction, we affirm the district court’s
    entry of summary judgment of no literal infringement.
    In light of this conclusion, we turn to Zircon’s claim of
    infringement under the doctrine of equivalents.
    III.
    The disclosure-dedication rule bars a finding of in-
    fringement under the doctrine of equivalents “when a
    patent drafter discloses but declines to claim subject
    matter. . . . [T]his action dedicates that unclaimed subject
    matter to the public.” Johnson & Johnston Assocs. Inc. v.
    R.E. Serv. Co., 
    285 F.3d 1046
    , 1054 (Fed. Cir. 2002). This
    rule is based upon the principle that “[a]pplication of the
    doctrine of equivalents to recapture subject matter delib-
    erately left unclaimed would ‘conflict with the primacy of
    the claims in defining the scope of the patentee’s exclusive
    right.’ ” 
    Id. Disclosure in
    the specification is sufficient to
    invoke this rule:
    25                                         ZIRCON   v. STANLEY
    [I]f one of ordinary skill in the art can understand
    the unclaimed disclosed teaching upon reading
    the written description . . . . This “disclosure-
    dedication” rule does not mean that any generic
    reference in a written specification necessarily
    dedicates all members of that particular genus to
    the public. The disclosure must be of such speci-
    ficity that one of ordinary skill in the art could
    identify the subject matter that had been dis-
    closed and not claimed.
    PSC Computer Prods., Inc. v. Foxconn Int’l, Inc., 
    355 F.3d 1353
    , 1360 (Fed. Cir. 2004).
    Here, Zircon does not dispute that the specification
    adequately disclosed the subtraction method. Instead,
    Zircon argues that the district court’s application of this
    rule was incorrect because the district court’s construction
    of “ratio” incorrectly limited the term to only division.
    Appellant’s Br. 26–27 (“However, as 
    discussed supra
    ,
    Zircon did in fact claim subtraction. Thus, the claims are
    consistent with the scope of the invention disclosed in the
    specifications and disclosure-dedication does not apply.”)
    As discussed above, however, the district court correctly
    construed “ratio.” Under this construction, the ’091
    patent does not claim a method or device for locating
    studs by subtracting capacitances. On these facts, it is
    clear that Zircon disclosed the subtraction method but
    failed to claim this method or an apparatus using it in its
    initial calculations in the ’091 patent. Because the dis-
    trict court properly granted summary judgment of non-
    infringement under the doctrine of equivalents, we affirm.
    Because resolution of this issue is an independently
    sufficient reason upon which to affirm the district court’s
    entry of summary judgment of no infringement under the
    ZIRCON   v. STANLEY                                 26
    doctrine of equivalents, we will not address Zircon’s
    arguments relating to prosecution history estoppel.
    CONCLUSION
    Because the district court correctly construed the
    term “ratio,” and properly determined that there was no
    infringement either literally or under the doctrine of
    equivalents, we affirm its judgment.
    AFFIRMED