657 F. Supp. 935 | N.D. Ill. | 1987
INPRO, INC., and David Orlowski, Plaintiffs,
v.
A.W. CHESTERTON COMPANY, INC., Defendant.
United States District Court, N.D. Illinois, E.D.
*936 Y. Judd Azulay, Azulay & Azulay, P.C., Chicago, Ill., for plaintiffs.
William J. McKenna, Hopkins & Sutter, Chicago, Ill., Charles C. Winchester, John M. Skenyon, Robert C. Nabinger, Fish & Richardson, Boston, Mass., for defendant.
ORDER
BUA, District Judge.
This order concerns defendant's motion for partial summary judgment pursuant to Fed.R.Civ.P. 56(c). Defendant maintains that Claims 1, 2, and 4 of plaintiffs' U.S. Patent No. 4,022,479 are invalid under 35 U.S.C. § 103 (obviousness) and 35 U.S.C. § 112 (failure to particularly point out and distinctly claim that which constitutes the invention). For the reasons stated herein, defendant's motion is granted.
I. FACTS
Plaintiffs Inpro, Inc. (Inpro) and David Orlowski, President of Inpro (Orlowski), instituted this action against defendant A.W. Chesterton Company, Inc. (Chesterton), alleging infringement on certain claims asserted in two patents owned by plaintiffs known as U.S. Patent No. 4,002,479 ('479 patent) and U.S. Patent No. 4,446,620 ('620 patent). Chesterton responds by denying infringement and counterclaiming for a declaration that the two patents are invalid. The '620 patent is owned exclusively by Orlowski, while the '479 patent is the property of Inpro. Since the filing of this action, plaintiffs voluntarily resubmitted the '620 patent to the U.S. Patent and Trademark Office (PTO) for re-examination and withdrew their claims of infringement regarding this patent. Thus, the dispute presently pending before this court concerns only those claims Chesterton is alleged to have infringed: Claims 1, 2, and 4 of the '479 patent.
The subject matter of the '479 patent concerns an allegedly novel design of a labyrinth seal. Labyrinth seals are used in bearing housings to prevent the egress of lubricating fluid and the ingress of contaminants. Often, labyrinth seals are used in conjunction with rotary pumps. A rotary pump is a type of pump in which the pumping action is created by a rotating drive shaft connected to motor. A stationary housing holds the bearings on which the drive shaft rotates together with a lubricant for the bearings. To prevent the lubricant from escaping from the housing and contaminants from entering the housing, a seal is employed. Depending on the nature and use of the pump, a number of types of seals may be used to contain and protect the bearing housing's lubricant from the external environment. Lip seals of felt are commonly employed to perform this sealing function. Felt lip seals typically involve the use of a fiberous ring made of wool and cotton that is affixed to the bearing housing so as to come into contact with the drive shaft as it rotates. The fiberous lip of the ring surrounds the circumference of the rotating shaft and affects a barrier to external contaminants as well as a containment for the housing lubricant. Because the life and effectiveness of such a friction based seal is somewhat limited, use of a lip seal is not always desirable.
Labyrinth seals are a second type of seal which is employed in a bearing housing to prevent the egress of lubricant and the ingress of contaminants. Labyrinth seals are noncontact seals and thus do not pose the friction or wear problems associated with lip seals. According to the prior art submitted by the parties, patented labyrinth seals have existed since at least 1939. See e.g., Young, U.S. Patent No. 2,281,905 (filed April 14, 1939).
The term "labyrinth" derives from the Greek word labyrintros and means a maze or obstructed passageway. Although labyrinth *937 seals can consist of many individual components, the type at issue in this case consists of only two basic parts: (1) a ring which is mounted on the stationary bearing housing; and (2) a cap which is fixed to and rotates with the revolving drive shaft. The stationary ring typically has one or more deep recesses in which a corresponding flange or set of flanges protruding from the rotating cap reside. When properly installed, the two parts fit together but never come into contact. Accordingly, friction is greatly reduced or eliminated. Yet, because the ring and cap do not contact, a potential exists for leakage through the narrow path between the parts. This leakage is minimized by the labyrinth arrangement of the path from which the seal gets its name. Fluids or particles traveling in such a path are less likely to enter the bearing housing if they encounter changes in direction or periodic pressure drops (caused by the path widening with its resulting change in flow velocity). The potential leakage path in any labyrinth seal includes at least one significant direction change and usually one pressure drop area.
The labyrinth seal disclosed by the '479 patent is reproduced in Appendix A. The first piece of the '479 patent is the ring member 40 which is fixed to the bearing housing 20 by means of an O-ring 45. The ring 40 does not touch the drive shaft 27. The second part of the '479 seal is the cap 41 which is attached to the drive shaft 27 by means of an O-ring 51. The cap rotates with the revolving drive shaft. The feature claimed in the '479 patent relating to preventing the seepage of contaminants from outside the bearing housing is the annular flange 52 on the cap which extends into the annular recess 48 on the ring cooperating with an exit hole 56 on the underside of the ring. As previously noted, the surfaces of the ring and cap do not touch. Fluids or particles from the housing's exterior can enter the gap between the ring and cap. Due to the design of the flange 52 and recess 48, however, any such seepage must make several direction changes before reaching the interior of the housing. The result is that the labyrinth minimizes the flow of outside contaminants through this path when the drive shaft is at rest. While the drive shaft revolves, the centrifugal action of the flange and cap prevents contaminants from entering through the narrow gap between the seal parts. Any fluids or particles which do find their way into the narrow gap are then expelled through the exit hole 56 which leads to the exterior of the bearing housing.
The second function the '479 seal serves is preventing the leakage of lubricants from the bearing housing along the drive shaft. The '479 seal minimizes this leakage by a series of three grooves 46 around the inside of the stationary ring 40. These grooves 46 are connected at the bottom by a drainback groove 47, which is connected to the interior of the bearing housing 20. When the drive shaft rotates, lubricant which creeps along the surface of the shaft enters the grooves 46 and experiences a velocity and pressure drop. As a result, the creeping lubricant migrates from the shaft into the grooves and is returned to the housing reservoir by means of the drainback groove 47. Thus, leakage of lubricant from the bearing housing is prevented.
II. DISCUSSION
A. Summary Judgment and Burdens of Proof
Summary judgment for a movant is appropriate under Rule 56 of the Federal Rules of Civil Procedure "only if the pleadings, depositions and affidavits fail to disclose a genuine issue of material fact." Gracyalny v. Westinghouse Electric Co., 723 F.2d 1311, 1316 (7th Cir.1983). In deciding the motion, the court must "resolve all doubts against the party seeking summary judgment." Id. Although Rule 56 applies to patent cases no differently than other types of cases, caution should be exercised in granting a motion for summary judgment in a patent case due to the need for expert testimony. SRI Int'l. v. Matsushita Elec. Corp. of America, 775 F.2d 1107, 1116 (D.C.Cir.1985) (in banc).
Once the PTO issues a patent for a certain subject matter, a presumption of validity *938 attaches. 35 U.S.C. § 282. This presumption attaches to each claim independently of the other claims. Id. The statute allocates the burden of establishing invalidity on the party challenging the patent. In carrying this burden, the challenging party must present clear and convincing evidence of invalidity. The ultimate burden of persuasion never shifts from the party asserting invalidity. Medtronic, Inc. v. Cardiac Pacemakers, Inc., 721 F.2d 1563 (Fed.Cir.1983).
Chesterton's counterclaim asserts that Claims 1, 2, and 4 of plaintiffs' '479 patent are invalid on two grounds. First, defendant argues the subject matter described by the three claims was obvious to those skilled in the art prior to the time the patent application was filed. 35 U.S.C. § 103. Second, defendant asserts Claims 1, 2, and 4 of the '479 patent fail to particularly point out and distinctly claim that which the alleged inventor considered as his invention. 35 U.S.C. § 112. These contentions will be addressed in turn.
B. Obviousness Under 35 U.S.C. § 103
Whether claims asserted in a patent are obvious to persons skilled in the art pertaining to the subject matter of the alleged invention is a question of law. Union Carbide Corp. v. American Can Co., 724 F.2d 1567, 1573 (Fed.Cir.1984). This legal conclusion is to be based on the factual inquiry outlined in Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S.Ct. 684, 693-694, 15 L.Ed.2d 545 (1966). According to the Court in John Deere, a determination of obviousness must rest on an analysis of three primary factors: (1) the scope and content of the prior art; (2) the differences between the prior art and the claims at issue; and (3) the level of ordinary skill in the pertinent art. In applying this test, the court is to decide whether the differences between the prior art and the claims at issue considered as a whole would be obvious to one possessing ordinary skill in the relevant art. John Deere, 383 U.S. at 17, 86 S.Ct. at 694. A determination of obviousness can also be influenced by certain secondary considerations. Id; Stratoflex, Inc. v. Aeroquip Corp., 713 F.2d 1530, 1538 (Fed.Cir.1983). These auxiliary factors include commercial success, long felt but unsolved need, failed attempts by others to achieve the claimed results or copying or acquiescence by competitors. The existence of secondary considerations is often the most probative evidence upon which a decision of obviousness can rest. Id.
The test under § 103 is not whether an improvement or use set forth in a patent would have been obvious or nonobvious. The test is whether the claimed invention, considered as a whole, would have been obvious or nonobvious to one of ordinary skill in the art. Carl Schenck A.G. v. Nortron Corp., 713 F.2d 782, 785 (Fed.Cir. 1983). Failure to consider the claimed invention as a whole is an error of law. W.L. Gore & Assoc., Inc. v. Garlock, Inc., 721 F.2d 1540, 1548 (Fed.Cir.1983) (error in considering claims in less than their entireties).
With the foregoing standards and considerations in mind, this court turns to first of three disputed claims in the '479 patent. Claim 1 of the '479 patent reads as follows:
1. A ring seal between a fixed housing and a rotating shaft comprising a first ring member having one end facing internally of the housing and a second end facing externally of the housing and being fixed and sealed to the housing and having a series of internal labyrinth annular grooves disposed adacent [sic] the shaft and an axially extending groove joining the annular grooves and opening to said first end and in communication with the housing, said first ring member further having an annular recess extending axially from and opening to said second end; and a second ring fixed to rotate with the shaft and having an annular flange extending axially from inner and outer radial surfaces and adapted for insertion in the annular recess with said radial surfaces bearing against the second end of the first ring member.
The language of Claim 1 essentially describes a two-piece labyrinth seal (the "first ring" 40 being the stationary part affixed to the bearing housing with a recess 48 and the "second ring" or cap 41 being attached *939 to and rotating with the drive shaft having a flange 52 designed to fit into the recess 48 of the first ring) with the groove 46 and drainback 47 features previously described for preventing the leakage of lubricant.[1] See Appendix A.
The parties do not appear to dispute the level of ordinary skill in the pertinent art. A person of ordinary skill in the act of labyrinth seal design is a graduate engineer with some experience in the field of seal design or a shop mechanic or machinist with many years of such experience. Defendant's Statement of Uncontested Facts ¶ 37.
In attempting to address the first two "primary factors" outlined in John Deere, Chesterton presents a series of examples of prior art together with deposition testimony of the inventor and co-plaintiff Orlowski to show each of the features asserted in Claim 1 existed in the art of labyrinth seal design prior to the January 2, 1976 filing of the '479 patent. With regard to the two-piece design of the '479 seal, Chesterton presents deposition testimony of Orlowski showing he was aware that certain labyrinth seals developed by the American Petroleum Institute (API) had incorporated the two-part concept prior to 1976. A review of that testimony indicates that Orlowski agreed that the concept of employing a stationary and rotating element to affect a seal was known in the prior art. (Orlowski Dep. p. 128.)
Similarly, Chesterton contends Orlowski's deposition testimony regarding the annular grooves 46 and drainback trough 47 in the stationary ring evidences Orlowski's knowledge that prior API seals had similar oil capturing grooves and return drains machined into the stationary portion of the seal. (Orlowski Dep. p. 118-119):
"Q Now, what I am asking is, did the API seals have similar labyrinth grooves in a stationary part?
"A Some do, some don't.
"Q Is your answer yes, some do?
"A Some do.
"Q Just talking about the ones that do, does when oil collects in one of these grooves, what happens to it in the API version of the labyrinth seal?
"A The grove inhibits the path and/or the exit of the lubricant.
"Q The API seal have any provision for getting the oil back from the grooves into the housing?
"A Some. Most have drilled holes at the root of the laby grooves.
"Q Do others have provisions?
"A Some have a milled trough.
"Q That create a groove similar to the Groove 47 shown in your Part 40?
"A I don't think I have ever seen one with a slope.
"Q They extended parallel to the shaft axis rather than sloping?
"A Yes.
"Q That was done prior to your invention?
"A Yes."
As to the last set of features described in Claim 1 concerning the flange 52 and corresponding recess 48, Chesterton argues that Orlowski's testimony recognized these features as part of the prior art (Orlowski Dep. p. 128):
"Q Then going down to line 66, it says a joint is provided between the two rings which is composed of an annular recess in the first ring that opens axially outward of the housing and an annular flange on the adjoining outer ring which fits with the annular recess of the first ring.
*940 "Now, as I understand it, the API designs included that concept also; is that correct?
"A Some of them to some extent."
Responding to this offer of proof, plaintiffs assert that Orlowski's cited deposition testimony is nothing more than a compilation of vague questions and answers which fails to adequately define the scope and content of the prior art so as to allow this court to determine if one skilled in the art could ascertain if the features in Claim 1 are obvious. Although this court does not believe Orlowski's testimony is as vague as plaintiffs assert, it nonetheless fails to give this court a sufficient basis to compare Claim 1 to the prior art. Chesterton does not point to one API seal which combines all the features described in Claim 1. The issue here is not whether the individual features existed in the prior art, but whether their combination was obvious in light of the prior art. Simply stated, Orlowski's deposition testimony does not provide a sufficient basis to judge whether Claim 1 is obvious under the John Deere analysis.
Next, Chesterton presents a series of diagrammed seals excerpted from various books and manuals published prior to 1972 to support the alleged obviousness of Claim 1. Each of the diagrammed seals allegedly illustrates the two-piece, flange-recess design recited in Claim 1 of the '479 patent. See Appendix B, Figures 1-3. Specifically, Figure 1 of Appendix B, which illustrates a seal excepted from V. Doughtie, Design of Machine Members 222 (4th Ed.1964), is alleged by Chesterton to embody all of the features contained in Claim 1 of the '479 patent.
Plaintiffs reject Chesterton's suggestion that Figures 1-3 in Appendix B illustrate the same type of two-piece labyrinth seal described in Claim 1. Plaintiffs note that Figures 2 and 3 suggest only a one-piece seal: a cap affixed to the rotating shaft which is fitted into a machined bearing housing. As the exhibits presented by Chesterton embodying these illustrations contain no discussion, this court is unable to determine whether a two-piece seal is, in fact, being pictured. Additionally, Figures 2 and 3 of Appendix B contain no drainback groove similar to groove 47 in the '479 patent. See Appendix A. Thus, what Figures 2 and 3 actually illustrate raise questions of material fact which this court is unable to resolve on a motion for summary judgment.
Plaintiffs next point out that the text in which Figure 1 of Appendix B appears does not describe the illustration as a labyrinth seal. Instead, plaintiffs note that the Doughtie text refers to Figure 1 as a "gasketed cover plate." The dictionary definition of gasket is "a piece or ring of rubber, metal, paper, etc. placed around a piston or joint to make it leakproof." Webster's New World Dictionary (College Edition, 1980). Plaintiffs contend that a gasket is simply a method of creating a seal by stuffing some material into the space to be sealed. Thus, plaintiffs assert that the illustration appearing in the Doughtie text by its very label does not describe a labyrinth seal, but some type of gasketed seal.
At first glance, the Doughtie text appears to illustrate a labyrinth seal similar to the '479 patent. However, Chesterton in its response memorandum, fails to explain why Doughtie refers to Figure 1 as a "gasketed cover plate" As noted by plaintiffs and not disputed by Chesterton, a gasketed seal is a friction based seal where some type of material is packed into the area sought to be sealed. Labyrinth seals are by definition noncontact seals. Resolving all doubts against the party moving for summary judgment, this court is unable to definitely state that the gasketed cover plate pictured in Figure 1 of Appendix B falls within the scope and content of the prior art of labyrinth seal design.
Plaintiffs assert that the validity of Claim 1 is further supported by the existence of certain secondary factors. Plaintiffs present affidavits from various individuals to show the '479 patent enjoyed commercial success and filled a long felt need. Plaintiffs' affiants state that labyrinth seals existing in the market prior to the '479 patent were not nearly as effective in keeping contaminants from entering the bearing housing or preventing lubricant *941 from escaping. Because of the increased efficiency the '479 patent offers users of labyrinth seals, the affiants assert the '479 seal has enjoyed considerable success in the market place. Finally, plaintiffs argue Chesterton acquiesced in the validity of the '479 patent. Plaintiffs claim that Mr. A.W. Chesterton contacted Orlowski to secure an agreement by which Inpro would manufacture and Chesterton would sell the '479 seals. Though no such agreement was ever concluded, plaintiffs assert Chesterton's actions further strengthen the presumption the '479 patent is valid.
Chesterton does not challenge the assertion that the '479 patent met with commercial success and filled a long felt need. Thus, these factors weigh in favor of the presumption of validity. As to plaintiffs' assertion of Chesterton's acquiescence, however, this court expresses some reservations.
The issue of public acquiescence is typically raised in a motion for a preliminary injunction against patent infringement pursuant to 35 U.S.C. § 283. In an attempt to prove the patent at issue is valid beyond question, the patentee will offer evidence of public acquiescence. Public acquiescence in the validity of a patent arises by a lack of substantial challenge to the monopoly rights of the patentee under circumstances which would support an inference that the patent has been examined and deemed valid. Zenith Laboratories, Inc. v. Eli Lilly & Co., 460 F.Supp. 812, 821 (D.N.J.1978). Establishing public acquiescence first requires proof that the patented item has been exploited commercially with great success. Second. the patentee must demonstrate the industry has respected the patent by noninfringement for a substantial period of time. Id. 8 Walker on Patents § 688 (Deller 2d Ed.1973). These requirements are designated to ensure that the industry possessed a motive to challenge the validity of the patent but decided upon careful examination that such efforts would prove fruitless. Carter-Wallace, Inc. v. Davis-Edwards Pharmacal Corp., 443 F.2d 867, 871 (2d Cir.1971).
In the present case, plaintiffs' assertion that Chesterton once approached Inpro to obtain a distributorship agreement does not address whether the industry has acquiesced in the validity of the '479 patent. Plaintiffs do not disclose whether the '479 patent has enjoyed undisputed recognition or has previously been the subject of infringement or invalidity litigation. Insufficient evidence is presented by plaintiffs to determine whether the doctrine of public acquiescence applies. Thus, this factor does not advance plaintiffs' position.
Viewing the evidence presented by Chesterton in the framework provided by John Deere, this court is unable to reach the conclusion that Claim 1 of the '479 patent is obvious as a matter of law. Orlowski's deposition testimony, although probative, does not conclusively establish that each of the features contained in Claim 1 existed in the prior art. Even if the deposition testimony could be read as such, the testimony does not make clear whether any prior API designs combined all the features of Claim 1 in a single seal. Thus, Orlowski's statements do not sufficiently establish the content of the prior art to allow this court to determine if any differences between the '479 patent and prior art existed.
Similarly, the illustrations of prior art submitted by Chesterton lack the clarity required for this court to declare a patent obvious on a motion for summary judgment. Based on the lack of description contained in the exhibits illustrating Figures 2 and 3 of Appendix B, it is uncertain whether one- or two-piece labyrinth seals are pictured. Since the two-piece feature to Claim 1 is essential to distinguishing the '479 patent from prior labyrinth seals, a material issue of fact is raised. Moreover, the characterization of Figure 1 of Appendix B as a "gasketed cover plate" presents difficulty in determining if such a device even falls within the scope of labyrinth seal design. Finally, plaintiffs' assertions that the '479 patent met with commercial success and filled a long felt need strengthens the presumption of validity. Although Chesterton's presentation of deposition testimony *942 and illustrations of prior art might well carry the day before the trier of fact, the issues of material fact raised prevent this court from granting its motion concerning Claim 1 on the grounds of obviousness. As Chesterton's arguments concerning Claims 2 and 4 are dependent upon this court finding Claim 1 invalid for obviousness, Chesterton's motion regarding the obviousness of Claims 2 and 4 is also denied.
C. Insufficient Claims Under 35 U.S.C. § 112
Chesterton's next attack on the validity of Claims 1, 2, and 4 of the '479 patent is based on the second paragraph of 35 U.S.C. §§ 112 which states in relevant part:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Chesterton directs this court's attention to certain statements made by Orlowski in his deposition which allegedly show several essential features required for the '479 patent to function properly were not incorporated in Claims 1, 2, and 4. Turning first to Claim 1, Chesterton presents the following excerpt of Orlowski's deposition which relates to a series of questions asking Orlowski what he regarded as his invention (Orlowski Dep. pp. 371-75):
"Q Were such compound seals new at the time you designed the seal that became that is shown in the '479 patent?
"A Compound labyrinth seals of various descriptions have been made as I submitted to you except that the specific arrangement as in '479 to my knowledge had never been accomplished.
"Q Which specific arrangement are you referring to or which features of the specific arrangement had never been accomplished before?
"A The-One very important one would be the method of contaminant expulsion.
"Q Providing the expulsion orifice?
"A In a specific arrangement, yes.
"Q Anything else? You might want to refer to some of the other drawings. I don't want to confine you to one.
"A I'd rather not discuss that patent as such.
"Q Well, I'd rather you did. Will you please answer the question.
* * * * * *
"A Okay. It will take a detailed discussion of the arrangement. Some features stand alone and others stand in combination with others.
"Q Well, why don't we Why don't you tell me which features stand alone first.
"A Well, the method of expulsion, that is, the slotting or vaning of the
"Q You are referring to the Slot 55 on the flange of the rotor?
"A Yes.
"Q That stands alone or does it work in combination with something else?
"A It works in combination with the expulsion arrangement.
"Q That was something different over the prior compound axial radial labyrinth seal we were talking about?
"A Yes. It made the product an absolute labyrinth rather than difficult path as
"Q Okay.
"A as before had been known. The depth of the trepanning grove No. 43 had never been to my knowledge ever attempted before.
"Q The depth of that groove?
"A The depth and function of that groove had never been exploited before to my knowledge.
"Q Well, let me ask you, what is significant about the depth of that groove? And what number is that, 4
"A 43.
"Q Or 48 perhaps?
"A Or 48.
"Q Okay. What's significant about the depth of that groove?
"A Well, if you don't have a deep enough groove, you don't have an opportunity to contain and then control and then expel contaminants as they appear.
"Q Is there some minimum depth for that groove?
*943 "A We think there is.
* * * * * *
"Q And what is the minimum depth that you feel is required for that groove?
"A I would say in the area of five hundred to six hundred thousandths of an inch
"Q Is that
"A would be optimum.
"Q Is that a depth you had worked out prior to the filing of the application for the 479 patent?
"A On first models that we used on tests had about five hundred thousandths and it seemed to work.
"Q That was your preferred depth at the time the application was filed?
"A That's right. That's what we felt to adequately accomplish our purpose."
As Chesterton notes, Orlowski identifies three features which he believed distinguished the '479 patent from prior labyrinth seals. These include the slots 55 in the flange of the rotating cap cooperating with the exit hole 56 to expel contaminants between the seal parts, and the recess 48 in the stationary ring having a certain minimum depth to contain and control contaminants. Additionally, when Orlowski was asked whether his invention would work satisfactorily without the exit hole 56, Orlowski responded that it would not. (Orlowski Dep. p. 137.) Chesterton essentially argues that because those features were not specified in Claim 1, it is invalid under § 112.
Responding to Chesterton's argument, plaintiffs advance two basic arguments. First, plaintiffs assert that Claim 1 must be read together with its dependent claims to determine if the requirements of § 112 are met. Second, plaintiffs contend that Claim 1 must be read in light of the drawings and specifications. These arguments will be addressed in turn.
In supporting their first argument, plaintiffs cite Application of Schutte, 244 F.2d 323 (C.C.P.A 1957) for the proposition Claim 1 must be read in light of Claims 2 and 3 which describe the exit hole 56 and slotted flange 55 features, respectively. Application of Schutte, however, only stands for the proposition that a dependent claim must be construed as including the claim on which it depends. This principle does not apply to Claim 1 since it is an independent claim. A feature called out in one claim is not read into any other claims except those dependent on it. An argument similar to that advanced by plaintiffs was made and rejected by the Federal Circuit in Kalman v. Kimberly-Clark Corp., 713 F.2d 760, 770 (Fed.Cir.1983):
KC argues that, in light of the Kalman disclosure, the independent claims must be read as limited to a process and apparatus which "[effect] movement" of a filter band or ribbon by differential hydrostatic pressure. The district court properly rejected this contention, for dependent claims 2 and 33 (not in issue) contain that very limitation, and it is settled and proper law that where some claims are broad and others narrow, the narrow claim limitations cannot be read into the broad whether or avoid invalidity or to escape infringement.
The second argument presented by plaintiffs rests on the proposition that descriptions of features ascertainable from the written specification or accompanying drawings are read into the claims. However, limitations in the written specification are, as a matter of law, not read into the claims which do not call out such features. Environmental Designs, Inc. v. Union Oil of California, Inc., 713 F.2d 693, 699 (Fed. Cir.1983). Moreover, while drawings accompanying the claims may be used as an aid in interpreting the claim, features not mentioned in the claim cannot be inferred from the drawing. Permutit Co. v. Graver Corp., 284 U.S. 52, 60, 52 S.Ct. 53, 55, 76 L.Ed. 163 (1931); see also 26 Fed.Proc., L.Ed. § 60:174 (1984).
Analyzing Claim 1 in light of the foregoing rules, it is clear that the language of the claim fails to identify the exit hole 56, slotted flange 55 or special depth of the recess 48. Since independent claims cannot be read in light of dependent claims, the features specified in dependent Claims 2 and 3 do not save Claim 1. The failure of *944 Claim 1 to mention the three stated features which work cooperatively to perform the expulsion function essential to the '479 seal's performance prevents any resort to the specification or drawings. Moreover, the minimum depth required for the recess 48 to properly contain and control contaminants is not mentioned anywhere in the written specification or discernable from the drawings. As Orlowski believed these three undisclosed features separated the '479 seal from the prior art, Claim 1 must be held invalid under § 112 for failing to point out and distinctly claim that which the applicant regards as his invention.
Dependent Claims 2 and 4 also suffer from the same defects. While dependent Claim 2 does add the exit hole feature to Claim 1, other deficiencies cause it to fail under § 112. The combination with Claim 1 still fails to recite the flange slot with which the hole works according to Orlowski, and it still contains, as part of defective Claim 1, a flange and recess arrangement without any reference to the minimum depth Orlowski believes is required for proper seal operation. As a result, Claim 2 does not comply with § 112. Claim 4 which is directly dependent on Claim 1 fails to call out the exit hole 55, slots 56 or the minimum depth for the recess 48. Claim 4 is also deficient in that it fails to mention an important formula developed by Orlowski designed to keep the O-ring used to attach the cap to the rotating shaft from deforming. During his deposition, Orlowski discussed what differentiated the O-ring aspect of the '479 patent from the prior art (Orlowski Dep. pp. 337-39, 341-43):
"Q Had you ever before you saw this article seen an O-ring drive connection of a rotor to a shaft other than your own?
"A Possibly.
"Q Well,
"A I think it's it has been done.
"Q How long ago do you think it was done?
"A Mechanical seals probably 30 years ago.
"Q You don't regard that as a contribution that you made to the art when you designed the seal in the 479 patent.
"A I don't know if there is a variation there or not. The O-ring groove relationship to the diameter of the O-ring and the protrusion of the O-ring may be peculiar to our usage of it.
"Q Well, at the time you filed your patent application for 479 patent, had you worked out some diameter or some relationships that are necessary for that O-ring groove in the O-ring?
"A Well, an O-ring generally is a sealing device and we are using an O-ring as a frictional drive member.
"Q Right.
"A And the parameters of the O-ring groove and the O-ring depth are important to use.
"Q When did you develop that relationship that you finally settled on as being satisfactory?
"A '76, '77.
"Q Prior to the application for the patent that the 479 patent?
"A Yes.
"Q What dimensions of the groove are important in that relationship?
"A Width and depth.
"Q Both the width and the depth? What function does the width or what is the importance of the width of the groove?
"A So that the O-ring is not allowed to deform in an axial direction.
"Q What about the depth, what's the
"A So that the O-ring has a prescribed amount of grip or frictional drive on the shaft.
"Q And what relationships did you develop? I mean, is there a particular formula that you follow or
"A Yes.
"Q followed? Is that a formula that you developed back at the beginning?
"A Yes.
"Q As I understand it, Mr. Orlowski, that formula you regard as being proprietary in Inpro; is that correct?
*945 "A I'd like to keep it that way as much as possible.
* * * * * *
"Q Well, I'm just asking you if you regard it, the formula, as being proprietary to your company. In other words, you have not previously published it
"A We don't
"Q to third parties?
"A We don't publish it.
* * * * * *
"Q Is the proprietary nature of the formula that reason the formula does not appear in your 479 patent?
"A Patent functionally describes it only.
"Q In what terms does it functionally describe it?
"A Describes the function of the O-ring drive.
"Q Well, it describes it as an O-ring and a groove that drivingly engages the shaft. But it doesn't describe your formula, does it?
"A No."
Accordingly, a critical formula essential to the proper functioning of the O-ring feature of Claim 4 is not only omitted from the claim but also kept out of the patent specification itself. Thus, it can hardly be said that the claims asserted against the defendant here, Claims 1, 2, and 4, "particularly point out and distinctly claim" what Orlowski regarded as his invention, as required by the statute. Instead, the claims omit critical limitations and elements, and generally, in an effort to be unduly broad, entirely avoid what Orlowski considered to be his contribution to the art. Accordingly, this court is compelled to find Claims 1, 2, and 4 invalid as a matter of law under 35 U.S.C. § 112.
III. CONCLUSION
For the foregoing reasons, defendants' motion for summary judgment is granted.
IT IS SO ORDERED.
*946 APPENDIX A
*947 APPENDIX B
NOTES
[1] Dependent Claims 2-4 of the '479 patent read as follows:
2. The invention defined in claim 1 further characterized by said first ring member having an opening on its underside opening to the annular recess and externally of the housing.
3. The invention defined in claim 3 further characterized by the first member having on its annular flange a slot that opens to the flange's outer surface and passes adjacent the opening in the first ring member as the shaft rotates.
4. The invention defined in claim 1 further characterized by the second ring member having an internal annular groove, and said second ring member is fixed to rotate with the shaft by means of an O-ring seated in the latter groove and engaging the shaft.
