Carl Zeiss Stiftung v. Renishaw PLC , 740 F. Supp. 1038 ( 1990 )

740 F. Supp. 1038 (1990)

CARL ZEISS STIFTUNG, Plaintiff,
v.
RENISHAW plc, Defendant.
RENISHAW plc, Plaintiff,
v.
CARL ZEISS STIFTUNG and Carl Zeiss, Inc., Defendants.

No. 87 Civ. 6748 (TPG).

United States District Court, S.D. New York.

July 5, 1990.

Stephen B. Judlowe, Roy C. Hopgood, Ira B. Winkler, Edward M. Reisner, Hopgood, Calimafde, Kalil, Blaustein & Judlowe, New York City, for Carl Zeiss Stiftung and Carl Zeiss, Inc.

James A. Oliff, William P. Berridge, Edward P. Walker, James C. Lydon, Oliff & Berridge, Alexandria, Va., for Renishaw plc.

OPINION

GRIESA, District Judge.

These are two consolidated actions involving claims of patent invalidity and patent infringement. In the first action Carl Zeiss Stiftung, a West German company, sues Renishaw plc, a British company. In the second action Renishaw is suing Zeiss Stiftung as well as Carl Zeiss, Inc., a United States subsidiary. For purposes of this opinion, the parties will be referred to simply as Zeiss and Renishaw.

The actions were tried to the court without a jury. This opinion constitutes the court's findings of fact and conclusions of law.

Facts

The action relates to a type of machinery used for the taking of highly precise measurements, called a "coordinate measuring machine," or "a CMM machine." A CMM machine can be automated so that a computer directs the measuring process and immediately records and prints the data received. Some CMM machines are capable of making measurements which are accurate to a few millionths of an inch. This equipment is expensive. The machines discussed in this case are generally sold for prices in the $150,000 to $400,000 range.

One essential part of a CMM machine is the probe. To the probe is attached a stylus and it is the tip of the stylus which actually makes contact with the surface of the object being measured, the "workpiece." This contact sets up certain reactions in the probe, which are conveyed to other parts of the machine and are highly important to the machine's measuring function.

Two or more contacts by the stylus/probe assembly are needed to make a measurement. If the desired measurement is, for instance, the diameter of a cylinder, the probe will make contact on one side of the cylinder and then be moved to make a contact on the other side of the cylinder. The distance between these points is the diameter.

It was formerly the case that the stylus was usually attached rigidly to the probe (the so-called "hard probe"). This was at a time when the measuring machines were normally operated by hand. The operator would manipulate the controls to move the probe to the desired position and that position would be recorded. However, automation brought about radical changes. In recent years automated machines have been developed involving the automated manipulation of the probe to establish numerous positions in very quick succession. The old-style hard probe, which was satisfactory for hand operations, was not suitable for the new automated machines. For one thing, if the stylus was rigidly connected and could not be deflected upon contact with the object being measured, there was great risk that the stylus would be broken. The need was to develop a probe with a stylus which would normally be deflected in some way upon contact with the object.

This need for deflection of the stylus gave rise to certain problems which led to the development of the Renishaw and Zeiss probes involved in this action.

Original Renishaw Probes

In the early 1970's David McMurtry, an Englishman who was then employed by Rolls-Royce, invented a "touch-trigger" probe which attempted to solve these problems. He used the concept of having "convergent surfaces" establish a "kinematic mount." This concept involves two objects which can move in and out of contact with each other. When the objects are brought into contact, this contact position is uniform and precise because the surfaces fit together at six points meticulously fashioned to prevent movement in any one of the six possible degrees of freedom which two bodies have with respect to each other. The concept of convergent surfaces to create a kinematic mount was age-old and in no way an invention of McMurtry, but he used these surfaces in an ingenious manner. McMurtry's convergent surfaces consisted of rings, in which there was an electrical circuit. The circuit was closed (i.e., the current flowed) when the convergent surfaces were seated. When the stylus was deflected upon contact with the object, and the convergent surfaces became unseated, this broke the circuit and a signal was generated, which caused the recording of the position of the probe at that instant. The electrical signal generated by the unseating of the convergent surfaces also stopped the forward drive of the machine. Thereafter the computer-driven motor directed the probe into position for its next contact.

The McMurtry probe was well adapted to automated measuring operations and proved to be enormously successful. McMurtry left Rolls-Royce and founded his own company, defendant Renishaw. Its headquarters is in Great Britain. Renishaw's sales of touch-trigger probes has grown from 35 probes in 1974 to 18,741 in 1989, of which about 12,000 were sold in the United States. Renishaw produces more than one model of touch-trigger probes, but by far the most successful is the "TP 2." The TP 2 is ½ inch in diameter and 1 3/8 inches in length. It normally carries a stylus about 7/8 of an inch long, making the whole probe/stylus assembly a little less than 2½ inches in length.

The McMurtry touch-trigger probes are not devoid of problems. These problems require some elaboration of the basic concepts of the CMM machine. In order to take a measurement, the probe and the rather large carriage to which the probe is attached move into position where the probe's stylus makes the contact with the workpiece. The idea is to determine the position of the tip of the stylus in space. This position is fixed with respect to an invariable reference point. More particularly, it will be a certain distance along the three Cartesian axes — the X, Y and Z axes. When the tip of the stylus makes contact, there are counters which record how far that tip has travelled, or been carried, from the reference point along each of the X, Y and Z axes.

This brings us to the problem, referred to earlier, of the need to have the stylus deflected somewhat upon contact with the workpiece. When the stylus is deflected, it creates movement in the "moveable part" of the probe. This movement is what unseats the convergent surfaces and creates the signal for the computer to record the travel distance along the X, Y and Z axes. But the fact that there is a slight delay from the precise moment of contact while the stylus is being deflected and until the signal is created — this circumstance means that the fixed part of the probe will travel slightly farther before the time of the signal than the exact point of contact. This is referred to as "pretravel" — i.e., the additional travel before the signal is given. Thus, at the time of the signal which measures the distance on the axes and fixes the position in space, the fixed part of the probe has travelled a little farther than the exact point to be measured — i.e., the point where the tip of the stylus contacts the workpiece. This difference is slight, but in the world of CMM machines, where great precision is needed, tiny variations become an issue.

Another related problem is "pretravel variation," arising from the fact that the pretravel may not be the same for each measurement. Such variation reduces "repeatability" and makes it difficult to compensate entirely for pretravel.

Where the stylus is short, as in the Renishaw TP 2 probe, the problems of pretravel variation and repeatability, while they exist to a small extent, are not of any real importance in the great bulk of the uses to which CMM machines are put. However, there are uses of CMM machines which require such extreme accuracy that these problems make the TP 2 less than optimal.

The shortness of the stylus of the TP 2 creates its own problem. Many times measurements must be taken within narrow and deep spaces. The stylus of the TP 2 cannot reach into such spaces. Renishaw has been able to compensate for this to some extent with extension devices for the probe, but the problem has not been entirely overcome. Also, it should be noted that the difficulties of pretravel and pretravel variation in a probe of the design of the TP 2 are magnified as the stylus becomes longer. Thus, there are definite limits to the length of a stylus which can be used on a TP 2 probe.

These considerations are summarized by Renishaw in its User's Guide for a probe of a different design:

Pretravel is the distance the probe travels between the stylus touching the surface and a trigger signal being sent to the CMM. Pretravel can be compensated by the measuring machine computer. However, due to the design of standard probes and the changes in force required to trigger from different directions, there is a small variation in this pretravel. This is called pretravel variation. On most CMM's using standard software and very long styli (e.g. up to 100 mm), pretravel variation can become a large factor in the probe's total margin for error.

Renishaw sells its probes, and certain related equipment, to manufacturers of CMM machines such as Brown & Sharpe Mfg. Co., Sheffield Measurement and LK Tool in the United States and other companies in various parts of the world. These companies sell the CMM machines to end users such as manufacturers of automobiles, aircraft, and a wide variety of other products.

McMurtry has obtained two patents in the United States relevant to the touch-trigger probes invented by him and marketed by Renishaw. U.S. Patent 4,153,998 (the "998 patent") was obtained on May 15, 1979. U.S. Patent 4,270,275 (the "275 patent") was obtained on June 2, 1981. The details of these patents will be discussed hereafter.

Zeiss Probes

Zeiss is a German company founded in the 19th century. It has a long history of manufacturing a variety of technically sophisticated products. For some years Zeiss has manufactured and sold CMM machines. It manufactures its own probes and, except in rare instances, uses only its own probes on its CMM machines. Also, with few exceptions, Zeiss does not market its probes to other manufacturers of CMM machines.

In 1973 Zeiss developed what is called an analog probe or a universal probe. The analog probe is large — about two feet long. The precise workings need not be described, except to say that it works on a different principle from what is involved in a touch-trigger probe. The analog probe is capable of the most extreme accuracy, greater than that of a touch-trigger probe. Also, it is capable of "scanning" a surface — i.e., registering a virtually continuous set of points along a surface. However, for the general run of measuring operations, the analog probe is not appropriate. In other words, the analog probe has its unique and very important uses, but it is no substitute for the touch-trigger probe.

In the late 1970's, Zeiss developed its own touch-trigger probe. The first model manufactured and used by Zeiss, and the one still most commonly used, is referred to as the Standard or ST. The ST probe is 4½ inches long and 3¼ inches in diameter.

The important feature of the Zeiss probe is that the signal is generated immediately upon contact with the workpiece, rather than waiting for the unseating of the convergent surfaces. This is achieved by the use of a piezoelectric element. When the stylus tip contacts a workpiece, the piezo element first senses an acoustic wave and then senses the pressure of the object on the stylus. This occurs in extremely rapid succession. The first time that either the acoustic wave or the pressure on the stylus reaches a certain level (called the "trigger level") the piezo will generate a position signal. As the probe moves forward following the generating of this signal, this movement causes the unseating of a kinematic mounting, generating a second electrical signal. This second signal is not used for purposes of fixing the position. It is used to "confirm" the validity of the earlier electrical signal as the position signal.

The confirmation operates as follows. The piezo element is extremely sensitive to noise. It is possible that the piezo might be activated and a signal generated by exterior noise such as the slamming of a door. If this is the case, the signal is not related to any measurement. Therefore, there is the need to confirm whether the signal represents an actual contact of the stylus with the workpiece. This confirmation is accomplished by the second signal involving the unseating of the convergent surfaces. If the second signal occurs within 100 milliseconds of the first signal, there is confirmation. Otherwise the first signal is rejected. However, the position of the probe at the time of the second signal is not recorded.

The unseating of the convergent surfaces performs another function — i.e., to stop the movement of the probe so that it can be sent to make another measurement. After these functions are carried out, the kinematic mount functions to bring the convergent surfaces and the stylus back to a precise rest position preparatory to the next measurement.

As is obvious from the above discussion, the Zeiss probe makes use of the concept of kinematic mounting involving convergent surfaces. Indeed, in the ST probe there are two sets of convergent surfaces. To some extent they perform functions that are the same as those performed in the Renishaw probe. In both the Renishaw and Zeiss devices, the unseating of the convergent surfaces generates a signal which is used to stop the probe so that it can be redirected. Also, the kinematic mount is used in both to return the convergent surfaces and the stylus to a precise rest position. However, in crucial respects Renishaw and Zeiss use the kinematic mounting differently. In the Renishaw probe the position signal is generated by the unseating. This does not occur in the Zeiss probe where the position signal is generated by the piezo. Also, the confirmation signal in the Zeiss probe has no counterpart in the Renishaw probe.

The fact that the position signal is generated immediately in the Zeiss probe virtually eliminates the problems of pretravel and pretravel variation involved in the Renishaw TP 2. This also means that a much longer stylus can be used in the Zeiss probe than in the TP 2.

Zeiss has recently developed its RST model. This makes use of the piezo and kinematic mount features in a manner basically similar to the ST. But it is smaller than the ST. The RST is 2¼ inches in length and 1 inch in diameter.

Zeiss has developed still another touch-trigger model using the piezo feature, called the ST 2. However, there have been no sales of this model in the United States, and a detailed description of it is unnecessary.

Zeiss obtained U.S. Patent 4,177,568 (the "568 patent") on December 11, 1979 for its touch-trigger probe design using the piezo feature.

As stated earlier, with rare exceptions, Zeiss only markets its probes in connection with the sale of its own CMM machines. During the years 1981-1989 Zeiss has sold a total of 154 CMM machines in the United States equipped with its touch-trigger probes, 151 with ST probes and 3 with RST probes, the latter sales being in 1989.

In connection with the machines carrying the ST model there have been a limited number of probes sold in addition to the one probe attached to the machine. The machines equipped with RST probes have more than one arm, and several probes are sold with each machine. The total number of Zeiss probes sold in the United States during the years 1981-1989 was 200, of which 178 were ST and 22 RST.

Later Renishaw Probes

Renishaw recognized the problems with its TP 2 probe and the corresponding advantages, for certain purposes, of a piezo probe. Therefore Renishaw set out to create its own piezo probe. This became its model TP 12.

The TP 12 uses the piezo element to immediately generate the position signal. It also uses kinematic mounting to confirm the signal, to stop the probe and to return the stylus to a precise rest position.

Originally Zeiss made a claim in this litigation that the TP 12 infringed its 568 patent. However, that claim has been withdrawn subject to renewal at a later time if the circumstances warrant. Thus the TP 12 is not at issue in this case.

However, Renishaw issued a User's Guide for the TP 12 which contains certain statements about the merits of piezo probes which are relevant to the issues remaining in the case relating to Renishaw's claims of infringement against Zeiss. One passage from the TP 12 User's Guide was quoted earlier and certain additional material from the Guide is now set forth.

The following description contrasts a "dynamic sensor," another term for a piezo element, with "standard probes," referring to Renishaw's earlier probes such as the TP 2.

The TP 12 probe works by using a dynamic sensor. This means that when the stylus on the end of the probe comes into contact with the surface to be measured, the dynamic trigger signal is sent immediately. This contrasts with normal method of working for standard probes, where the movement of the stylus displaces the patented seating mechanism, hence triggering the probe. Although this standard method is very accurate, the dynamic sensor is able to virtually eliminate the slight bending errors introduced when measuring using standard trigger probes.

Elsewhere in the User's Guide there is a statement regarding the advantage of a piezo probe in allowing the use of longer styli.

High accuracy when using long styli is one of the TP 12's outstanding features. It gives users of Coordinate Measuring Machines the opportunity to measure inside narrow bores etc. deep within a workpiece, using a combination of the longest probe extensions (e.g. 300 mm) and the longest styli (e.g. 100 mm), whilst still maintaining high accuracy and repeatability.

Renishaw has recently developed its model TP 7. It uses a semi-conductor instead of a piezo element. A detailed description will not be given because it is not involved in any patent claim in this case.

The Issues

Renishaw contends that the Zeiss touch-trigger probes — models ST, ST 2 and RST — infringe claims 2 and 15 of the 998 patent. Renishaw also contends that Zeiss infringes claims 1 and 3 of the 275 patent. Zeiss denies infringement of any of these claims. With respect to claim 3 of the 275 patent, Zeiss's defense is that the claim is invalid.

Legal Conclusions

The basic statute regarding infringement is 35 U.S.C. § 271(a).

Except as otherwise provided in this title, whoever without authority makes, uses or sells any patented invention, within the United States during the term of the patent therefor, infringes the patent.

A patent consists of a "specification," concluding with the "claims." 35 U.S.C. § 112. Technically, the claims are part of the specification, but the cases frequently speak of the specification and the claims as two different parts of the patent. In theory each claim describes an invention. Jones v. Hardy, 727 F.2d 1524 (Fed.Cir. 1984).

It is the law that infringement is determined by comparing an accused device, not with the actual physical invention of the patentee, or with the embodiments of the invention as described in the specification, but with the language of the claims. SRI Int'l v. Matsushita Elec. Corp., 775 F.2d 1107, 1121 (Fed.Cir.1985). To prevail in an infringement action, the patentee must show by a preponderance of the evidence that the accused has infringed a claim of its patent. Lemelson v. United States, 752 F.2d 1538 (Fed.Cir.1985).

An accused product may infringe a patent claim either literally or through the doctrine of equivalents.

Literal infringement requires a two-step analysis. First, the court must determine the meaning and scope of the claim. Second, the court decides whether the claim reads on the accused device. Literal infringement requires that an accused device embody every element of the claim. Mannesmann Demag Corp. v. Engineered Metal Prod. Co., 793 F.2d 1279, 1282 (Fed. Cir.1986).

Although it is the claim and not the specification which defines the invention for purposes of determining infringement, nevertheless the specification can be used in interpreting the claim. D.M.I., Inc. v. Deere & Co., 755 F.2d 1570 (Fed.Cir. 1985). Also, the history of the prosecution of the patent — the so-called "file wrapper history" — may be used in construing the claim. SRI Int'l v. Matsushita Elec. Corp., 775 F.2d 1107, 1118 (Fed.Cir.1985). Finally, the language of other claims may be used in determining the scope of the claim under consideration. Fromson v. Advance Offset Plate, Inc., 720 F.2d 1565 (Fed.Cir.1983).

There is a special statutory rule of construction, dealing with the situation where a claim describes how an operation is to be performed without describing the precise structure which is to accomplish this. This is referred to as "means plus function" language. In regard to such language, 35 U.S.C. § 112 provides:

An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof.

Thus, where a claim merely contains "means plus function" language, reference can be made to the specification to determine what structure is in fact embraced in the claim.

One oddity of patent law is that literal infringement can be based on a type of equivalence, which is held to be different from the "doctrine of equivalents" previously mentioned. The former arises under 35 U.S.C. § 112, which concludes with the phrase "and equivalents thereof." Thus, the statute provides that where a claim has "means plus function" language without a recital of the structure, the claim shall be construed to cover the structure described in the specification and equivalents of that structure.

This brings us to the doctrine of equivalents, which is judicially created and goes beyond what is involved in § 112. If the accused device does not fall literally within the patent claim, the court may nonetheless find it to infringe the patent under the doctrine of equivalents. Under this doctrine, if the accused device performs substantially the same function in substantially the same way to achieve substantially the same result as the patented invention, then the device is said to infringe. See Graver Tank & Mfg. Co. v. Linde Air Prod. Co., 339 U.S. 605, 608, 70 S. Ct. 854, 856, 94 L. Ed. 1097 (1950); Perkin-Elmer Corp. v. Westinghouse Elec. Corp., 822 F.2d 1528 (Fed.Cir.1987). The purpose of this equitable doctrine is to prevent the unscrupulous copyist from making unimportant changes which take the accused device outside the literal words of the claims. R. Harmon, Patents and the Federal Circuit 123 (1988).

Alleged Infringement of Renishaw 998 Patent

This patent lists David R. McMurtry as the inventor. It recites that the application for the 998 patent was filed on November 5, 1976, and that this application was a continuation of two previous applications filed in 1973 and 1975, which were abandoned.

The specification summarizes the invention as follows:

This invention relates to probes for use in measuring apparatus. In particular the invention relates to a probe for determining at what point in space contact is established between a stylus and an object.

The specification describes different embodiments of the invention illustrated by 14 drawings.

The basic principles common to all the embodiments are stated in an Abstract which defines the probe as:

A probe for determining at what point in space contact is made between an object and a stylus. The stylus is located in an accurately defined rest position relative to a housing and slight deflection of the stylus away from this rest position is detected preferably by electrical switches.

Every embodiment describes how the stylus is deflected from a rest position (obviously by contact with the object to be measured) and how this deflection is detected. When this occurs information is transmitted to a device (not itself involved in this patent) "to read and record the coordinates of the probe at that time." Over and over in the specification, in connection with each of the embodiments, and regardless of the differences in their structure, there is consistent invariable reference to the deflection of the stylus (or equivalent language such as "movement of the stylus away from its rest position") accompanied by detection of that event (or equivalent language such as "sensing" or "determining").

The first embodiment, illustrated by Figures 1-3, contains a kinematic mount formed by convergent surfaces. The position signal is generated when the deflection of the stylus unseats the convergent surfaces, breaking an electric circuit and providing a pulse to an automatic switching arrangement. This embodiment is basically what Renishaw manufactures as the TP 2 touch-trigger probe.

Other embodiments, illustrated in Figures 4-8, involve other methods of creating a rest position for the stylus, deflecting the stylus from this rest position and an electrical signaling system appropriate to this arrangement. Here the rest position is created by the interaction of two springs — one vertical and one horizontal, or by the constraint of the side wall of the fixed member.

There are other embodiments which have no relevance to the present case.

The specification also deals with the question of precisely how the signal is generated when the stylus is deflected. The preferred method is through the use of an electrical switching system. However, other means are described as follows:

The probes herein described have been provided with a detector comprising an electrical switching system but it will be appreciated that the electrical switching system could be replaced by fluidic or other known switches and that the detector may comprise an optical projection system for magnifying a mismatch between two pointers, the mismatch providing an indication of movement of the stylus.

Although this description is somewhat open-ended in referring to "other known switches," nevertheless there is no departure from the concept that the signal occurs upon the deflection of the stylus. Thus, this description ends by referring to the "movement of the stylus."

There is nothing in the specification in regard to any embodiment which describes a signalling device which is activated by a piezoelectric element or other means immediately upon the contact of the stylus with the workpiece prior to the stylus being deflected.

Turning now to the claims themselves, the two claims of the 998 patent allegedly infringed — claims 2 and 15 — are quoted in full in the Appendix to this opinion. Claim 2 starts with the following description:

A device for mounting a stylus in position-determining apparatus wherein said device and an object are movable relative to each other for providing a signal when said stylus engages said object thereby indicating the position thereof....

Claim 2 goes on to state that the device has a fixed member, and also a movable member to which the stylus is connected. There is then a description which refers to the fact that the fixed member and the movable member are connected by convergent surfaces which are unseated when a force is applied to the stylus and reseated on cessation of the force. The concluding description in claim 2 is of a

means for providing said signal when said movable member is removed from its rest position.

For our purposes, perhaps the most crucial point in claim 2 is that the signal is generated upon the unseating of the convergent surfaces. Thus (to use the language from the introduction), the device provides "a signal ... indicating the position" of the object and (referring to the concluding language) such a signal is provided "when said movable member is removed from its rest position." This is precisely what is referred to in the specification as deflection of the stylus.

The Zeiss probes do not infringe claim 2. They do not generate a signal upon the unseating of the convergent surfaces. They do not create a signal in any equivalent manner. Zeiss uses an entirely different method for generating the signal — an acoustic wave or pressure on the stylus which activates the piezoelectric element immediately upon the contact of the stylus with the workpiece.

The unseating of the convergent surfaces is used in the Zeiss probes to generate an electrical signal, but this signal is not for the purpose of fixing the position (the kind of signal spoken of in Claim 2). In the Zeiss probes the signal from the unseating of the convergent surfaces is used to confirm the validity of the piezo signal and to stop the probe movement so that it can move to the next measurement.

Turning to claim 15, the introductory description is:

A device for mounting a stylus in position-determining apparatus wherein said device and an object are movable relative to each other for engaging the stylus with the object and by such engagement sense the position thereof....

Claim 15 goes on to state that the device has a fixed member, and also a movable member to which the stylus is connected.

One of the principal differences between claim 15 and claim 2 is that claim 2 makes no express reference to electrical circuitry, whereas claim 15 does so. Thus, in claim 15 there are said to be "electric contact elements" in both the fixed member and the movable member. Claim 15 goes on to state that there is a "bias means" for urging the movable member into engagement with the fixed member so that all the electrical contact elements are connected and are in a "rest position."

Claim 15 further describes a "constraining means" — i.e., a "means acting directly on said movable member" for constraining such member against movement in a horizontal direction, although it is intended that, when a force is applied to the movable member (obviously referring to the stylus striking the workpiece), the movable member is displaced from the rest position. On this displacement the contact between the electrical elements is broken. When the force on the movable member ceases, the "bias means" and the "constraining means" cooperate to return the movable member to the rest position.

Finally, claim 15 states that there is an electric circuit connected to the electric contact elements which is "to change state" when the engagement between these elements is broken.

Renishaw argues that the Zeiss probes infringe claim 15. Renishaw contends that claim 15 is broader than claim 2 in a crucial respect — that it does not specify a precise type of signalling device. Renishaw recognizes, of course, the language in claim 15 about an electric circuit changing state when the engagement between the electric contact elements is broken. However, Renishaw notes that there is nothing in this description which states that this is the cause of the position signal, and contends that it is simply a general statement that, for some unspecified purpose, the electric circuit is broken when the contacts are unseated. Thus, according to Renishaw, this description is broad enough to cover the operation involved in the Zeiss probes — that of having the unseating of the convergent surfaces offer the confirming signal and cause the stopping of the movement of the machine.

Zeiss contends that claim 15 is narrower. In the first place, Zeiss argues that claim 15 is limited to the embodiment of the 998 patent illustrated in Figures 4-8 of the specification. This embodiment, as described earlier, involves the use of a spring or the wall of the fixed member to constrain horizontal movement of the movable member, rather than the use of convergent surfaces as illustrated in Figures 1-3. Zeiss argues that its probes in no way correspond with the embodiment illustrated in Figures 4-8. In addition, Zeiss argues that the language in claim 15 dealing with electrical circuitry must be read as having to do with the position signal. Zeiss thus interprets claim 15 as describing a position signal, and one that is different from what is involved in the Zeiss probes.

It is the court's view that Zeiss is correct in both of its arguments.

Zeiss's first argument relates to the "constraining means" portion of claim 15. Since this is a "means plus function" description, the court must look to the specification for purposes of interpretation. 35 U.S.C. § 112. Also, it is proper to look at the "file wrapper history" in construing the claim. SRI Int'l v. Matsushita Elec. Corp., 775 F.2d 1107, 1118 (Fed.Cir.1985).

Claim 15 describes a "bias means" and a "constraining means." The bias means operates vertically and the constraining means operates horizontally. The specification makes it clear that this description of the two "means" in claim 15 does not refer to the convergent surfaces kinematic mount in Figures 1-3, but to the other types of kinematic mount illustrated in Figures 4-8. The specification states:

In [the Figs. 1-3] embodiment it will be understood that the rest position of the stylus is both defined by and deflections from this rest post position detected by, the same means. In FIG. 4 an alternative embodiment is shown in which part of the task of defining the rest position of the stylus is achieved by means separate from the means for detecting deflection of the stylus away from this position.

During the prosecution of the application for the 998 patent, the applicant represented to the patent examiner that the claim which is now claim 15 (then claim 43) was directed to the embodiment of the invention shown in Figures 4-8.

The court finds that the "constraining means" portion of claim 15 refers to the structure illustrated in Figures 4-8 and does not refer to the structure shown in Figures 1-3. Since the Zeiss probes do not correspond with Figures 4-8, they do not infringe claim 15.

Renishaw points to dictum in Rolls-Royce Ltd. v. GTE Valeron Corp., 800 F.2d 1101, 1106 n. 5 (Fed.Cir.1986), indicating that claim 15 covers all the embodiments in the 998 patent. However, that statement cannot be taken as dispositive in the present action. Zeiss was not a party in Valeron. Moreover, the issue in the present case is sharply different from what was presented in Valeron.

Zeiss is also correct on the meaning of the language in claim 15 dealing with electrical circuitry. The introductory language in the claim specifies a device for mounting a stylus "in position-determining apparatus" in which the stylus engages an object, and by such engagement senses its position. The claim goes on to describe what the device is comprised of. The final element comprising the device is described as an electric circuit which changes state when the engagement (used in a different sense than in the introductory language) between the electric contact elements is broken.

The issue between the parties is whether the electric circuit, as described in claim 15, is or is not for the purpose of creating a position signal. Zeiss contends that it is. Renishaw contends that it has no such specific meaning.

It is clear that a device for sensing a position must include some element which creates a signal as to that position. There can be no position-sensing device without such a signal. If claim 15 were to constitute a description of everything other than the essential element, it would be a most odd attempt to describe an invention of a position-sensing device. The only possible way to make sense of this claim is to interpret the description of the electric circuit as referring to the position signal feature.

Therefore, claim 15 must be interpreted as describing an electrical signal for fixing the position, and that description refers to the unseating of the electric contact elements as generating that signal.

Since the Zeiss probes do not generate the position signal in this manner, they do not infringe claim 15. There is no infringement literally or under the doctrine of equivalents.

Alleged Infringement of Renishaw 275 Patent

The 998 patent was granted on May 15, 1979. Shortly before that date — on April 18, 1979 — McMurtry applied for another patent on his touch-trigger probe, which became the 275 patent and was granted on June 2, 1981.

The 275 patent has exactly the same specification as does the 998 patent. However, the claims are stated differently. The claims in 275 attempt to go as far as possible in eliminating any specific reference to a particular method of generating the position signal.

It would appear clear that Renishaw sought the 275 patent in order to deal with probes designed by other parties with different signalling methods.

Because of the relationship between the 275 patent and the 998 patent, Renishaw filed a "terminal disclaimer" with regard to 275, with the result that the 275 patent will expire on the expiration date of the 998 patent, May 15, 1996.

Renishaw alleges that Zeiss infringes claims 1 and 3 of the 275 patent.

Claim 1 of 275 is identical to claim 2 of 998, except for the final clause. Whereas claim 2 of 998 ends with the statement that the position signal is provided when the movable member is removed from its rest position, claim 1 of 275 closes with a reference to the position signal but in much more general terms. The language is:

means for providing said signal when said stylus engages the object.

Since this is a "means plus function" clause, reference must be made to the specification in order to interpret it. 35 U.S.C. § 112. The specification, as already noted, makes it clear that the position signal is generated when the stylus is deflected. This refers to the unseating of the convergent surfaces. Thus, despite the attempt to couch it differently, claim 1 of 275 is in substance the same as claim 2 of 998. Zeiss does not infringe either.

The final claim to be dealt with is claim 3 of the 275 patent, which commences:

A device for mounting a stylus in position-determining apparatus wherein said device and an object are movable relative to each other and the stylus is engageable with said object, said device comprising. ...

The claim goes on to describe the fixed member, the movable member, and the convergent surfaces used to seat and unseat the stylus. Claim 3 is completely devoid of any description referring to the position signal.

Zeiss's defense as to claim 3 is that it is invalid.

It has been held that where a claim fails to recite an essential element necessary to make the claimed combination operative, the claim is invalid. General Electric Co. v. United States, 572 F.2d 745, 755, 215 Ct. Cl. 636 (1978); Shelco, Inc. v. Dow Chemical Co., 322 F. Supp. 485, 517 (N.D. Ill.1970), aff'd, 466 F.2d 613 (7th Cir.1972). The General Electric case held this to be so even for a so-called "open" claim (i.e., one using the word "comprising" rather than the word "consisting"). However, in Bendix Corp. v. United States, 600 F.2d 1364, 1369, 220 Ct. Cl. 507 (1979), the court stated that it is not necessary that a claim recite each and every element needed for the practical utilization of the subject matter, as long as what is described in the claim has utility separate and apart from the other elements.

Claim 3 commences by referring to a "device for mounting a stylus in position-determining apparatus." The "device" is the probe. Claim 3 goes on to say that the device comprises certain things. But what is then listed does not describe an operable probe, and does not describe McMurtry's invention.

The McMurtry touch-trigger probe invention did not consist merely of moving the stylus back and forth between a rest and an unseated position, which is the function presented in claim 3. Such activity, in and of itself, cannot fix any position and cannot measure anything. There must be the position signal in order for there to be a probe usable for measuring purposes. As the specification points out repeatedly, the invention involves the two essential and inseparable functions of deflecting the stylus and detecting that deflection; there must be the unseating of the stylus and the generation of the signal caused by that unseating; and, of course, the stylus must return to its rest position thereafter.

The unseating and reseating function described in claim 3 has no utility separate and apart from the signalling function. The unseating of the convergent surfaces does not happen as some isolated event. This unseating breaks an electric circuit, thus creating a signalling impulse. To try to separate the unseating and the signal is wholly artificial and unrealistic. After the unseating, the convergent surfaces reseat, or resume their rest position. But again, this is not an independent event in and of itself. It involves reestablishing the electrical circuit so that the probe is ready for the next signalling operation.

In sum, the arbitrary presentation of part of an invention does not constitute a claim of a valid invention. Claim 3 is invalid.

Even if it is assumed that the seating and unseating operation can, by itself, be the subject of a claim, nevertheless claim 3 is invalid for obviousness. The relevant statute provides that a patent may not be obtained

if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains.

35 U.S.C. § 103. The cases have set forth factors to be considered.

Determining obviousness/nonobviousness under 35 U.S.C. § 103 involves factual inquiries into: (1) the scope and content of the prior art; (2) the level of ordinary skill in the art; (3) the differences between the claimed invention and the prior art; and (4) objective evidence of nonobviousness, e.g., long felt need, commercial success, failure of others, copying, unexpected results.

Perkin-Elmer Corp. v. Computervision Corp., 732 F.2d 888, 894 (Fed.Cir.1984). A patent is presumed valid, 35 U.S.C. § 282, and obviousness must be shown by clear and convincing evidence. Perkin-Elmer, 732 F.2d at 894.

Although claim 3 does not literally use the words "kinematic mount," it has been conceded by both sides, and assumed throughout the trial, that what is described in claim 3 is a kinematic mount created by convergent surfaces.

The concept of a kinematic mount involving convergent surfaces goes back at least to the 19th century. It is described in Mechanical Design of Laboratory Apparatus, a book written by H.J.J. Braddick of Manchester, England, published in 1960. Chapter 2 lists conditions for satisfactory operation of instruments containing a "number of parts which co-operate," one of these conditions being:

(a) Kinematic or geometrical — the parts must be held in the required relative positions or move relative to one another in the required way.

This is followed by a discussion of kinematics. Where two parts of an instrument have "mating surfaces" kinematics makes the relative position of the parts definite, subject to "deflexions produced by applied forces." The text asserts that the principle of kinematic or geometric design was stated formally by Maxwell in 1876 "and has been applied by instrument designers in increasing measure."

Braddick describes the fact that the position of one rigid body relative to another has "six degrees of freedom," and states:

It follows that if one element is to be located in a fixed position relative to another the kinematic design of the connexion between the two involves contact at six points properly chosen.

What this means is that six points on one surface converge with six points on the other surface. One of the illustrations in the Braddick text shows two round surfaces with three balls attached to one surface fitting into three slots in the other surface. Each ball in a slot forms two points of contact. The three balls/slots are 120 degrees apart. A spring may be used to hold the two pieces together when they are not being deflected out of the closed position.

Claim 3 of the 275 patent describes the fixed member and the movable member being constructed so as to form mutually convergent surfaces which will be removed from a rest position and returned to that position. The specification describes three locations of convergent surfaces 120 degrees apart, consisting of three members extending out from the stylus engaging three vee shaped slots formed by pairs of ball bearings. In this manner the six points of contact, described by Braddick as constituting a kinematic design, are formed. There is a "bias means," such as a spring, to urge the two parts together after the stylus has been deflected.

The inevitable conclusion is that Braddick disclosed in plain textbook terms what is described in claim 3 of the 275 patent. Braddick did not, of course, literally talk about using a kinematic mount in a probe for a CMM machine. But Braddick spoke generally of applying the principle in scientific instruments.

Zeiss contends that the court should also consider a German book on mechanical design —Automatologie by Wolfgang Schmid, published in 1952. This book illustrates and describes what is called a tracer. Part of this is the tracer spindle which runs over the surface being traced, and is analogous to the stylus in the probe of a CMM machine. The design in Automatologie shows a tracer spindle which is connected to a structure with three arms 120 degrees apart, through which there is an electric circuit. The spindle can be deflected and cause the breaking of the circuit.

There is no kinematic mount described in Automatologie. However, this text does illustrate the concept of a spindle (or stylus) in a position-sensing device, which deflects and returns to a rest position. Indeed Automatologie is important in teaching a method of mounting a spindle or stylus so that it can be deflected in any direction.

When Braddick and Automatologie are combined, there is teaching of the deflection (unseating) of a stylus in a position-sensing device, the reseating of that stylus, and a means for making the seated position definite and precise through kinematic mounting.

It must be concluded that someone ordinarily skilled in the art of designing precision measuring instruments had available to him in 1972 all the information necessary to lead to the concept of a kinematic mount involving convergent surfaces to solve the problem of the deflection and reseating of a stylus in a precise manner.

The case law requires consideration of "objective evidence" on the question of obviousness, such as long felt need and commercial success.

It has been vividly demonstrated that McMurtry's probe has satisfied a basic and important need in the precision measuring industry, and has been an enormous commercial success. However, claim 3 of the 275 patent does not describe McMurtry's probe. There is not the slightest evidence that what is set forth in claim 3 — something which can wander about seating and unseating without taking measurements — has answered any industrial need or has had any commercial success whatsoever.

Under all the circumstances, the court concludes that claim 3 is invalid for obviousness.

Conclusion

Renishaw has not proved that the Zeiss probes infringe claims 2 and 15 of Renishaw's 998 patent or claim 1 of the 275 patent. Zeiss has proved that claim 3 of the 275 patent is invalid and thus cannot be infringed by Zeiss.

The parties should submit a judgment in accordance with the rulings in this opinion.

SO ORDERED.

APPENDIX

Claim 2 of Patent for 4,153,998

2. A device for mounting a stylus in position-determining apparatus wherein said device and an object are movable relative to each other for providing a signal when said stylus engages said object thereby indicating the position thereof, said device comprising:

a fixed member,

a movable member to which said stylus is connectable, said movable member being supportable on said fixed member at a plurality of spaced-apart locations, one of the movable and fixed members having at each of said locations a pair of mutually convergent surfaces and the other one of said members being engageable with said convergent surfaces,

bias means for urging said movable member into contact with said fixed member so that all of said convergent surfaces are engaged thereby positively defining a rest position for said movable member, said movable member being removed from said rest position in opposition to said bias means when a force is applied to said stylus, said bias means and convergent surfaces co-operating, on cessation of said force, to return said movable member to said rest position, and

means for providing said signal when said movable member is removed from its rest position.

Claim 15 of Patent for 4,153,998

15. A device for mounting a stylus in position-determining apparatus wherein said device and an object are movable relative to each other for engaging the stylus with the object and by such engagement sense the position thereof, said device comprising:

a fixed member having an axis,

a movable member to which a stylus is connectable, first electric contact elements connected to said fixed member at three locations spaced apart around said axis,

second electric contact elements connected to said movable member in positions confronting said first elements in the direction of said axis,

bias means for urging said movable member into engagement with said fixed member at all said first and second electric contact elements thereby positively defining a rest position for said movable member,

means acting directly on said movable member for constraining said movable member against movement in a direction transverse to said axis when said movable member is in said rest position, said movable member being removable from said rest position in opposition to said bias means by a force applied to said movable member thereby to break engagement between said elements at at least one of said locations, said bias means and said constraining means cooperating to return said movable member to said rest position on cessation of said force, and

an electric circuit connected to said contact elements to change state responsive to engagement between said elements at at least one of said locations being broken.

Claim 1 of Patent 4,270,275

1. A device for mounting a stylus in position-determining apparatus wherein said device and an object are movable relative to each other for providing a signal when said stylus engages said object thereby indicating the position thereof, said device comprising:

a fixed member,

a movable member to which said stylus is connectable, said movable member being supportable on said fixed member at a plurality of spaced-apart locations, one of the movable and fixed members having at each of said locations a pair of mutually convergent surfaces and the other one of said members being engageable with said convergent surfaces,

bias means for urging said movable member into contact with said fixed member so that all of said convergent surfaces are engaged thereby positively defining a rest position for said movable member, said movable member being removed from said rest position in opposition to said bias means when a force is applied to said stylus, said bias means and convergent surfaces cooperating, on cessation of said force, to return said movable member to said rest position, and

means for providing said signal when said stylus engages the object.

Claim 3 of Patent 4,270,275

3. A device for mounting a stylus in position-determining apparatus wherein said device and an object are movable relative to each other and the stylus is engageable with said object, said device comprising:

a fixed member,

a movable member to which said stylus is connectable, said movable member being supportable on said fixed member at a plurality of spaced-apart locations, one of the movable and fixed members having at each of said locations a pair of mutually convergent surfaces and the other one of said members being engageable with said convergent surfaces, and

bias means for urging said movable member into contact with said fixed member so that all of said convergent surfaces are engaged thereby positively defining a rest position for said movable member, said movable member being removed from said rest position in opposition to said bias means when a force is applied to said stylus, said bias means and convergent surfaces cooperating, on cessation of said force, to return said movable member to said rest position.