Opinion
Patent Appeal No. 7483.
October 27, 1966.
Virgil E. Woodcock, Philadelphia, Pa., James H. Littlepage, Washington, D.C., for appellant.
Melvin M. Goldenberg, Washington, D.C., Richard E. Hosley, Lynn, Mass. (Frank L. Neuhauser, Washington, D.C., of counsel), for appellees.
Before RICH, Acting Chief Judge, and MARTIN, SMITH, and ALMOND, Judges, and Judge WILLIAM H. KIRKPATRICK.
United States Senior District Judge for the Eastern District of Pennsylvania, designated to participate in place of Chief Judge WORLEY, pursuant to provisions of Section 294(d), Title 28, United States Code.
Phillips appeals from the decision of the Board of Patent Interferences, adhered to on reconsideration, insofar as it awarded priority to Lynch and Larew (hereinafter Lynch) as to counts 18 and 19. Priority was awarded to Phillips as to counts 4, 17, and 20 but Lynch did not appeal and that award stands.
The sole issue before us is Phillips' right to make counts 18 and 19, the board's award of priority resting solely on its decision that Phillips' application lacks support for these two counts. Both parties relied on their filing dates and neither took testimony. The issue turns on the interpretation given to the counts and to Phillips' specification.
It was twice decided by the "Primary Examiner" that Phillips does support counts 18 and 19. These claims first appeared in Lynch and Larew patent No. 2,824,240, issued Feb. 18, 1958, on an application filed Nov. 30, 1953. Phillips copied the claims in an application filed July 2, 1958, serial No. 747,324 for the reissue of his patent No. 2,754,429 issued July 10, 1956, to Leeds and Northrup Company as assignee. The Lynch and Larew patent issued to General Electrict Company as assignee. Initially the Primary Examiner held, on a motion by Lynch to dissolve, that Phillips could not make any of the 22 claims he copied from the Lynch patent. In response to a request for reconsideration, the examiner held that Phillips could make counts 4, 17, 18, 19, and 20. Lynch then requested reconsideration and the examiner replied in an opinion nearly three times as long as the one petitioned from, saying:
While all three decisions on motion are signed by the same "Primary Examiner," appellant's counsel stated at oral argument that they were in fact three decisions by three different examiners. If this be so, two examiners were of the opinion Phillips supports counts 18 and 19, with official approval by a third.
It is still held that Phillips' disclosure fully supports the counts in question without any straining of terms or applying any equivalency of terms.
The Lynch and Larew request has raised many questions relating to the lack of support in Phillips' application.
The above remarks showing how the counts clearly read on Phillips in effect answer these questions * * *.
The case then went to the board for final hearing, on Lynch's request, on an order to show cause against Lynch as to why judgment should not be entered against him on the five counts remaining in the interference.
The invention of the counts is a system for the centralized, automatic control of a plurality of electric generating stations through mechanical, electrical, and electronic control means which receive from the power lines a signal indicating a deviation from the desired norm, called an error signal, which may result from either change in load or change in frequency or both. This error signal is automatically utilized to send out to the several generating stations a modulated carrier signal, known as the control signal, which acts through receiving equipment in the generating stations to operate the controls of the prime mover which powers the generator, such as a steam turbine or diesel engine, so as to correct the deviation from the desired norm. There are controls which may be adjusted by the dispatcher at the central control location so that generators may be selected to pick up the load in such a way as to use first those generators which operate most efficiently. The signals which go from the central control station to the several generating plants are carrier currents transmitted over the power lines themselves, being alternating current carrier frequencies which are different for each station. The carrier currents are modulated by superimposed frequencies which convey the information necessary to effect control of the generating plants, the modulation frequencies being varied in accordance with the error signal received and other control considerations. We quote from appellant's brief as to further aspects of the system:
If all stations are to be loaded equally, then the carrier currents may all be modulated by the same frequency, say 90 cycles per second. However, if one station is to be loaded to a greater or lesser extent than the others, then the modulation frequency will be made greater or less than 90 cycles per second. At each station, the carrier current is demodulated and the control frequency then utilized to change the generation until the system condition has been returned to 60 cycles per second [the desired norm, deviation from which produces the error signal].
* * * When deviation in tie-line load is utilized, a concurrent deviation in frequency may also be used to determine the magnitude of the error signal. Such a system has been termed tie-line load control with frequency-bias.
Sometimes less than the whole of an error signal will be utilized. For this purpose provision will be made for selection of a proportionate amount of the error signal for each of the several generating stations.
Both parties disclose systems of the foregoing kind.
It will suffice to quote count 19 inasmuch as count 18 is exactly like it with omission of that part of clause "(2)" which is emphasized, the number and letter designations, which are alternatives, being those used in the briefs and opinions in this case below. Count 19:
An electric power control system for controlling the generation of a power system having a plurality of generating stations, the control system comprising
(1) means for producing an error signal proportional to the difference between a system condition and a desired condition,
(2) means for selecting a proportionate amount of said error signal (a) and producing a control signal for each generating station whose frequency varies in accordance with the amplitude and polarity of said selected proportionate amount of said error signal,
(3)(b) means for varying the frequency of at least one said control signal to produce a modified control signal for at least one said generating station,
(4)(c) and means for varying the generation of each generating station in response to its respective control signal to tend to bring said system condition to said desired condition.
The only disagreement between the parties, and between the examiners and the board, is with respect to the clause designated (3) or (b). Appellees admit in their brief here, and the board found, that Phillips has express support for every other element of the two counts. The examiner so found and also found support for clause (3)(b) but when the board rendered its decision it appears to have been persuaded to a different view by appellees' arguments as to counts 18 and 19 while agreeing with the examiners on the other three counts.
There is general agreement as to the applicable rule of law as expressed by this court in Martin v. Friendly, 58 F.2d 421, 19 CCPA 1181 (1932):
The counts are broadly drawn and, not being ambiguous, must be given the broadest construction which they will reasonably permit. Limitations, not expressly defined, can not be read into them for the purpose of avoiding the issue of priority. [Cases cited.]
And in Deibel v. Heise, 46 F.2d 570, 18 CCPA 907 (1931):
Had appellant desired to limit his claims to a particular kind or type of kraft paper he might have done so. However, having deliberately elected to claim the invention broadly, he is not in a position to insist that limitations be read into his claims (the counts in issue) for the purpose of avoiding the issue of priority. [Cases cited.]
The board referred to these two quotations. At oral argument appellee's counsel stated that he made no contention counts 18 and 19 are ambiguous.
In the absence of ambiguity, it is neither necessary or proper to look to the specific apparatus described in the Lynch et al. patent in the interpretation of the counts. It appears to us that the board did this in construing the counts. We think it should not have done so but do not find it necessary to discuss this point further.
To enable us to discuss the precise issue before us it is necessary to have a diagram of the various "means" referred to in the counts, wherefore we have created our own, the patent drawings being too complex for our needs. What we are describing is what Phillips discloses since support for the counts depends thereon.
The parallel lines at the left represent a power line fed by multiple generating stations to be controlled, designated as stations M, H, and K. The central control system is represented by a block diagram consisting of various elements receiving information from the power line and feeding instructions back to it for transmission to the various stations. Only the controls for station M are illustrated, those for the other stations being duplicates, but operating on different carrier frequencies. However, the same error signal simultaneously actuates all station controls.
"Load" and "Frequency" represent apparatus for measuring those elements of power line condition, variations from the predetermined normal being fed to an electrical network which produces an error signal in the form of d-c voltage which, in turn, controls a reversible electric motor the shaft rotation of which is indicated by the broken line marked "Drive."
The carrier current of the control signal is produced by an oscillator FM which feeds the control signal to the line for transmission to station M. This carrier might have a frequency of 1000 cycles per second, for example. The control information or intelligence which is carried to the generating station by the carrier is in the form of a modulating frequency imposed on the carrier and generated by the modulating oscillator 66M. The frequency of the modulation (e.g. from 80 to 100 cycles) is what conveys the control information in the control signal and it is determined by a station controller 10M which adjusts the frequency of the modulation oscillator. Associated with controller 10M are three adjustable resistances or "slidewires." Two of them, 54M and 57M, can be manually set by the dispatcher. The third, 50M, is motor-driven from the error signal network. A change in the adjustment of any of them has an effect on the modulation frequency produced by oscillator 66M.
Slidewire 54M controls the percentage participation of station M. For example, the dispatcher can decide that when more power is needed it will be obtained 50% from M, 25% from H, and 25% from K and he sets his 54M, 54H, and 54K controls accordingly. Slidewire 57M is called the "maneuvering point" control, a point predetermined for station M suited to "its connected generating capacity, local load and like factors," according to Phillips' specification. At argument it was explained this control can be used to determine at what percentage of its capacity station M would normally be operated. Slidewire 50M (and corresponding slidewires 50H and 50K, not shown) is not set manually but responds to the error signal, being operated thereby through its motor drive, it being understood that the controls for each of the other stations are simultaneously adjusted.
Each of slidewires 50M, 54M, and 57M acts through a common electro-mechanical control forming part of 10M to adjust a pair of frequency-controlling resistors in the modulation oscillator 66M and, this being so, each of them exercises independent control over the modulating frequency produced by 66M and imposed on the carrier produced by FM to control the station M. Another way to state this is that control is exercised over the station by changing the setting of any one of these slidewires, such control being automatic and continuous in the case of slidewire 50M's automatic response to the error signal. Slidewires 54M and 57M are manually set according to the dispatcher's discretion and are not responsive to the error signal. Once set, they will remain constant during continuous control of control signal modulation frequency by error signal actuation of 50M, until purposely reset.
We are aware of appellee's argument that each of these slidewires does not separately and independently determine the signal frequency and we agree. We are saying that each independently has a control over frequency, the actual frequency produced being, of course, determined by the setting of all three slidewires.
We will now consider the reasons given by the board for its holding that the Phillips disclosure does not support clause (3) (b) of the counts, which reads (our emphasis):
means for varying the frequency of at least one said control signal to produce a modified control signal for at least one said generating station
In its first opinion the board said (our emphasis):
We think that a straightforward reading of the counts in question [18, 19] requires that the means for varying the frequency specified in clause (b) must be in addition to the means of clause (a) for producing the control signal whose frequency varies as specified. We are of the opinion that the slidewires 54M and 57M are a necessary part of the last mentioned means. At least we do not find in the Phillips specification any suggestion that either may be omitted. It therefore follows that neither of these elements can be regarded as properly supporting clause (b). Moreover, we do not believe it to be proper to describe an element of a system by a function which is not determinable at the output of the element. In this case it cannot be determined from an examination of either slidewire 54M or 57M that a variation in their contact position will result in varying the frequency of the later-produced control signal. We, therefore, are of the opinion that the Phillips reissue application does not support counts 18 and 19.
On petition for reconsideration the board said (our emphasis):
Although we are not convinced of any error in our decision, in view of the arguments presented we deem it advisable to restate our conclusion. On the representative channel M slidewires 54M and 57M directly modify the error signal, not the control signal. They affect the latter only as an adjustment of the error signal and their effect on the frequency of the control signal is a remote, not a proximate function or result. We do not believe that such a relationship provides proper support for clause (b) of counts 18 and 19, broad though it may be.
The board's reference to "a remote, not a proximate function" seems to be the same thought as that expressed in its first opinion in the sentence beginning with "Moreover." To attempt to make clear what the board was talking about, from other statements made it would appear that the board's idea was that slidewires 54M and 57M affected only the position of the shaft of a motor (14M) in controller 10M which, in turn, mechanically adjusts the frequency-determining resistors in the modulation oscillator 66M. The board also seems to have felt that the settings of 54M and 57M determined the amount of the error signal which would be permitted to affect the frequency of the control signal, thus modifying the error signal rather than the control signal. This seems contrary to the examiners' view that:
Count 18 finds support in the Phillips reissue application in the same manner as count 4. The limitation as to the modified control signal [clause (3)(b)] is 57M or 54M which when adjusted cause motor 14M to readjust the frequency applied to carrier of Fm, and thereby adjust station Fm [M?] independently of other stations.
In the third examiner's opinion this same view was reaffirmed in the statement:
See note 1. This is said to be by a different examiner.
Count 18 contains a limitation not previously discussed namely "means for varying the frequency — to produce a modified control signal". This limitation is rather broad and is fully answered by slidewire 57M or 54M. Either will modify the signal affecting the adjustment of the modulating frequency, thus produce a modified modulation of the frequency applied to FM. This obviously modifies the signal received at station M, alone to the exclusion of the other signal producers.
It will be noted from what the board said, above, that underlying its conclusion of no support was the proposition that if slidewires 54M and 57M were necessary to meet clause (a) of the counts, then the same elements could not be relied on a second time to provide support for clause (3)(b). Assuming its premise, the board said that the means to support the latter must be "in addition to" the means to support (a).
This seems to have been approached in accordance with the rule against the double inclusion of elements in claim interpretation, a rule which is not of universal application and has its exceptions. See Ellis, "Patent Claims," sec. 17; McCrady, "Patent Office Practice," 4th ed., sec. 82, which says, in part:
Where the elements of a claim are in the form of means plus a statement of function it is permissible that some elements be common to two such means. [Cases cited.]
Appellant has a clearcut answer to the board's argument:
After conceding that either of slidewires 54M or 57M changes the frequency of the modulating oscillator 66 M, the Board reached the erroneous conclusion that in the absence of a disclosure in the Phillips specification that either of these slidewires could be omitted, both are necessary to meet the requirement of clause (2), designated clause (a) by Lynch and Larew. We think the conclusion does not follow. If either slidewire varies the frequency, then surely one of them, as 54M, may be used to satisfy means clause (2) of Count 19, and the other may be utilized to satisfy means clause (3). [Emphasis ours.]
We agree with that view. The way we match Phillips' disclosure to elements (2)(a) and (3)(b) is this: In count 19, the narrower one, slidewire 54M meets the proportioning requirement; clause (a) is met by carrier oscillator FM + modulation oscillator 66M + its operating motor and the necessary elements in its control circuit including slidewire 50M with the means which cause movement of 50M; (3)(b) is met by slidewire 57M, the maneuvering point control, when it is adjusted differently from corresponding controls for other stations, 57H and 57K, so that according to clause (3) of the count one control signal to one generating station is modified. It will be seen that we have not used the same element twice to satisfy different "means" clauses.
In discussing count 17 containing a proportioning limitation, the board held it to be supported by 54M. It said, "We do not understand that Lynch et al. contend otherwise."
We do not consider as valid the board's argument that slidewires 54M and 57M have only a "remote" effect on the control signal, rather than affecting it as a "proximate" function. First, we do not consider this a proper legal distinction. If a "means" produces a function as a direct consequence of its operation, it would not seem to matter that the function is "not determinable at the output of the element," as the board required. No authority contrary to our view has been cited. Furthermore, a "means" is not necessarily " an element." It may include several. Secondly, Phillips' circuitry is such that changing the setting of either 54M or 57M directly affects the control signal frequency. There are three variable resistors to vary the output frequency of modulation oscillator 66M, namely, 50M, 54M, and 57M and only the first is operated in accordance with the error signal. The other two are "means" for varying control signal frequency manually, acting through associated electro-mechanical devices.
Appellees seem to base much of their argument on the undisputed fact that the ultimate control signal frequency is determined by the settings of all three controls but we do not see the force of the argument in view of the fact that each of them separately affects the frequency of the control signal. We also regard as specious their argument, apparently adopted by the board, that the adjustment of a resistance has effect on the error signal rather than on the control signal. In this electro-mechanical system what happens when the dispatcher changes 54M or 57M, or when the error signal automatically changes 50M, is that the control signal frequency is changed and that is the sole purpose served by all three of these controls. While it is true that the direct result of a change in the setting of 54M or 57M by the dispatcher is to vary the extent to which the error signal voltage introduced through automatic changes in 50M is applied to the motor in controller 10M, the ultimate effect of such changes is to act through the motor to change the frequency of the control signal.
For the foregoing reasons we find counts 18 and 19 to be supported by Phillips' application disclosure and the decision of the board as to those counts is reversed.
Reversed.
It is clear from a reading of the respective specifications before us that we are dealing with inventions having different embodiments in electric power control systems. The counts in issue, however, are not drawn to the specific systems disclosed but include several "means" clauses.
Thus, the "invention" defined by the counts is much broader than the specific embodiments disclosed in the respective specifications. It is the "invention" as defined by the counts which must be considered in resolving the issue before us: does appellant's specification disclose the "invention" defined by the count? Both parties have urged us to examine both specifications in resolving the issue. Doing so I find that the result reached by the majority is correct. The invention as defined in the counts, when considered in relation to the respective specifications, warrants the majority in reversing the decision below. I therefore concur in this result.