Technical Development Corp. v. United States

PER CURIAM; This case comes before the court on plaintiffs’ exceptions to the recommended decision (including opinion and findings) filed February 23, 1978, by Trial Judge Francis C. Browne. Having considered the case on the briefs and oral argument of counsel, the court agrees with the trial judge’s opinion (which is set forth infra, with two slight modifications) and adopts the same (as so modified), together with the following paragraphs and the trial judge’s findings, as the basis for its judgment in the case.1 We agree with the trial judge that the Government is licensed under the claims-in-suit of the two involved patents because (a) the crossover circuit invention of the ’809 patent (claims 9-11, 19, 25) was first reduced to practice in the performance of government contract T5202; (b) the maximum fuel limit circuit invention of the ’809 patent (claim 2) was conceived in and during the performance of the ’15202 contract; and (c) the temperature-controlled circuit invention of the ’908 patent was both conceived and first reduced to practice in the performance of the ’15202 contract. At the oral argument plaintiffs stressed, with respect to the first reduction-to-practice of the claims-in-suit of the — 132 — crossover circuit invention of the ’809 patent, a point which the trial judge did not discuss because (as we shall show) it was not properly raised before him. That new point is that (a) certain of the involved claims of the ’809 patent, most particularly claim 19, do not concern an engine or comparable motor at all,2 (b) therefore as to those claims it is irrelevant that the Packard tests (prior to the ’15202 contract) failed to involve true acceleration (which is said to be significant only for engines or comparable motors), and (c) as to those claims first reduction-to-practice occurred in the Packard tests or earlier in bench tests which did not involve operations under actual or adequately simulated service conditions. This belated argument must be rejected. First, we consider it clear that plaintiff did not present this case on that basis to the Trial Division; instead, it argued its case in terms of the operation of an engine. The first paragraph of finding 5 of the first decision in this case, Technical Development Corp. v. United States, 202 Ct. Cl. 237 (1973), cert. denied, 416 U.S. 983 (1974), commences with the flat statement, "The invention described in the ’809 patent relates to electrical circuits for controlling the fuel' impact to gas turbine engines, particularly jet aircraft engines” (emphasis added), and the second paragraph of that finding goes on to describe the invention of the ’809 patent in terms of an engine. 202 Ct. Cl. at 257-58. At the plaintiffs’ specific request, the first paragraph of old finding 5 (with its express reference to "gas turbine engines”) was incorporated (as finding 7) into the findings now before us in this second Technical Development case.3 More than that, in their brief to the trial judge in the present phase of the case (pp. 6-7) plaintiffs commenced their Summary of Argument with this statement: "Plaintiff contends that Claims 19 and 25 of the ’809 patent broadly define apparatus for alternatively selecting one of — 133 — two parameter signal voltages related to engine operation, such as engine shaft speed, gas temperature or acceleration rate, or fuel flow, to control the operation of the engine” (emphasis added). Also, in the portion of plaintiffs’ opening brief to the trial judge on first reduction-to-practice, in connection with defendant’s alleged license under the ’809 patent (pp. 83-87b), as well as in the parallel portion of the reply brief (pp. 11-14, 28), there is no discoverable hint of the contention now made; again, the argument is wholly in terms of an engine or comparable motor. It is no wonder that the trial judge did not treat with the new contention now presented. The case was simply not tried before him on that basis, and there was no reason for him to consider the environment of the ’809 claim as anything other than an engine or comparable motor (involving the factor of acceleration). Conversely, we consider that he properly evaluated the issue of reduction-to-practice of those claims with respect to the environment of such an operating engine. Second, plaintiffs have not convinced us, on this record, that .the ’809 patent covers any device not operated in connection with an engine or comparable motor. The title of the patent is "Electronic Engine Speed Control System” and the specification relates almost entirely to engines. All that plaintiffs point to is a single sentence which says:4 "However, it is to be understood that the invention is not limited to this particular application ['engines of the jet propulsion type’] but is deemed to also cover such modifications and use as come within the scope of the invention as expressed in the appended claims.” But in view of the almost-complete focus of the specification on engines, of the "fundamental” principle that "claims are to be construed in the light of the specifications and both are to be read with a view to ascertaining the invention” United States v. Adams, 383 U.S. 39, 49 (1966), and of the plaintiffs’ presentation below which was geared to engines and motors, we are not persuaded that the ’809 patent covers the type of invention (not at all involving the components of speed, movement, and acceleration) now pressed upon us. — 134 — On the basis of the trial judge’s opinion (as slightly-modified) and his findings (modified in one particular), as well as on the basis of the foregoing additional discussion, the court concludes that plaintiffs are not entitled to recover and that the petition must be dismissed. OPINION OF TRIAL JUDGE BROWNE, Trial Judge: Plaintiffs seek reasonable and entire compensation under 28 U.S.C. § 1498 (1976) for the unauthorized manufacture and use by or for the United States of jet engine fuel control systems described in and covered by two patents1 owned by plaintiff Franklin F. Offner (hereinafter "Offner”). Offner is the inventor to whom both patents were issued. Technical Development Corporation (hereinafter "TDC”) is Offner’s exclusive licensee and, by virtue thereof, a coplaintiff in this suit. The patents relate to electronic fuel control systems suitable for use in jet engines such as those used in aircraft. The ’809 patent is entitled "Electronic Engine Speed Control System.” The ’908 patent is entitled "System for Accelerating Engines to Selected Speeds and Maintaining the Speed Selected.” Plaintiffs contend that the claims presently in suit are infringed by the integrated electronic fuel control system designed for and used on jet engines designated by defendant as the J-47 engine. The accused J-47 engines were manufactured for the United States by the General Electric Company. Defendant, in addition to asserting defenses of nonin-fringement and invalidity of the claims in suit, contends that it is licensed under the claims in suit and, therefore, is not liable to plaintiffs. We hold that defendant is licensed under the claims in suit, and, therefore, is not liable to plaintiffs. I. Introduction This is the second phase of the trial on liability in this case, the first phase having been limited by agreement of — 135 — the parties to the issues of validity and infringement of claims 3, 4 and 5 of the ’809 patent and claims 11 and 35 of the ’908 patent. The first phase of the trial on liability resulted in a dismissal of the petition as to those claims, the court having determined that defendant is fully licensed under each of those claims. The court remanded the case to the Trial Division for further trial proceedings and an opinion on the liability of the defendant on the remaining claims in suit. Technical Development Corp. v. United States, 202 Ct. Cl. 237 (1973), cert. denied, 416 U.S. 983 (1974).2 The claims in the present phase of the trial on liability are claims 2, 9,10,11,19 and 25 of the ’809 patent, and claims 1, 9, 10, 21 and 30 of the ’908 patent. A. The Licensed Claims Claims 3, 4 and 5 of the ’809 patent are drawn to a null-balance speed control circuit which operates in conjunction with a temperature override circuit to control the flow of fuel to a jet engine. Claims 11 and 35 of the ’908 patent are drawn to a more sophisticated stall suppressor circuit. This court in its decision dated June 20, 1973, determined that claims 3, 4 and 5 of the ’809 patent were licensed to the Government because the first actual reduction to practice of the inventions defined thereby occurred in the performance of contract W33-038ac-15202 between the Government and plaintiff, TDC. — 136 — The court also held that claims 11 and 35 of the ’908 patent were licensed to the defendant because the first actual reduction to practice of the inventions defined thereby occurred in the performance of contract W33-038ac-20375 between the Allison Division of the General Motors Corporation (hereinafter Allison) and the United States and purchase order subcontract, H-2076 between Allison and Thompson Products Corporation (hereinafter Thompson) issued pursuant to the terms of contract W33-038ac-20375. Thompson, Offner and Offner Electronics, Inc. were determined to be joint adventurers and as such were bound by the patent rights and obligations of the subcontract to the same extent as Thompson Products. B. The Claims in Suit Counsel for the parties, apparently in the belief that the court would ultimately rule on the issues of validity and infringement of claims 3, 4 and 5 of the ’809 patent and claims 11 and 35 of the ’908 patent, stipulated that: 4. The parties agree that if any of claims 3 to 5 of Reissue Patent No. 24,809 are held to be valid and infringed by the Court (as distinguished from the Commissioner) there will be no further trial proceedings irrespective of how the Court (as distinguished from the Commissioner) rules with respect to claims 11 and 35 of Patent No. 2,697,908. 5. If all of claims 3 to 5 of Reissue Patent No. 24,809 are held to be invalid or non-infringed, then the parties further agree that the trial will be resumed with respect to all claims involved in suit with respect to claims 2, 9, 19 and 25 of Reissue Patent No. 24,809, and claims 1 and 10 of Patent No. 2,697,908. The court, having ruled that defendant was licensed under the claims submitted for adjudication, did not hold the claims to be valid or infringed. Neither did the court hold that the claims were invalid or not infringed. Thus the literal terms of the stipulation did not seem to cover the situation which resulted! The result, however, was that defendant was not liable with respect to those claims and, therefore, the effect of the ruling was the same as if the court had held the claims were not infringed, within the contemplation of the above-quoted paragraph 5 of the stipulation. — 137 — Accordingly, a further trial was expected to be conducted as to all issues relating to claims 2, 9, 19 and 25 of the ’809 patent and claims 1 and 10 of the ’908 patent. Prior to trial, however, the parties agreed to exclude claim 9 of the ’809 patent and claim 10 of the ’908 patent from consideration. II. The Issues The parties have agreed to use claims 2, 19 and 25 of the ’809 patent and claim 1 of the ’908 patent as representative claims as to all issues in the case. Plaintiffs assert validity and infringement of these claims, while defendant presents a variety of defenses, most of which narrow down to invalidity, noninfringement and license. Accounting has been deferred, pending a determination of liability, if any. For the reasons set forth in the following opinion, we hold that the United States is licensed under all of the claims now at issue. Accordingly, the issues of validity and infringement are not reached. Mine Safety Appliances Co. v. United States, 176 Ct. Cl. 777, 364 F.2d 385, 150 USPQ 453 (1966).3 III. Comparison of the Claims of the Patents in Suit The two patents in suit present this court with what seems to be a novel (or at least uncommon) situation. While only two patents are involved in this liability litigation, at the completion of the trials this court will have effectively determined the liability issue with respect to five related but specifically different fuel control circuits. As will become evident, each of the five circuits, as claimed by the respective patents, have separate dates of conception and reduction to practice. The court in the initial trial held that the Government is licensed as to two of these circuits, and, — 138 — therefore, these circuits will not be further discussed. The three other circuits and their respective claims remain to be adjudicated. Both the ’809 patent and the ’908 patent relate, in general, to electronic circuits for controlling the flow of fuel into the combustion chambers of gas turbine engines, particularly jet aircraft engines. A jet aircraft engine basically comprises a cylindrical housing, open at both ends, in which is located: (a) a series of air compressors toward the forward end of the housing, (b) a combustion chamber aft of the compressors, and (c) a gas turbine aft of the combustion chamber. The turbine and compressors rotate about a common axis. In operation, air enters the engine through the open front end of the housing and is compressed through a series of compression stages by the air compressors. The compressed air is then fed into the combustion chamber wherein fuel is injected and, when ignited, is burned. The hot, compressed combusion gases thus produced expand rearwardly through the turbine, which drives the compressors, and out the open rear end of the engine, thereby creating the desired forward thrust of the engine. Introduction of too much fuel into the combustion chamber at a given speed or load will cause the engine to develop an overtemperature condition which can damage the turbine and tailpipe, a result to be avoided. A useful fuel control system for a jet aircraft engine must be capable of maintaining the engine speed at a selected number of revolutions per minute (rpm), and yet be capable of performing quick and efficient accelerations whenever a greater engine speed (higher rpm) is selected. The control must provide sufficient fuel in a sufficient time period to accomplish these speed selection and acceleration goals without allowing the engine either to stall or to exceed its maximum operative temperature. When Offner turned his attention from designing and manufacturing electronic measuring instruments to the field of gas turbine fuel controls in 1943, most aircraft fuel control systems were hydromechanical in nature. Offner learned from a friend in the Navy that there was a need for a suitable fuel flow control system for aircraft jet engines then being developed by the Navy. Through research of his — 139 — own and through further knowledge gained during performance of Government contracts (including experimentation on Government-owned jet engines), Offner invented the various embodiments of the electronic jet engine fuel control systems described and claimed in the two patents in suit. The three interrelated electronic control circuits remaining at issue in this case are: (1) a crossover circuit which selects either a speed control signal or an acceleration signal (claims 9, 10, 11, 19 and 25 of the ’809 patent); (2) a maximum fuel limit circuit (claim 2 of the ’809 patent); and (3) a temperature-controlled acceleration circuit (claims 1, 9, 10, 21 and 30 of the ’908 patent). A general description of these circuits follows. A. The Crossover Circuit The ’809 crossover circuit selects either of two signals: (1) a speed control signal, or (2) an acceleration signal. The selected signal passes through the control system so that the chosen circuit regulates the flow of fuel to the engine. The function of the speed control circuit (1) is to adjust the opening of the engine’s fuel valve in response to the selected aircraft speed as determined by the aircraft’s throttle position. The function of the acceleration signal (2) is to accelerate the engine from its initial speed to a higher selected speed as quickly as possible without causing the engine to reach overtemperatures or perform inefficiently. Generally, as engine speed (rpm) increases, more air flows through the engine. The increased air flow tends to cool the turbine. Therefore, as the speed increases the engine can accept more fuel without overheating. The acceleration control circuit in the ’809 patent produces a rate acceleration signal which increases exponentially with the increase in the engine speed. If a control were designed to monitor only the speed control signal, a throttle movement calling for a quick and large acceleration would likely cause an overtemperature. The speed signal would, under such conditions, call for maximum fuel flow. This maximum flow, particularly at low speeds would cause the engine to overheat. Therefore, the Offner crossover circuit solves the potential overheating problem by selecting the acceleration control signal — 140 — when high acceleration rates are called for since this signal is designed to schedule fuel flow according to the engine temperature characteristics. When the engine speed approaches the selected speed for which the throttle is positioned, the crossover circuit operates to switch the control system back to the speed control circuit. When the engine speed equals the intended value, the control operates to keep the fuel flow constant and the engine remains in a steady state at that desired speed. The selection of the speed signal or the acceleration signal is accomplished through the use of two diodes and a variable control voltage.4 The findings of fact accompanying this decision describe the operation of the electronic speed control, acceleration control and crossover circuits in more detail. For a general understanding of the system, description of a theoretical and somewhat oversimplified acceleration of an engine using the control should suffice. For the purposes of this example, it will be assumed that the crossover control circuit is designed so that the control crosses into the acceleration circuit when a speed change is required of greater than 20 percent of the initial speed, and crosses back to the speed control circuit when the actual speed reaches about 80 percent of the selected or desired speed.5 If the throttle of the engine is pushed forward (toward the "open” position), calling for an increase in speed less than the minimum 20 percent, only the speed control signal will control. If, however, the throttle is quickly pushed further forward, thereby calling for an increase in speed greater than the 20 percent minimum, the crossover circuit will — 141 — block the speed signal and pass only the acceleration signal until the speed attains 80 percent of the selected speed. A second crossover will then occur, causing the acceleration signal to be blocked and allowing only the speed signal to pass to control the fuel flow. B. Maximum Fuel Limit Circuit In order to further sophisticate his control system, Offner designed and invented a maximum fuel limit (mfl) circuit. Offner found that at certain engine speeds, the fuel valve should only be opened to a limited degree. Any opening beyond that limited degree, at a given speed, resulted in an overtemperature of the engine. Claim 2 of the ’809 patent claims this aspect of the Offner invention, and the findings of fact accompanying this decision describe the operation of this circuit in greater detail. Generally, however, the mfl circuit compares a voltage representation of the engine speed with a voltage representation of the fuel valve opening. Whenever the fuel valve opening is considered too large for a given speed (and would therefore tend to cause an engine overtemperature), a diode passes an mfl signal which serves to decrease the fuel flow. Generally, if the opening is not considered too large, the mfl signal will not be passed. Whenever the mfl signal is passed, it joins either the speed signal or the acceleration signal (depending upon the signal then being selected by the previously described crossover circuit). The signals thus joined combine to control fuel flow, but the mfl signal is designed so that it will, in effect, override any speed or acceleration signal calling for increased fuel. Thus, while the two signals are combined, the mfl controls and fuel flow will decrease. Once the fuel flow decreases to the point that the fuel valve opening is maximum for the existent speed, the mfl signal is blocked and the speed or acceleration signal will again control. C. The ’908 Temperature-Controlled Acceleration Circuit The third circuit presently before this court is the temperature-controlled acceleration circuit as described and embodied in claims 1, 9, 10, 21, 28 and 30 of the ’908 patent. The invention described in the ’908 patent is an — 142 — improvement on the control system described in the ’809 patent. The ’908 patent describes a crossover circuit which selects either: (1) a speed control signal, or (2) an acceleration signal, in much the same manner as the ’809 patent. The principal difference between the two crossover circuits is that the acceleration control in the ’908 patent is maintained through the utilization of the exhaust gas temperature as a parameter. In his studies, Offner found that the use of the exhaust gas temperature as a parameter rather than a speed parameter, such as described in his ’809 patent, offered a more sophisticated acceleration control. The operation of the crossover between the speed signal and the temperature-controlled acceleration signal is, for description purposes, the same as that already described in reference to the ’809 crossover circuit. IV. Background of the Invention in Suit The inventions defined by the patented claims in suit find their genesis in an electronic fuel control system developed by Offner over the 4-year period 1944 to 1948. In 1943, Offner first became aware of the potential need for fuel controls for military gas turbine engines. An officer in the U.S. Navy advised Offner that the Navy was developing a gas turbine engine for use in jet aircraft and suggested that Offner might be able to contribute to development of a fuel control system for the engine. Utilizing his own time and funds, as well as those of his small electronics company, Offner, in 1944, built several models of an electronic fuel system incorporating a speed control circuit and temperature override control circuit for a turbine engine. In early 1945, Offner visited the Power Plant Laboratory at Wright Field, Dayton, Ohio, with a view to soliciting governmental interest in the concepts of his control system. He succeeded, and between March and June 1945, Offner developed and built a model control unit for preliminary tests on an aircraft jet engine. In June 1945, Offner took the model to Wright Field and ran tests on an aircraft jet engine bench installation. The test unit included the speed control circuit and a temperature — 143 — override circuit. The tests showed that the control unit, although still in the development stage, had potential for further bench and flight testing. Following the test at Wright Field in June 1945, Offner continued to do further work on the jet engine fuel control system. During the remainder of 1945, Offner made improvements in his system including, among other things, incorporation of the principles of signal selection or "crossover” in the control circuit, using the acceleration rate as the parameter for one control signal and speed as the parameter for the other. During this time period, Air Force personnel suggested to Offner that he contact Packard Motor Car Company who had been working on the development of aircraft jet engines for the Air Force. Packard in early 1944 had entered into a contract with the Government for engineering and development work in connection with Packard’s manufacture and test of the Rolls-Royce V-1650-9 aircraft piston engine for the Government. In January of 1945, the Government authorized Packard to proceed with a change in the contractual scope of work so as to include preliminary design studies of a turbojet engine for pilotless expendable aircraft. In July of that year, Offner submitted a quotation to Packard for Offner electronic fuel controls. On August 9, 1945, Packard issued a purchase order to Offner Electronics, Inc. for the delivery of six electronic speed control units with experimental direct current tachometers. Negotiations and work continued, and in December of 1945, Offner visited Packard and delivered his control model 630X3 to Packard for its use. The control contained Offner’s null-balance speed control circuit and acceleration rate control circuit with crossover signal selection. Packard’s personnel performed simulated laboratory bench tests with the control. The results of the tests were promising enough to warrant bench testing of the control on an 1-40 aircraft jet engine. The control was engine-tested on that model between December 12, 1945 and January 7, 1946. The Offner control was not tested on an XJ-41 engine because that engine was in an early stage of development and was not then considered reliable. While working on the Packard purchase order, Offner kept in constant communication with the Government — 144 — concerning further development of his control. On June 20, 1946, Offner Electronics, Inc. and the United States entered into contract W33-038ac-15202 (hereinafter the T5202 contract) for developing and building an electronic fuel control for the J-33 turbojet aircraft engine, the control to be suitable for engine and flight tests. The contract was in reality a development contract since it set out only performance specifications and called for the contractor to provide a final report covering the details of design, development and testing. The contract included a "patent rights” clause by which the contractor, Offner Electronics, Inc., agreed to grant to the United States a license to any and all inventions "conceived or first actually reduced to practice * * * in the performance of the contract.” Between June and December 1946, Offner and Offner Electronics, Inc., proceeded to develop and build an electronic fuel control pursuant to the contract. Sometime before November 1946, Thompson Products, Inc., of Cleveland, Ohio, became interested in the project and supplied Offner with some of the mechanical elements to be used as parts of the fuel control system. Offner also visited the Allison Division of General Motors Corporation, Indianapolis, Indiana, to discuss arrangements for testing the electronic control on a J-33 gas turbine engine. Allison was then working with the Government to develop the J-33 engine for military aircraft. The control unit (except for the fuel valve) developed in performance of the ’15202 contract was actually completed and bench tested by Offner in September 1946. By December 1946, Offner sent the completed fuel control to Allison, where it was bench tested on a J-33 jet engine between January 13-17,1947. The particular control tested consisted of (1) a speed control circuit capable of producing an error signal voltage whose magnitude and polarity were determined by the difference between actual and desired engine speed; (2) an acceleration rate circuit capable of producing an acceleration rate signal increasing with engine speed; and (3) a crossover circuit capable of selectively switching from the speed signal to the acceleration signal or vice versa. The control included additional improvements such as integrated feedback circuits to — 145 — increase system stability. The tests indicated that the control operated the engine successfully during severe accelerations. On February 17, 1947, Offner sent a specification and drawing to his patent attorney in order to assist him in the preparation of a patent application. That application included both the crossover and the mfl circuits and was the basis for the ’372 patent (the patent from which the ’809 reissue patent subsequently resulted). After the January tests at Allison and as a result of the test on January 16, 1947, Offner made a tentative conclusion that since temperature appears to be sensitive to small changes in acceleration, direct temperature control over acceleration may be preferable to limiting temperature by the acceleration rate control circuit tested at Allison. Offner subsequently developed a temperature-controlled acceleration circuit. After the January tests at Allison, by agreement of the parties concerned, the Offner control was adapted for use with the J-35 jet engine (an axial flow compressor as compared to the centrifugal flow compressor of the J-33 engine). In May 1947, Offner returned to Wright Field and brought with him the 630X3 control, used at Allison, and also brought a separate piece of equipment that included the temperature-controlled acceleration circuit he had devised. Offner disconnected the original acceleration control circuit and wired the temperature-controlled acceleration circuit to the 630X3 control. The unit was bench tested in this form at Wright Field in May 1947. The testing of the Offner control between May 7-9, 1947, at Wright Field on the J-35 engine covered a period of 22 hours. On the morning of May 7, 1947, the control did not satisfactorily operate since nine overtemperatures were recorded during 40 engine accelerations. However, by the conclusion of the testing on May 9,1947, a final setting was achieved in which the accelerations were successfully made. The May test included rigorous snap throttle accelerations. After the May 1947 tests, Offner no longer actively participated in contract ’15202 and the contract was formally terminated on July 10, 1947. The following excerpt from the trial commissioner’s opinion, as adopted by the court in Technical Development — 146 — Corp. v. United States, 202 Ct. Cl. 237, 250-51 (1973), cert. denied, 416 U.S. 983 (1974), is the law of the case and is hereby adopted as a part of the present opinion of the present trial judge. Sometime after mid-1946 and while Offner was developing the fuel control under Contract W33-038 ac 15202, Offner and Thompson Products, Inc. decided to collaborate in their efforts to pursue further development of the Offner control. On March 5, 1947, Offner Electronics Inc. and Thompson Products entered into a contract by which Thompson Products got an exclusive license under Offner’s fuel control inventions (and any patents which might be granted thereto); and Thompson Products agreed to manufacture fuel control systems using electronic components developed and built by Offner Electronics Inc. The control units were thereafter referred to as the "Thompson-Offner fuel system.” On March 21, 1947, a meeting was held at the Power Plant Laboratory, Wright Field, to discuss a program for testing fuel control systems on jet engines then under development by Allison for the Government. Offner attended the meeting along with representatives of Allison, Thompson Products and the Government. The upshot of the meeting was that Allison was to be responsible for procuring and testing fuel controls for its jet engines, one such control to be the Thompson-Offner fuel system. Several days later, Thompson Products submitted a proposal to Allison to furnish the Thompson-Offner fuel system for testing. Allison in turn submitted a proposal to the Government for "* * * Procurement and Testing of Five (5) Thompson-Offner Turbo-Jet Engine Controls.” The proposal noted that "* * * it should be recognized that this is a joint research and development program on new type controls involving the Government, control manufacturers, and this contractor; which no doubt will incur numerous changes in design which we will have to negotiate from time to time.” After negotiations extending throughout the balance of the year 1947 and early 1948, Allison and the United States, on March 15, 1948, entered into Contract W33-038 ac 20375 for "Electronic Power Controls, Tests, Reports and Data.” The contract provided for procuring, testing (both bench tests and flight tests), and evaluating the Thompson-Offner fuel system, as well as a competing system then under development, known as the Eclipse-Pioneer control. The Government was to furnish the jet engines for the tests. The contract contained a Patent Rights — 147 — clause similar to the one in the earlier Offner-United States contract. Then, on May 11, 1948, Allison entered into a subcontract with Thompson Products, pursuant to the terms and conditions of the principal contract, including the Patent Rights clause. Neither Offner nor Offner Electronics Inc. was a signatory to the Allison-Thompson Products subcontract. However, Offner received a copy of the subcontract, and he continued to collaborate with Thompson Products in all phases of development, testing and reporting required by the subcontract’s terms. Furthermore, Thompson Products and Offner agreed to split the proceeds payable under the subcontract ($48,900) on the basis of 55% to Thompson Products and 45% to Offner Electronics Inc. Offner Electronics Inc. in fact received $21,305 as its share of the subcontract price. The Thompson-Offner controls were flight tested at Muroc Air Force Base, Muroc, California between May and September 30, 1948. Personnel of Thompson Products and Offner Electronics, Inc., went to California to supervise and oversee installation and flight testing of the fuel control. The tests were conducted with Air Force planes and with the assistance of Government personnel and material. These tests were successful, and the control utilized contained a speed control circuit, a temperature-controlled acceleration circuit, and a crossover circuit. V. License Issues A. Introduction Against this factual backdrop, defendant contends that the United States is licensed under the inventions as claimed in both patents in suit. Defendant contends that: (1) the tests at Packard from December 12, 1945 through January 7, 1946, did not constitute a reduction to practice of the crossover circuit and that, even if it did, the United States would be licensed either under the terms of the purchase order between Packard and Offner or under the theory of implied license; (2) the crossover circuit was first actually reduced to practice during the tests in January and May 1947 under the T5202 contract; (3) the mfl circuit was conceived in the performance of the ’15202 contract; (4) the temperature-controlled acceleration circuit was con — 148 — ceived in the performance of the ’15202 contract and first reduced to practice during the bench tests in May 1947 which were part of the ’15202 contract; or (5) in the alternative, the temperature-controlled acceleration circuit was first reduced to practice in the performance of the ’20375 contract (Offner being bound by the patent rights clause of the contract by its joint venture relationship with Thompson or under an implied license theory). Plaintiffs deny each of defendant’s assertions and argue that: (1) the tests at Packard prior to January 7, 1946 did constitute the first reduction to practice of the crossover circuit; (2) Offner was not bound by any explicit patent rights clause or by any implied license theory; (3) the mfl circuit was first conceived prior to the inception of the ’15202 contract; (4) the temperature-controlled acceleration circuit and the May 1947 tests were not conceived or conducted "in the performance” of the ’15202 contract; and (5) the Government is not entitled to a license based on the reduction to practice of the temperature-controlled acceleration circuit during the ’20375 contract since Offner is not bound to grant the United States a license either under a joint venture theory or an implied license theory. We conclude that each of the three inventions embodied by the aforementioned representative claims were either conceived and/or first reduced to practice in the performance of the ’15202 contract and therefore we need not reach the thicket of more theoretical contract interpretations and implied license theories to which defendant would subject the Packard or Allison contracts. Accordingly, we hold that defendant is licensed under all claims in suit in the ’809 and ’908 patents. As to defendant’s alternative claim of license for the ’908 patent claims under the ’20375 contract, we will not discuss that matter at length since it is unnecessary to a resolution of the case, but feel compelled to state that the court’s first opinion and decision in this case adequately applied defendant’s theory to the facts in this case. While findings of fact detail our considerations, brief analysis of the law and the facts leading to our conclusion follow. B. Legal Theory The license defense is based on a contractual theory. Often in a Government contract, the supplier of services or — 149 — goods agrees that any inventions conceived of or first actually reduced to practice in the performance of the contract may be used by the Government on a royalty-free, nonexclusive basis. When the subject matter is clearly within the reasonable scope of the services or goods called for by the contract, the patentee-supplier is clearly obligated to grant the Government the agreed royalty-free, non-exclusive license. Obviously, cases in which the parties are bound by clear legal (contractual) relationships and in which the subject matter of the invention is clearly within the substantive terms of the contract are seldom litigated. On the other hand, the litigated cases arise out of situations in which the Government and the patentee have entered into several contracts of different scope, the parties have disagreed as to the scope of the performance called for in one or more contracts, or the patentee, although a beneficiary of the contract, was not the prime contractor nor a direct party to the contract at issue. We are now confronted with such a complex situation. If the patentee is bound by a contract under which the Government is expressly granted a royalty-free license for any inventions "made in the performance” of the contract, the sole remaining issue is whether the patented invention in question was so made. This court has construed the general phrase "in the performance of’ liberally, reasoning: Inventions made under a Government contract are the product of expenditures from the public treasury in the course of a governmental function; the public, having in a sense ordered and paid for the invention through its representatives, should not again be taxed for its use, nor excluded from its use, nor permitted to use it upon restrictive conditions advantageous to no one but the patent owner. Mine Safety Appliances Co. v. United States, 176 Ct. Cl. 777, 789, 364 F.2d 385, 392 (1966) (quoting Investigation of Government Patent Practices and Policies, Report and Recommendations of the Attorney General to the President, Vol. I, pp. 88-89 (1947)). Under such a liberal construction, it is enough that a significant feature of the invention was, itself, within the contractual scope, or resulted directly from the course of the contract performance. Id. at 787-88, 364 F.2d at 391. — 150 — The Government has the right to use, royalty-free, those inventions which have a "close and umbilical relationship” to the work and research funded by the United States and were crystallized during performance of the federal contract. Technitrol, Inc. v. United States, 194 Ct. Cl. 596, 613, 440 F.2d 1362, 1372, 169 USPQ 732 (1971). If the invention is so tied to the work to be done under the contract as to contribute significantly to the results anticipated by that agreement, the Government is entitled to a license. Mine Safety Appliances Co. v. United States, 176 Ct. Cl. 777, 787, 364 F.2d 385, 391 (1966). When determining whether or not an invention was conceived or first reduced to practice within the performance of a contract, this court does not focus only upon the final physical product or device which the Government receives. In research and development contracts, the Government is also paying for advancement in technologies and is entitled to crystallized ideas, improvements and inventions emerging from the process of ongoing study, inquiry and creation. Technitrol, Inc. v. United States, 194 Ct. Cl. 596, 615, 440 F.2d 1362, 1373 (1971). If the Government and the patentee entered into several continuing research and development contracts and the invention was not separate from or independent of the ongoing program, the Government is entitled to a license. The liberal construction of the contractual phrase "in the performance” is not without limit. That phrase precludes the licensing of postcontract developments which are new and not obvious. For example, there is no license when a nonobvious invention is conceived after termination of the inventor’s federal connection even though that separate invention may have a close connection with the preceding work. Id. at 618, 440 F.2d at 1374-75. In order to establish the close and umbilical connection, required by Technitrol, the defendant must do more than broadly allude to various connections or contracts between an invention and a Government contract. It must supply hard facts to support its position. Cf. Lockheed Aircraft Corp. v. United States, 202 Ct. Cl. 787, 485 F.2d 584 (1973). Where the facts surrounding the development of the invention, the contract language, and the parties’ contemporaneous construction of the contract lead to the conclusion that the — 151 — invention was not within the performance of the contract, no license accrues. See Rel-Reeves, Inc. v. United States, 209 Ct. Cl. 595, 534 F.2d 274 (1976). Finally, where a patentee is not a direct party to a contract containing a license clause, the Government may nevertheless be licensed under an "implied license” theory. Such a theory should not be applied broadly to trap an unwary inventor, but its application is appropriate in limited situations. For example, where a patentee has been intimately involved in all phases of performance of a Government-funded contract during the performance of which the invention was conceived and/or first actually reduced to practice, and the work was, in fact, paid for in substantial part or entirely out of Government funds, the Government is entitled to a license. Technical Development Corp. v. United States, 202 Ct. Cl. 237, 255 (1973), cert. denied, 416 U.S. 983 (1974). See Ordnance Engineering Corp. v. United States, 68 Ct. Cl. 301 (1929), cert. denied, 302 U.S. 708 (1937). C. Burden of Proof The defendant bears the burden of proof on the license defense and must establish by a preponderance of the evidence that a conception or a first actual reduction to practice occurred in the performance of a Government contract. Cf. Mine Safety Appliances Co. v. United States, 176 Ct. Cl. 777, 364 F.2d 385, 150 USPQ 453 (1966). This court, in its first opinion in this case, defined conception as follows: An invention is conceived, within the meaning of the patent law, when the inventor has a "definite idea of a complete and operative invention as it is thereafter to be reduced to practice. * * * Technical Development Corp. v. United States, 202 Ct. Cl. at 308, finding No. 53. This court, in the same opinion, defined actual reduction to practice as follows: An invention is actually reduced to practice when it is put into physical form and shown to be operative in the environment of its practical contemplated use. * * * Laboratory tests, rather than tests under actual use or service conditions, may be sufficient to constitute actual — 152 — reduction to practice if the conditions of the test adequately simulate the conditions of practical use. * * * Technical Development Corp. v. United States, 202 Ct. Cl. at 308, finding No. 54. If the defendant proves that the invention was reduced to practice in the performance of such a Government contract, then the burden shifts to the plaintiff to prove that a first reduction to practice of the invention occurred prior to the award of the contract. The proof of conception and/or reduction to practice is a heavy one for either party and requires more than self-serving testimony or uncorroborated records and documents. If the defendant succeeds in proving by a preponderance of the evidence that conception has occurred in the performance of a Government contract, then the Government is entitled to a license. Plaintiff, with the single exception of the crossover circuit, has failed to prove a conception of any of the three inventions before the date in which Offner entered into the 15202 contract. Plaintiff has established a conception date of July 28, 1945 for his crossover circuit, as evidenced by a witnessed entry of his notebook diagrams which explain a crossover circuit. Plaintiff has presented other documents, notebook entries, and testimony to prove conception of the temperature-controlled acceleration circuit and the mfl circuit, but this evidence fails to establish conceptions of the respective circuits, as defined by the quoted law of this case. Most of the testimony of Offner on this point is self-serving and uncorroborated. None of the documents produced show or explain a definite idea of the complete and operative invention as required by the definition of conception as established by the first decision in this case. (The findings of fact accompanying this opinion set forth the facts relating to conception in detail.) Similarly, plaintiff has failed to convince the court that any of its laboratory tests of the control on engine simulators constituted a reduction to practice of any of the patented circuits now in issue. It is important to note that the "Offner electronic controls” were the first electronic controls used on jet engines. There was little contemporaneous knowledge or reliable data to establish whether or not such controls — 153 — could successfully control a jet engine, the engines themselves being in the early stages of development. These simulators, as found in the earlier trial, did not fully simulate the characteristics and environment of a full scale, operative aircraft engine in its practical, contemplated setting. The record indicates that during the laboratory tests conducted with Government-owned engines and facilities, Offner did not have sufficiently complete knowledge or data to have devised complete models even if he had desired to do so. During the bench tests of the control on actual engines in the Government facilities, the controls suffered some design and technical failures. Both during and after these tests, Offner and other personnel frequently changed the component values within a given system and redesigned the control circuits and mode of operation to function within those values. These facts support the conclusion reached herein that the simulated laboratory tests did not constitute an actual reduction to practice of the circuits in issue. Plaintiffs have argued that the tests at Packard in January of 1946 establish the first actual reduction to practice of the crossover circuit. The record establishes that the control used at the Packard tests on the 1-40 engine did include a crossover circuit capable of switching between a speed control and an acceleration rate control circuit, but plaintiffs have failed to prove by a preponderance of the evidence6 that the crossover circuit actually operated or was proven to be operative in the environment of its practical contemplated use in an aircraft jet engine. The operation of the crossover circuit, as described earlier, is such that the speed of the engine can be increased at a certain rate without ever causing the control to cross over from the speed control mode to the acceleration mode. A quick movement of the throttle calling for an incrementally large increase in speed would cause the control to cross from the speed control mode into the acceleration control mode. However, a slow enough throttle movement — 154 — would not cause this crossover. The evidence introduced at trial does not establish whether or not the control ever performed the crossover function during the Packard tests. Two witnesses of the test, after an interval of some 30 years since the test, offered contradictory testimony but each line of testimony was supported by equally persuasive reasoning. One testified that the throttle movement was slow in order to avoid the acceleration mode, while the other testified that the movement was quick in order to determine if the acceleration mode worked. The data concerning the tests results is equally inconclusive. The engine did not reach an overtemperature during the tests, but the parties agree that at a relatively slow throttle movement, no overtemperature would occur even if the control stayed in the speed mode. Other factors which might have led to an ultimate conclusion that the crossover function actually occurred (such as the normal speed response of the engine, the degree of damping found in the speed circuit, and the speed of the servo motor which controlled the opening and closing of the fuel valve) are lacking in the record. In short, the record does not establish that the crossover did or did not occur. The party upon whom the burden rests must suffer the loss. In view of the above, it is concluded that at the time that Offner Electronics entered into the T5202 contract on June 20, 1946, Offner had conceived of the crossover concept but had not conceived of the mil or the temperature-controlled acceleration crossover circuit. At that point in time none of the inventions had been reduced to practice in the environment of any practical contemplated use. D. Terms of Contract Performance The T5202 contract was a development contract calling for one J-33 fuel control unit at a price of $7,500. The description of the article included: CONTROL, power, automatic, for J-33 turbo-jet engines, suitable for engine and flight tests, and in accordance with the requirements of Exhibit "A” attached hereto. Exhibit "A” noted tentative specifications for the control. It did not specify the circuitry to be used or the means by — 155 — which the control was to perform within the specifications. The contract, in effect, called for the development of a control which would achieve the desired results. The ’15202 contract included a patent rights clause which granted to the United States any invention, improvement and discovery conceived or first actually reduced to practice in the performance of the contract. We hold that the crossover circuit was first reduced to practice, the mfl circuit was conceived, and the temperature-controlled acceleration circuit was both conceived and reduced to practice in the performance of the ’15202 contract. Perhaps the most controversial issue which had to be decided in reaching the above result is the scope of the ’15202 contract both as to duration and subject matter. Plaintiffs assert that the ’15202 contract covered only the single "J-33 control” which it ultimately handed over to the Government. That control included the crossover circuit between speed control and acceleration control, but did not include either the mfl circuit or the temperature-controlled circuit. Plaintiffs further contended that the contract was completed before the May 1947 tests took place. We disagree with both of these contentions. The ’15202 contract was a development contract calling for an electronic fuel control. In research and development contracts, the Government normally is contracting for more than the final product. It is also contracting for design, development, improvements and advancements in technologies. Thus it is entitled to at least a "shop right” in or license to use the "crystallized ideas, improvements and inventions emerging from the process of ongoing study, inquiry and creation.” Technitrol, Inc. v. United States, 194 Ct. Cl. 596, 615, 440 F.2d 1362, 1373 (1971). As the findings of fact show in more detail, the conception of the mfl circuit and the temperature-controlled acceleration circuits took place during the period that Offner was working under the ’15202 contract. The United States is entitled to the use of those ideas for which it not only paid but also assisted in the making. Thus, the United States is licensed to use the inventions defined by claims 2, 19 and 25 of the ’809 patent and claim 1 of the ’908 patent. Even if the contract was deemed to be completed before the May tests, we find that the mfl and temperature156 controlled acceleration circuits were conceived before that time. Furthermore, even if not fully conceived, these two inventions were not separate and independent circuits developed from independent knowledge and research, rather they were tied to the work to be done under the T5202 contract and the knowledge gained from that contract. Mine Safety Appliances Co. v. United States, 176 Ct. Cl. 777, 787, 364 F.2d 385, 391 (1966). The fact that the May 1947 tests were conducted on a J-35 engine instead of a J-33 engine, as originally called for, does not, in our view, establish that the contract was completed before these tests. Viewing the contract as a whole and weighing contemporaneous thoughts of the parties involved, we find that the contract was not completed, as to Offner Electronic’s performance until after the May tests. The contract was not formally terminated until July 10, 1947. The major goal of the contract was a working fuel control, proven by engine testing. Major control components were not changed from use in one engine to the other, and it is apparent from the patent specification and the nature of the invention that the temperature-controlled acceleration circuit, in general, could operate on either engine. We admit that, in all likelihood, the temperature-controlled acceleration circuit would have to be fine tuned to the particular engine it was to control, but the general concepts and basic circuitry would nevertheless remain the same. The parties mutually agreed to change to the J-35 engine. The Government representatives, throughout their paperwork, treated the two tests as being the same contract, and there is no evidence that plaintiff considered the May tests to be separate or that separate pay was requested or received for the May tests. The evidence as a whole supports the defendant’s theory that the May 1947 tests were made in the performance of the ’15202 contract. In the event that actual flight tests would be required to actually reduce any of the inventions to practice (a theory to which we do not subscribe), the Government would be licensed as to the crossover circuit and the temperature-controlled circuit due to the flight tests performed at Muroc Air Force Base under the ’20375 contract. The findings detail the basis upon which this result is reached. — 157 — VI. Conclusion Considering both the January 1947 and May 1947 tests to be in the performance of the contract, we hold, as detailed in the findings, that the crossover circuit was first reduced to practice in the January 1947 test and the temperature-controlled circuits were first reduced to practice during the May 1947 tests. Also, the maximum fuel limit circuit was conceived in and during the performance of the ’15202 contract. The United States is therefore licensed under each of these inventions, as claimed in claims 2, 19 and 25 of the ’809 patent and claim 1 of the ’908 patent. Plaintiffs’ petition is dismissed. CONCLUSION OF LAW Upon the trial judge’s findings and opinion, both of which are adopted by the court (with a few minor modifications), and upon the court’s per curiam opinion, the court concludes as a matter of law that plaintiffs are not entitled to recover and the petition is dismissed. Footnotes [1] The court adopts the trial judge’s findings (with a single change in finding 164) but those findings are not printed herein because of their length. An order of this date indicates the change made in finding 164. [2] Claim 19 is said not even to involve a rotating member. [3] In its first decision in Technical Development Corporation, the court found no pre-1947 reduction-to-practice of the ’809 claims then in suit because those "tests” were not run on a gas turbine engine and did not simulate the environmental conditions of gas turbine operation (or of any other practical device). See 202 Ct. Cl. at 253, 308-09. It is plain that, in the first phase of this case, the ’809 patent was considered as wholly concerned, for the purposes of this case, with a gas turbine engine or comparable practical device. [4] After stating: "The invention is well suited for speed control of rotating members such as engines and other types of power plants and is herein illustrated and described in its application to engines of the jet propulsion type.” [1] U.S. Patent Reissue No. 24,809 (hereinafter the ’809 patent) reissued in 1960, originally issued as U.S. Patent No. 2,662,372 (hereinafter the ’372 patéht) in 1953 and U.S. Patent No. 2,697,908 (hereinafter the '908 patent) issued in 1954. [2] The first decision of this court concerning claims 3-5 of the ’809 patent and claims 11 and 35 of the ’908 patent in Technical Development Corp. v. United States, 202 Ct. Cl. 237 (1973), cert. denied, 416 U.S. 983 (1974), provides a thorough background discussion of the inception of the Offner fuel control and the conception and reduction to practice of some of the concepts to which the patents are directed. Many of the findings of fact made by the trial judge and approved by the court, as well as many of the conclusions of law, are relevant to and constitute the law of the case in this, the second phase of the trial on liability. The plaintiffs attack some of the previous findings as being clearly erroneous and collaterally questions the propriety of applying some of the court’s previous legal conclusions to the issues now before the court. The trial judge not only disagrees with plaintiffs’ assertion that certain findings are clearly erroneous, but also, in fact, concludes that the findings are fully supported by the record. Even if the trial judge disagreed with any of the findings, he had no authority for retrying those facts, de novo. To do so would have the effect of overruling the court’s approval of the previous trial judge’s findings (which were necessary for the court to reach its conclusions in the earlier phase of the trial of this case). Thus, the previous findings and conclusions relevant to this case stand, and are incorporated by reference, whenever relevant, in the second phase of the trial on liability. [3] The validity and infringement issues are not decided in this case although all of the evidence necessary to a determination of those issues is already in the case. In the absence of the Mine Safety case and this court’s initial Technical Development decision, we would have been inclined to submit a decision determining all issues. However, the course followed in those decisions, at least where the license issue is clearly dispositive, serves the desired goal of judicial efficiency. In the event the court does not sustain the trial judge’s opinion on the dispositive license issue, no further trial will be required to determine validity and infringement since the record is complete and no further evidence is required as to those issues. [4] Plaintiffs assert that the crossover concept embodied in claims 9-11, 19 and 25 covers the use and selection of the temperature override signal. However, that contention is faulty. The claims covering the crossover require a selection between the signals of two different engine parameters, namely, speed and acceleration. The diode used in the temperature override circuit, held to be licensed in the initial case, compares a signal representative of the then existing engine temperature with a signal representative of a preselected maximum temperature. These signals are of the same engine parameter, namely, temperature. If this were not the case, then it would appear that defendant would be licensed to the general crossover concept by virtue of his license to practice the temperature override circuit as decided in the first decision of this court. [5] Many of the Offner working controls used this 20-80 percent relationship. The patent itself does not specify the exact crossover point. [6] Plaintiffs normally do not have to prove the first reduction to practice, but, as the decision will show, defendant has proven by a preponderance of the evidence that a crossover circuit was actually reduced to practice in the ’15202 contract. Thus, the burden was shifted to plaintiffs to show that the reduction to practice in the performance of the ’15202 contract was not the first reduction to practice.