Opinion
No. 155, Docket 20443.
March 7, 1947.
Appeal from the District Court of the United States for the Southern District of New York.
Action by Maxwell M. Bilofsky and another against Westinghouse Electric Supply Company and another for patent infringement. From a judgment dismissing the complaint, the plaintiffs appeal.
Affirmed.
Samuel E. Darby, Jr., of New York City (Paul Kolisch, of New York City, of counsel), for plaintiffs-appellants.
Victor S. Beam, Thomas F. Byrne, and William F. Kelly, all of New York City, and Ralph H. Swingle, of East Pittsburg, Pa., for defendants-appellees.
Before AUGUSTUS N. HAND, CHASE, and FRANK, Circuit Judges.
The appellant Bilofsky is the owner of U.S. Patent No. 2,341,520 which was granted to Maynard A. Babb on February 15, 1944, on his application filed February 20, 1941, for a starting and operating circuit for gaseous electric discharge devices. The circuit may be used in starting hot cathode, gaseous, fluorescent lamps but the patent is not so limited. The other appellant is a corporation controlled by Bilofsky which is the exclusive licensee under the patent. The owner and licensee brought this suit in the District Court for the Southern District of New York against the appellees who manufacture and sell a starter for hot cathode fluorescent lamps which is alleged to infringe. The trial court did not find any infringement and dismissed the complaint without passing upon the question of validity.
Fluorescent lamps are not lighted at once when the electricity is turned on as are incandescent lamps but have first to be started. The type here featured consists of a glass tube with an electrode at either end made of tungsten wire and separated by approximately the distance of the length of the tube which is filled with mercury vapor to which is added, to make starting easier, some rare gas like argon, neon or helium. This combination of vapor and gas within the glass tube is a very poor conductor when cold but when hot it becomes a good electrical conductor through which current flows readily from one electrode of the lamp to the other. By starting is meant the creation of conditions at the electrodes and throughout the tube which will permit the steady flow of current, at the voltage ordinarily supplied for commercial or domestic consumption, through the mixture in the tube. For simplicity, no description will be given of the remainder of what goes on within the tube to produce, by its effect upon the fluorescing coating on the inside of the glass, the steady illumination desired. We need now know only that to start the lamp it is necessary to heat the column of gas within the tube until it becomes a good electrical conductor; that this requires a stepping up of the input current from a voltage ordinarily of either 110 or 220 to the order of 350 and upwards; and that it is desirable to limit unsuccessful attempts to start an inoperable lamp.
Dench in U.S.Pat.No. 2,200,443 had, before Babb, shown how to start a fluorescent lamp. He did it by putting into the circuit a ballast coil and an automatic switch that would turn the current off and on continuously at predetermined intervals, choosing to use what is called a glow switch for this purpose though there are other known types which will serve as well. The effect of such turning off and on of the current which comes in from the source of supply at too low voltage to start the lamp is to create current surges in the ballast coil which becomes in effect a transformer that steps the current up to the required voltage for striking an arc across the space between the electrodes of the lamp. When, after a little lag, such voltage is obtained, the lamp, if it is in good condition, will be started when the arc has struck and heated the column of gas and will thereafter operate on the lower normal voltage of the supply current. Dench provided only for attempts to start a lamp which will continue to be made unsuccessfully when a lamp tube is out of order until the current is turned off by manually opening the current supply circuit. Until that is done there will be a flickering every time there is an attempt to strike an arc which will not only cause annoyance to people affected by it but prolonged repetition of such attempts are harmful to the starting system.
To do away with these disadvantages in Dench, Gref in U.S.Pat.No. 2,293,897 disclosed, before Babb, the introduction of a thermostatic switch into the starting circuit which would, after a limited number of unsuccessful attempts to start the lamp had been made, open the circuit and prevent more. When this was done the thermostat in his added switch would cool because no current flow kept it warm and, if not held from so doing, the bi-metallic bar would then have reversed its action and closed the starting circuit. That would have let more attempts to start the lamp be made until the thermostatic switch, actuated by its coil, had been heated again by the current and again opened the starting circuit. To prevent the limitation of his improvement to merely increased delay in continuous intermittent action, Gref used a spring latch so placed that it would catch the bi-metallic bar as it moved. This latch held his thermostatic switch open until it was released manually by pushing a button on the lamp fixture.
What Dench and Gref did was enough to make fluorescent lamps start when the tubes were in good condition and to cut off unsuccessful attempts to light them before undue annoyance, or unnecessary damage to the lamp's starting circuit, were caused. But no further attempts could be made to start the lamp until Gref's latch had been manually released to allow his thermostatic switch to close the circuit. If a lamp is in a more or less difficult place to reach, as is often the case, this requirement of a manual release of the latch is inconvenient for often it is desirable to have further attempts made to start the lamp before the replacement of the tube makes it necessary to go to it. Sometimes operable lamps fail to start and later will start and continue to be usable for a considerable time. They do not usually "burn out" once and for all like an incandescent lamp.
It was at this point in the development of starter control that Babb came in. What he did was to substitute something else for the latch of Gref which kept the Gref-added thermostatic switch from closing the circuit when the thermostat in it cooled after the current was cut off. Babb used a thermostatic switch which opened the circuit at the proper time, just as Gref's did, and had with it, instead of a spring latch, the following means for holding the bar from reversing. When the bi-metallic bar moved under the influence of the heater in the starting circuit, as Gref's did, that same movement brought it, not against a spring latch which held it from returning to its normal position when cool but against a terminal to close another circuit through which sufficient current then flowed to heat another coil. The heat from this second coil kept the bi-metallic bar from cooling and reversing its action to close the starting circuit and also kept it immovable against the added terminal to keep that circuit closed so long as the supply, or wall, switch was closed. A reduced current which would supply only the heat needed to keep the bar immovable could be used for heating the holding coil though the full circuit current would serve as well except to conserve current. The result was that after the permitted efforts to start the lamp had been made unsuccessfully the starting circuit would be opened and be held open so long as current reached Babb's holding heater coil. But when the supply switch, i.e., the wall switch, was operated to shut all current off the lamp Babb's holding coil would cool. That let the thermostat it had been holding move to its normal position when cool and close the starting circuit of the lamp. This permitted a recurrence of the starting operation whenever the current was again turned on the line in which the lamp was located. By turning off and turning on the current at the supply switch as many controlled attempts to start the lamp could be made as might be desired before the lamp itself was replaced. On one of these attempts the lamp might start and when it did the controls would be by-passed as in Dench and there would be steady illumination. If, and when, later the lamp failed, additional attempts to start would follow automatically, but continue, if unsuccessful, only as permitted by the controls.
But in Babb the closing of the supply, or wall, switch was not the only way to "reset" the starter. That would, of course, not only cut the current off the defective lamp but would also extinguish any other lamps which might be in the same circuit. Such extinguishing of other lamps in the circuit could be avoided, when desired, by opening only the starting circuit of the defective lamp. As the input current flowed through one of the electrodes in the tube the removal of the tube opened Babb's heater circuit and allowed his holding heater coil to cool. That let the thermostat return to its normal position where it closed the break in the lamp's starting circuit at that point. The insertion of a new tube closed the circuit which had been opened by the removal of the old and energized the starting circuit again. In this way the mere removal and replacement of a defective fluorescent tube produced the same effect that the removal and replacement of a defective incandescent lamp does wherever such lamps are used. Presumably the new lamp would start but if not, the controls would go into action to halt the starting process and keep it halted as before until what was wrong was remedied.
The accused device does not open the starting circuit and, as a part of the same movement of a bi-metallic bar, close a heater circuit as in Babb. It uses a heater, as in Gref or Babb, to open the starting circuit with some actual, but unneeded, aid from another heater which is always in a live circuit, and consequently heated, while the supplying current is on, the additional coil being fed a comparatively small amount of current through an always closed by-pass circuit partially blocked by a resistance. The heat thus generated is not, however, enough to actuate the bi-metallic bar alone and so cannot open the starter system but it does supply heat enough to hold the bar immovable when the combined heat from the two coils has moved it to open the starting circuit. This effect will be done away with whenever the supply switch is opened, which will occur, as in Babb, whenever the wall switch is closed or the tube is removed, both electrodes of the lamp being in the circuit with the resistance through which this holding coil is supplied with current. Thus Babb's result was obtained by the accused device in substantially the same way but without utilizing the movement of the bi-metallic bar which opened the starter circuit to close a holding heater circuit, the latter being continuously in a circuit and always heated and ready to play its holding part whenever required.
We cannot agree that Babb's claims relied upon, of which 1, 2 and 5 were designated as typical, must be too narrowly construed to cover the accused control. We think that control by means of a continuously heated coil in defendant's always closed by-pass circuit is the fair equivalent of Babb's control by means of a normally open by-pass which is closed to heat a coil which does the same thing in the same way when heated, and that the patent, if valid, is infringed.
It is clear enough that Babb was the first to add to Gref's automatically opened starting circuit a device for holding it open which would be released automatically when, and only when, closure was desired. It is also apparent that this had some useful features not found in the latch-holding means of Gref since it eliminated the need for going to the lamp until a new tube was to be inserted. But so much alone is less than enough to support the patent. What is disclosed must, to be patentable, be new in the sense that Babb was the first to accomplish that result in substantially the way he did it. Pennsylvania R.R. Co. v. Locomotive Truck Co., 110 U.S. 490, 4 S. Ct. 220, 28 L.Ed. 222; Blake v. San Francisco, 113 U.S. 679, 5 S.Ct. 692, 28 L.Ed. 1070; Continental Scale Corporation v. Harrison Wholesale Co., 7 Cir., 132 F.2d 463.
While Babb was the first to use in the starting system of a fluorescent lamp an additional by-pass circuit which was normally open and would be closed and held closed electrically so long, and only so long as desired, he was not the first to do that very thing in the field of discharge lamp starting systems to which his contribution to the art belongs. Spaeth in U.S.Pat.No. 1,881,976 disclosed a starter for mercury arc lamps and the like, which had a normally open by-pass circuit that would be closed automatically whenever the lamp failed while the supply current was turned on and so prevent any attempt to start the lamp "until it is manually restored to operation at an opportune moment." Here we find the disclosure and use of a normally open by-pass circuit which would be automatically closed by a magnetic switch, the equivalent of a thermal switch, to by-pass the starting circuit. This by-pass circuit was held closed by a magnetic holding coil until current was turned off the magnet by opening the wall switch. Doing that was the manual restoration to operation to which Spaeth alluded in the above quotation. The magnet would then cease for want of current to hold the by-pass circuit closed just as Babb's thermal switch did, and the starting circuit would become operable when the current was again turned on it. The adaptation of Spaeth's normally open by-pass circuit by Babb required no more than a comparatively simple arrangement of well known electrical parts to substitute a thermal switch and holding coil for the equivalent magnetic switch and holding coil which, when placed as they were relative to each other, became a substitute for Gref's spring latch. There is ample evidence to show that this was the routine work of a competent electrical engineer who would require only the will to do it as a spur to achievement. Though complete anticipation of Babb was not shown, his improvement of Gref's control by installing the slight adaptation of Spaeth's automatic re-set as a substitute for Gref's manual one in the way described was not the substantial step beyond the prior art which is invention that will support a valid patent. Cuno Corp. v. Automatic Devices Corp., 314 U.S. 84, 62 S.Ct. 37, 86 L.Ed. 58. "The mere aggregation of a number of old parts or elements which, in the aggregation, perform or produce no new or different function or operation than that theretofore performed or produced by them, is not patentable invention." Lincoln Co. v. Stewart-Warner Corp., 303 U.S. 545, 549, 58 S.Ct. 662, 664, 82 L.Ed. 1008. The alternative way of opening the supply circuit without using the wall switch, by the removal of the tube and its closure by the insertion of a tube was but the normal and, indeed, inevitable result of having one of the tube's electrodes in the circuit which was, of course, old.
Judgment affirmed.