Hall Signal Co. v. General Ry. Signal Co.

169 F. 290, 94 C.C.A. 580, 1909 U.S. App. LEXIS 4582

• Court: Court of Appeals for the Second Circuit
• Decided: March 16, 1909
• Docket: No. 241
• Status: Published

Opinion

COXE, Circuit Judge.

This is an equity action for the infringement of letters patent No. 470,813 granted March 15, 1892, to Adoniram J. Wilson, assignor to the complainant, for an improvement in automatic electric block signals for railways. The application was filed November 9, 1891, which must be regarded as the date of the invention.

The drawings represent, diagrammatically, the invention applied to a series of seven blocks on a line of railway, the sections varying in length to meet the requirements of different railways. The intermediate blocks—those between the first and the seventh as shown—may be duplicated indefinitely. The specification describes, by reference to the drawings, the successive operation of a train upon the signal devices in passing over the track sections represented. In the preferred arrangement shown in the drawings it is intended that each alternate signal shall be merely a cautionary signal which the engineer may pass but he does this with notice to look out’ sharply for the next signal. The intervening signals are absolute signals, which he should never pass except when they stand at safety, and if they do not clear before him he must stop until they do. In the broad invention the rail circuits are normally closed and the signaling circuits are normally open at a circuit closer which is constructed to be closed by the action of the train upon one of the rail circuits. The signaling circuit' is also capable of being broken by one or more normally closed circuit breakers which are constructed to be opened by the action of the train on one or more., of the next succeeding rail circuits of the line.

The claim involved is as follows:

“1. In a block signaling apparatus, the combination, with a series of normal-ly-elosed rail-circuits, of a series of normally open or broken signaling-circuits,. each signáling-cireuit including a normally-open circuit-closer constructed to be closed by the action of the train upon one of the rail-circuits and also including a normally-closed circuit-breaker constructed to be opened by the action of the train upon the next succeeding rail-circuit of the line, substantially as and for the purpose set forth.”

It is a combination claim for a block signaling apparatus, having the following element's:

(1) A series of normally closed rail circuits.

(2) A series of normally open signaling circuits.

(3) Each signaling circuit including (1) a normally open circuit closer, constructed to be closed by the action of the train upon one of the rail circuits and also including (2) a normally closed circuit breaker constructed to be opened by the action of the train upon the next succeeding rail circuit of the line.

The defense is that the claim, if broadly construed, is invalid in view of the prior art, and, if limited to the device shown and described, is not infringed.

The patent relates to the so-called normal danger system as distinguished from the normal safety system, it being contended that the patentee was the first to make the former a safe, reliable, simple and workable system—having marked advantages over the systems of the prior art.

The record presents' such a- wilderness of irrelevant matter that it is not easy to deduce therefrom a perfectly clear understanding of the respective advantages and defect's of these rival systems. We understand, however, that the normal danger system is equipped with nor- . mally open signaling circuits so arranged that the force which sets the signal at safety never operates except when a train is approaching. .The signal is held by gravity in the position indicating danger so long as no trains are present and is sent to safety by the on-coming train.

• Under the normal safety system, when the engine driver approaches a block the semaphore at the entrance is in an inclined position by day • and shows a green light by night indicating that the track ahead is clear and that he may safely proceed unless the signal rises to a horizontal position, or shows a red light. In other words he is to assume that the track ahead is clear and safe until the contrary appears by the appearance of a danger signal.

The normal danger system, on the contrary, always shows a red light at night and a horizontal semaphore by day, indicating danger; ..the safety signal appearing only on the approach of a train and when • the block to be protected is clear. In the first system the assumption is that all is safe ahead till “danger” appears, and, in the second, that .all is unsafe ahead until “safety” appears.

In the one, if there be no movement of the signals, the engine driver keeps on, in the other, unless there be movement of the signals, he .stops.. To quote from the pamphlet said to have been written Iw the president of the defendant:

“Consistency argues that all the signals shall indicate danger, except where they are cleared for the passage of a train.”

From the viewpoint of safety it seems to us that the preponderance of evidence is clearly in favor of the complainants’ system. If, through climatic or other influences, the signals cease to operate' it seems obvious that it is much safer that they should clog in the danger rather than in the safety position. In the one case the failure of the signal to operate might result in the unnecessary stopping of-a train, in the other it might result in a rear-end collision. It is not at all improbable that laymen may give greater weight to this feature of the system than practical railroad men, but it' seems to us that an improvement which reduces, in any degree, the chances of derailment and collision, and thus safeguards life and property, is entitled to greater consideration than one which deals only with economics.

We do not intend to intimate that the Wilson system does not, from a practical point of view, compare favorably with the normal safety plan. Quite the contrary appears. As the electric current is in operation during the comparatively short time that the signal stands at clear it is obvious that there is very much less consumption of battery material. As one of the witnesses points out, the expense of maintenance increases only in proportion to the amount of use, whereas in the normally clear system the expense is greatest when apparatus is least used. Scientific men, versed in the art of railway signaling had early recognized that the logic of the situation was all with the normal danger theory. Thus, in 1842, Sir William Cooke writes:

“I think it highly desirable that the ordinary or quiescent condition of the station signal should be a state of danger, and not a state 'of safety, so that a train should never run into a station without special guaranty that it was prepared to receive it.”

This was before the days of automatically controlled block signaling but it shows that early in the art of railroading it was recognized that the ideal position of the signal was one indicating normal danger. The difficulty was not with the theory but with the means to make the theory practicable. This was an easy task when applied to a manually operated system but the record shows that its practical application to an automatic system was obstructed by impediments which it was found impossible wholly to remove, although many skillful and scientific men were at work on the problem. At the date of Wilson’s invention, in 1891, the art of railway signaling was in an embryonic condition. Men of genius and skill, cognizant of the advantages of the normal danger plan, were endeavoring to produce a practical system, but none, prior to Wilson, succeeded in perfecting a plan to which hard headed railroad men were willing to intrust the lives and property committed to their care.

The defendants rely upon patents granted to five inventors—Robinson, Spang, Pope, Gassett and Westinghouse—to defeat or fatally limit the claim in suit. William Robinson, in 1872, received a patent, relating to a normal safety system and in 1879 he received a British patent for a normal danger system. It is not pretended that either of' these patents anticipates but, if we comprehend the defendant’s contention, it is that Robinson possessed the knowledge, as demonstrated by the description and drawings of his patents, to convert the plan of the earlier patent into a successful normal danger system. A sufficient answer is that patents are not defeated by what prior inventors might have done. They, like other men, must be known by their works, and no one pretends that the plan described by Robinson in the British patent 3,479, in 1879, was operative. At least no one ever attempted to operate it. No reliable protection is shown and it is doubtful whether it could be operated in connection with a block system.

Robinson is conceded by all to have been one of the most accomplished signal engineers of his time. The fact that he failed to convert his 1872 system into a normal danger system is mute but persuasive testimony that it was not an obvious thing to do. The defendant’s argument would be more specious if the 1879 patent were not here to confute it. If this distinguished engineer had not made the •attempt it might be said with considerable force that had he done so hie certainly would have succeeded. But what is left of the contention in view of the fact that he did attempt it and failed? Would the court be justified in assuming that a skilled mechanic could do what Robinson could not do?

Four patents. Nos. 164,227, 164,228, 168,059 and 208,995, were issued to Henry W. Spang, from 1874 to 1878, describing with great particularity eight normal danger plans which never went into use. Being radically wrong in theory they could never be made practically useful. The initial errors in Spang’s system were insufficient length of the clearing section and lack of efficient protection for the train while on that section. The defendant’s counsel deny this but we are not convinced by their argument. We agree with the judge of the Circuit Court .in holding that the protection afforded by the Spang patent, 164,227, was insufficient “owing to the magnetization of a magnet which in this type of signaling was necessary in order to hold the signal- in its danger position.” In other words the danger signal was not held in position by gravity but by electric energy which if withdrawn by breakage or for any other reason was liable to cause disaster by producing the misleading impression that the track was clear.

Frank L. Pope in 1873 was granted a patent, No. 143,529, covering a normal danger system or, more accurately, two such systems. Here too we have the fatal defect of clearing sections but 50 feet in length and wholly unprotected. The plan was never put into actual use and is admitted to be wrong in principle.

Oscar Gassett, in 1882, was granted a patent, No. 251,867, for a normal danger system, which like all the others which preceded it was never a practical success. The language of the description is somewhat indeterminate upon the question of duplication if the entire track is to be protected, and this has led to sharp differences of opinion among the experts which the court cannot undertake to decide. It is enough that a patent which was respected by competitors for 13 years and which covers a system which has been in successful operation during its entire life cannot be invalidated by the ambiguous language •of a patent which has added nothing of value to the art. In such circumstances, unusually clear and perspicuous language is necessary. Success cannot be anticipated by failure. When the problem of aerial navigation is finally solved by the construction of a secure dirigible air-ship, it is safe to predict that the inventor’s patent will not be invalidated by a prior structure, no matter how perfect it may be, which was never known to fly. The difficulty with Gassett’s plan is the same as with his predecessors, viz., short and insufficient protection for the clearing sections. When he attempts to afford entire protection he has recourse to track instruments.

George Westinghouse, Jr., in 1887 was granted a patent (No. 360,-638) for a signaling apparatus. Like the Robinson patent. Westinghouse shows normally closed track circuit, but in a normal safety system. Like defendants he mounts both his home and distant signal on the same post but in other respects he adds nothing to the prior art. The judge of the Circuit Court has considered the Robinson. Spang and Gassett patents more in detail and we do not deem it necessary to repeat his observations regarding them. We have endeavored to understand the combinations of the patent, of the defendant and of the various patents which have a direct relation to the controversy; but to describe them in detail, without recourse to diagrams and reference letters, is impossible. Such descriptions even though the material were at hand would be of no general interest and would serve no useful purpose. We must content ourselves by giving our conclusions without entering into minute details.

We have, then, a situation where a problem is taken up by a number of skilled specialists, all working for its solution, each contributing something of value but failing to produce a safe, workable plan. If we may judge the prior systems from the fact that none of them, except in one or two tentative instances, went into actual use, the inference is plain that railroad men were unwilling to take the risk of installing them. When Wilson took up the work it had virtually been abandoned by the others, they had tried and failed and there was ‘no reliable normal danger plan then in existence. That Wilson solved the problem we have no doubt, the systems installed under his patent are successful and are rapidly growing in popularity. We do not consider him a pioneer in the sense that he discovered a new art. The idea of a normal danger system was old but he was the first to harness it and set it to work. So much he has contributed and to this extent he is entitled to protection.

The subject is so abstruse and complicated and the opinions of those who are skilled in the art are so diametrically opposed upon almost every mooted question that we are forced to bring to our aid presumptions drawn from facts about which there is no dispute. The normal danger system is a meritorious one and, concededly, has some advantages over the rival system. Theoretically it is the ideal system and was the one employed when signals were manually operated. Repeated efforts had been made to secure a perfect electrically controlled normal danger system but always without success. When, therefore, the question is whether a prior system described and diagrammatically illustrated is practically operative, any doubt which arises should be resolved in favor of the working plan and against the paper plan. It matters not that all the elements of the claim were old when segregated. Undoubtedly they were old, but they were never combined before to accomplish the result which Wilson accomplished.

In approaching'the question of infringement it must be remembered that the controversy is confined to the first claim which covers the broad invention and has no connection with the second claim which relates to the specific means described and diagrammatically shown. Having found that Wilson was the first to devise a successful normal danger system affording full protection to the train in front and rear for every inch of track, it is manifest that a construction should be placed upon the claim as broad as the invention and that he who uses the invention without license should be held to infringe no matter what else he may use. It must also be borne in mind that the combination of the claim is expressly related to “a block signaling apparatus” as understood by those skilled in the art. In other words the claim cannot be construed to refer to “a succession of unrelated units” but to “a series of consecutive blocks” which are necessary in a block signaling apparatus.

Even were we able to do so, we find it unnecessary to follow the discussion, found in the record and briefs, of the differences between the plan of the patent and the defendant’s plan for the reason that infringement does not, in our judgment, depend upon the use of the precise apparatus shown by the patentee in illustrating his system. The installation by the defendant on the Illinois Central Railroad is certainly “a block signaling apparatus” on the normal danger plan. It would seem that when this system was installed the defendant was aware of the fact that it would infringe the claim of the Wilson patent if liberally construed. After an examination, the expert of the defendant reported that in his judgment, the claims “would be held to be invalid or so limited in their- scope that the circuits which we proposed to use would not infringe.” The president of the defendant assented to this view. The inference is plain that they concluded to fake the risk, knowing that they would be held to infringe unless a limited construction was given to the first claim.

The defendant’s system is an embodiment of the Wilson system. It has all the elements of the claim operating in substantially the same way to produce a like result. In the defendant’s apparatus there is a series (nine) of normally closed rail circuits and a series of normally open or broken, signaling circuits. Each signaling circuit has a normally open circuit closer and a normally closed circuit breaker operating as described in the claim. The approaching train, if the track be clear, produces a “clear” indication of a signal ahead, while a signal in the rear stands at “danger” thus affording complete protection.

The defendant locates its home and distant signals on the same post, but this, though quite likely an improvement, is a matter of detail in no way changing the operation of the combination which produces the same result upon the series of distant and home signals respectively, as in Wilson’s system. Such inconsequential changes do not go to the essence of the invention and do not avoid the claim. The defendant has moved the distant signals as shown in the .drawings of the patent and has located them on the posts of the home signals in the rear. The differences between its system and that of the patent upon which the defendant bases its contention of non-infringement are stated by complainants’ expert, Mr. Wagner, as follows:

“Defendant does not infringe:

“(1) Because it uses subsections.

“(2) Because it clears distant signals too late.

“(3) Because it clears borne signals too soon.

“(4) Because it clears distant signal through home signal on ahead.

“(5) Because it does not clear home signal by a circuit closer ‘a’ on the block to the rear, but only holds home signals to- clear by this circuit closer ‘a.’

“(6) Because it puts distant signal to danger by closing a shunt through circuit closer ‘a.’ ”

His answers may be briefly paraphrased, in the same order, as follows:

First. Defendant’s blocks were divided into subsections because they were too long to be safely and economically operated each as a single rail section.

Second. The distant signal in defendant’s system is cleared at a point 2,000 to 3,000 feet to the rear and remains in the clear position while the train traverses this distance which is amply sufficient to clear the signal.

Third. The defendant’s home signal is cleared throughout the entire block to the rear as shown in the patent by a normally open circuit closer.

Fourth. The ciixuit of the distant signal is closed as described in the patent independently of the home signal on ahead. The actual clearing of the distant signal is merely arrested until the home signal has moved to the clear position and by doing so has closed a shunt.

Fifth. Defendant’s contention numbered 5, supra, requires the diagram to which it refers to explain it satisfactorily. But without the diagram it is plain that it is based upon a somewhat technical construction, the fact' being that the system would be practically useless were it not for the circuit closer “a,” which closes and holds the signaling circuit closed until the train reaches and enters the block protected by the signal.

Sixth. The distant signal is not sent to danger by a shunt closed by the circuit closer “a” but by opening the normally closed circuit breaker “f.”

We think it unnecessary to dwell further upon the question of infringement. The subject is fully discussed in the opinion of the Circuit Court and is illustrated and elaborated by means of a diagram showing the defendant’s system.

The defendant argues that, in any event, the costs of the rebutting testimony should be borne by the complainants. When it is remembered that the issue is confined tó a single claim of a patent' having but one drawing and three pages of description, we are convinced that at least half of the record of 2,400 pages is irrelevant and immaterial. As, however, the judge of the Circuit Court’ has considered and decided this question and has awarded this complainant but two-thirds of the amount expended in printing and taking testimony, we do not feel disposed, assuming the question to be properly before us, to interfere with his adjustment of the costs, which is certainly not unreasonable.

The patent having expired, the paragraph of the decree providing for an injunction should be modified by the Circuit Court.

The decree is affirmed with two-thirds of the costs of this court.