Westinghouse Electric & Mfg. Co. v. Prudential Ins. Co. of America

155 F. 749, 1907 U.S. App. LEXIS 5296

• Court: U.S. Circuit Court for the District of New Jersey
• Decided: August 12, 1907
• Status: Published

Opinion

LANNING, District Judge.

The question presented by this case is whether the defendant has infringed complainant’s patent No. 582,481, dated May 11, 1897. The patent was issued to the complainant as assignee of Edwin E. Nolan. It is entitled “Fastening Means for Core-Plates df Electrical Machines.” In his specification Nolan says:

“My invention relates to electric generators and motors, and has particular reference to means for fastening the laminse of the cores of such machines in position. The object of my invention is to provide a simple, inexpensive, and: efficient means for fastening the laminse of the cores of electrical machines together, and to the casting constituting the support of the same, and one which may be readily inserted and as readily removed when it is desired to dismember the machine for any purpose.”

The claims are four in number, and the charge is that each of them is infringed. They are as follows:

“1. A core for electrical machines, comprising a casting having a cylindrical surface and provided with a circumferential flange at or near one end and a circumferential groove at or near the other end, an annular plate adjacent to said groove, laminae clamped between the flange and the annular plate, and a fastening-ring located partially in said groove and bearing against the outer side of said annular plate.

“2. A core for electrical machines, comprising a casting having a1 cylindrical surface and provided with a circumferential flange at or near one end and a circumferential groove at or near the other end, an annular plate provided with a shoulder on its outer side, laminae clamped between the flange and the annular plate, and a fastening-ring located partially in said groove and bearing against the shoulder on said plate.

“3. A core for electrical machines, comprising a casting having a cylindrical surface and provided with a circumferential flange at or near one end and a circumferential shoulder of less diameter at or near the opposite end, an annular plate adjacent to said shoulder, laminae between said annular plate and. the flange on the casting, and a spring-ring interposed between said annular plate and said shoulder.

‘•4. A core for electrical machines, comprising a casting having a cylindrical surface and provided with -a flange adjacent to one end and' with a shoulder-adjacent to the opposite end, an annular plate provided with a shoulder, laminae between said annular plate and the flange on the casting, and a spring-ring interposed between the shoulder on the annular plate and the shoulder on the casting.”

The figures annexed to the patent are as follows:

The parts referred to in the first claim are the casting having a cylindrical surface, 2; the circumferential flange, 3; the circumferential groove, 4; the annular plate, 5; the laminae, 6; and the fastening-ring, 8. The parts referred to in the second claim are the casting having a cylindrical surface, 2; the circumferential flange, 3; the circumferential groove, 4; the annular plate, 5; the shoulder, 7, on the outer side of the annular plate; the laminas, 6; and the fastening-ring, 8. The parts referred to in the third claim are the casting having a cylindrical surface, 2; the circumferential flange, 3; the circumferential shoulder, 4a; the annular plate, 5; the laminae, 6; and the spring-ring, 8. The parts referred to in the fourth claim are the casting having a cylindrical surface, 2; the flange, 3; the shoulder, 4a; the annular plate, 5; the shoulder, 7, with which the annular plate is provided ; the lamina;, 6 ; and the spring-ring, 8.

In the specification Nolan further says:

“In order to assemble the parts in the position shown in Fig. 1, the laminee constituting the outer portion of the core are first built up upon the casting, 2, and the annular plate, 5, is then placed in position and pressed inward against the plates by any suitable mechanism to a point slightly beyond that shown in the drawings. The ring, 8, is then sprung into place in groove, 4. when the plate, 5, is released and springs outward against the said ring. The parts are thus rigidly clamped in position, and by reason of the shoulder, 7, there is no possibility of the fastening-ring, 8, being thrown out of its normal position by centrifugal force due to rotation of the armature. It 'will be readily seen that, if it is desired to separate the parts of the core, it will he merely necessary to subject the plate, 5, and laminae, 6, to sufficient pressure to permit the ring, 8, to pass the shoulder, 7, when it may be sprung out of the groove, 4.”

The distinguishing feature between the patent in suit and the defendant’s device is that in the patent in suit the casting, 2, is provided with a cylindrical surface, w'hile in the defendant’s device the casting is provided with three arms having curved ends, and not a continuous cylindrical surface. This difference, however, is clearly an immaterial one. Indeed, Prof. Kennelly, one of the experts who testified for the defendant, himself says:

“The blue print of the Bullock construction [the defendant’s devicej shows a spider of three arms provided with one integral flange, D, and one movable flange, A. This spider is not a cylinder, but may be regarded as the equivalent of a cylinder. It may be called a cylindrical spider. On the cylindrical spider are mounted a number of core-plates, assembled and fastened between the flanges. The movable end-flange is clamped by a lock-ring resting in a groove, or in a series of three grooves, one on each arm of the spider. If the spider were a true cylinder, this groove would be a cylindrical groove. The defendant’s Bullock construction, as represented in the blue print, is the ordinary construction of the prior art, of a fixed flange and a movable flange, with a clamping device of the calendering-rolls of the prior art, viz., the lock-rings of the Perkins, Granger, and Ingram patents. The structure of the Bullock armature blue print differs from the structure of the figures in the Nolan patent in suit in certain details. For example, the Bullock lock-ring has square cross-section. The Nolan lock-ring has rectangular cross-section, being relatively ■ tall and narrow. The Bullock groove has a width about twice as great as the width of the lock-ring; whereas, the Nolan groove in the cylinder is but little wider than the lock-ring. The recess in the outer side of the movable flange has a depth of about one-sixth of the width of the lock-ring in the Bullock construction, and about half the width of the lock-ring in the Nolan construction. The Bullock supporter is a spider of three arms; whereas, the Nolan supporter is a spider of three arms covered by a cylinder. All of these differences I regard as immaterial, and as matters of mere detail.”

The record of the case shows that from 1882 down to 1897, when the patent in suit was granted, three different methods had been employed for holding in their places the laminae of an armature. The first maybe designated as the “pin method,” the second as the “shaft-nut method” and the third as the “bolt method.” The method described in the patent in suit may be called the “spring-ring method.” The pin method, illustrated by the Weston patent, No. 401,669, applied for September 22,1882, in the very early stage of the art, is unsatisfactory, because the laminae, which are held in position by pins driven into a wooden hub surrounding the shaft, are likely to be displaced by the jarring and rapid rotation of the armature. The shaft-nut method is one by which the lamina are held in position by a nut screwed on the shaft. This method is ill-adapted to very large armatures, in which the lamina are not seated on the shaft, but, as in the patent in suit, on the outer surface of a spider rigidly mounted on the shaft. By the bolt method the lamina are held in place by bolts running parallel with the shaft through the lamina. This method has a disadvantage in the fact that the bolts must be placed so near to the shaft or to the outer surface of the spider as to be out of the magnetic circuit, or, if they be within the magnetic circuit, must be insulated. Another disadvantage, is that, as the bolts in large machines are necessarily many in number (there is an illustration in the record of a machine having 40 or 50 of them), there is difficulty in threading the laminae on the bolts. A disadvantage common to both the shaft-nut method and the bolt method is that the nuts are likely to loosen, so that the laminae are not held firmly in their positions. It is possible that the use of the spring-ring method is not always practicable; but, where it is, its advantages over the other methods in use are obvious. The spring-ring fitted in a groove, and held in position in the manner described in the patent in suit, cannot recede from its position. It is not necessary to insulate it. It is applicable to, large machines. It is easily placed in position, and easily removed, so as to permit, with slight expense, the assembling and removal of the laminae. ■ The device of the patent in suit is therefore a useful one; but the question is: does it rise to the dignity of invention ?

The evidence shows that a large armature sometimes revolves with a speed, at its surface, of one mile or more per minute. In such an armature the centrifugal force is so great that it is desirable to have the spring-ring, where it is used, held firmly in the groove. Claims 1 and 3 omit any reference to a shoulder on the outer side of the annular plate, and they are' fully satisfied by annular plates resting against the inner surface of the spring-ring, without any shoulder on the annular plates for holding the spring-ring in its groove. While nothing in the prior art relating to electrical machines has been referred to which clearly shows the use of a spring-ring for holding the laminae in position, the use of such a ring for holding together the parts of a machine has long been well known in other mechanical arts. An illustration of such use is to be found in the Granger patent, No. 407,858, issued in 1889. That was a patent for a cloth-calendering roll. The roll, consisting of corn husks and cotton, is built up on a shaft and solidified by hydraulic pressure. At each end- of the roll is a metallic head-piece, and the roll, thus constructed, is locked in its position by a split-ring, adjacent to the outer part of each of the head-pieces and fitted in an annular groove on the shaft. The Perkins patent, No. 355,026, granted in 1886, is for a paper-calendering roll. It describes a roll mounted on a shaft, with a metallic head-piece at each end. The head-pieces, with the intervening roll, are held in position by two split-rings, located in annular grooves on the shaft just outside of the head-pieces. There is also a German publication of Ganz & Co., referred to in the record, which appeared in 1893. A drawing accompanies the publication, and shows a small armature with the laminae seated directly on the shaft and clamped between two end plates. It also shows that the plate at one end of the laminae is held in position by a device the nature of which is not explained in the publication, but which, it is clear, is not a nut. It seems to me to illustrate a split-ring. I think the use of a split-ring for holding the parts of a machine in position in the manner described in claims 1 and 3 of the patent in suit, was, at the date of issuing the patent, an old and very well-known use. .Whether such a ring be used for keeping in their proper positions materials mounted on a shaft for the construction of a cloth-calendering roll or a paper-calendering roll, or a core for an electrical machine, it has but the one function. In my judgment, therefore, claims 1 and 3 are invalid for want of patentable novelty.

But claims 3 and 4 each contain an additional element, namely, the shoulder on the outer side of the annular plate. This shoulder extends over the split-ring, and its function is to keep the ring from being thrown out of its groove by the centrifugal force of the rapidly revolving core. Nothing like this device is disclosed in any prior patent to which reference has been made. The Westinghouse patent, No. 583,-494, granted on the same day as the patent in suit, shows a groove in the outer surface of the casting, and also a groove in the inner periphery of the annular plate. When the annular plate is placed on the casting against the laminae and pressed to its position, the two grooves are opposite to each other. Into these grooves molten Babbitt metal or other fusible metal is poured, through openings provided for the purpose, where it quickly solidifies and forms a ring holding the annular plate in its position. The Lusk patent, No. 508,177, the Weisell patent, No. 537,569, and the Harris & Browning patent, No. 61,630, are for improvements in nut-locks. Notwithstanding the insistment of counsel for defendant that these patents disclose anticipations of claims 3 and 4 of the patent in suit, I think they fall far short of it. They might, perhaps, be regarded as anticipations of claims 1 and 3; but they embody no device for overcoming the effect of centrifugal force. The devices described in them simply prevent the loosening of nuts on bolts and shafts.

In my opinion the complainant’s patent is good as to claims 3 and 4, and invalid as to claims 1 and 3. I think, also, the defendant’s device is an infringing one as to claims 3 and 4.

There will be a decree in accordance with these views.