The Obry device for steering torpedoes is an application of the principle of the gyroscope to the Whitehead torpedo. These torpedoes, before the installation of this device, had no vertical rudders, their course in a straight line being insured by permanently clamping two vertical vanes upon the horizontal tailpiece, the proper angle being determined by a series of trial runs. The Obry gear requires the placing of two rudders in the vertical tail-piece, one in the upper and one in the lower half. The combined area of these rudders is equal to twice the area of the fixed vanes, so that their action will overcome that of the vanes. It is found in practice that it makes no difference whether the vanes are on or off, or, if on, at what angle they are set.
In the Howell torpedo the driving mechanism is itself a gyroscope, which, revolving in the plane of the line of fire, tends to keep the torpedo in that line. If for any cause it be turned aside, the resultant of the forces causes the torpedo to roll, which, by the action of a transverse pendulum, moves the vertical rudders and brings the torpedo back to its course. In the Obry device the gyroscope is mounted on gimbals, so that, no matter how the torpedo may turn, the gyroscope remains in the plane in which it was first started. The torpedo turning about this, or, what is relatively the same thing, the gyroscope turning in the torpedo, moves the valve of a steering engine or servo-motor which works the vertical rudders.
This is the principle upon which the device works. The arrangement of the parts by which this is accomplished is not very complicated, but the greatest care in the making and assembling is necessary. All weights are counterbalanced, and the bearings are such as to reduce friction to a minimum.
In the earlier types the device was mounted in the after balance chamber on the bulkhead abaft the engine-room, and the winding socket was in the side of the torpedo. The new torpedoes for this Government, however, will be fitted with the device mounted on the floor of the balance chamber, accession being given by means of a door in the bottom, similar to the engine-room door, except that it is water-tight. The winding socket also will be in the bottom of the torpedo. This change of position necessitates some changes in the arrangement of the parts, notably in the actuating mechanism, but the general principle remains unchanged, and I shall endeavor in what follows to describe briefly the arrangement, noting where the old and new types differ.
The device consists essentially of three parts—the gyroscope, the actuating gear, and the servo-motor that works the vertical rudders.
The gyroscope or fly-wheel is made of close-grained, homogeneous bronze, carefully cast so that the metal is of the same density in all parts, and then turned down smooth, great care being taken to get the geometrical center and the center of gravity exactly coincident. The axle is mounted in jeweled bearings in a gimbal ring, which is in turn mounted in a second ring, and the whole mounted on a bronze standard. The gyroscope is thus left free to move in any direction irrespective of the course, inclination or roll of the torpedo.
The fly-wheel is caused to revolve rapidly by means of a pinion on its axle which engages with a large cog-wheel, on the spindle of which is wrapped a stiff steel wire spring. There is a piece cut out from the rim of the cog-wheel, making it rather a toothed segment than a complete wheel, and in this space the pinion revolves after being started by the toothed part of the wheel.
In the earlier devices the spindle of the cog-wheel on which the spring is wound was horizontal, being above and parallel to the axle of the gyroscope. The cog-wheel thus engaged directly with the pinion. The end of the spindle was fitted in a bearing in the side of the torpedo, with a squared recess sunk in it to take the wrench for winding the spring. The device as fitted to our torpedoes has the spindle vertical and the winding socket in the bottom of the torpedo. In order that the cog-wheel may engage with the pinion it is changed into the segment of a crown wheel, or circular rack, shaped like an inverted saucer, with teeth around the edge, and a piece taken out of one side. The spindle, which is hollow, passes up through the center and has a rod sliding in it which operates an important part of the mechanism.
This is the arrangement of levers for locking the gyroscope, so that when it is not in operation it will be held in a plane parallel to the axis of the torpedo. After it has been given its rapid revolution, and is capable of maintaining its plane of rotation unchanged, it is released, so that the torpedo may be free to turn about it as it will.
The first locking of the gyroscope is done by the motion of the rod before mentioned, the inner end of which projects beyond that of the spindle about one-quarter of an inch. The rod is held in this position by its outer end being pressed against a covering plate screwed into the shell of the torpedo over the end of the spindle. As this covering plate is screwed out, the rod follows it under the influence of a spring until its inner end is flush with that of the spindle. This motion allows two little levers with flat ends or shoe-pieces to come down under the action of springs on the two gimbal rings over their pivots, holding them at right angles to each other and placing the gyroscope in a plane parallel to the axis of the torpedo. The gimbal rings are held thus as long as the covering plate remains off, that is, while the spring is being wound.
The winding is done by a square-ended wrench fitting in the socket in the end of the spindle, which wrench is bored out to fit over the protruding end of the rod. A little less than one turn suffices to wind the spring, in doing which the segment is turned so that its teeth just engage on the edge of the blank space with those of the pinion. While the winding is going on. the gyroscope is held steady in its plane by the little levers on the gimbal rings, but it is necessary to provide some device for holding it after the covering plate is screwed in (which pushes in the rod in the spindle and lifts the little levers), and especially while the toothed segment is sweeping across the pinion and giving the gyroscope its momentum. This is done as follows:
There is, on the spindle and turning with it near its outboard end and just within the shell of the torpedo, a plate containing on its flat surface a cam on which rests the short end of a bell- crank lever. The other end of the bell-crank lever is fitted with a little hopper-like receptacle which takes over a stud on the inner gimbal ring in a prolongation of the axis of the gyroscope. This effectually prevents any motion of the gyroscope except that of revolution, and the cam is so shaped that the short end of the bell-crank lever is not moved (which action releases the gyroscope) until the segment has swept across the pinion and come to the blank space, where it brings up against a stop. In other words, the cam is a circle with one little jog in it at a place corresponding to the blank in the segment. Thus, when the spring has been wound and the covering plate screwed in, the gyroscope is held only by the hopper on the bell-crank lever. When the torpedo is fired, and after the gyroscope is set spinning, this is drawn back, leaving the gimbals perfectly free.
The releasing of the spring when the torpedo is fired, and the stopping of the segment at the proper place, is also done by the plate containing the cam. Screwed into this plate near its circumference is a small pin, which, when the spring is wound, is caught and held by a pawl. A series of levers connect this pawl with the main starting lever above the valve group. When the torpedo is fired, the motion of this lever is transmitted to the pawl, releasing the pin and allowing the segment to revolve under the influence of its powerful spring. The pinion being so small is thus given a very rapid rotation, which is the speed of the gyroscope.
Since the motion of the starting lever also starts the gyroscope, it must be remembered to make all tests of the torpedo before the Obry device is wound. If this is not done and the gyroscope is set spinning during the tests, one must wait fifteen or twenty minutes until it stops, to prevent which it should not be wound up until just before the torpedo is pushed home in the tube.
To insure the stopping of the segment at the right moment with the pinion in the blank space, and at the same time to obviate any shocks to the mechanism, the plate above-mentioned on the spindle is made cam-shaped in its periphery as well. The cam is approximately a circle placed eccentrically to the spindle, but for lightness, all except the part farthest from the center of rotation, which is the only part used, is cut away.
When the segment is released and revolves until the pinion reaches the blank space of the segment, the cam begins to take against a small rubbing piece mounted on a block of rubber, which, as it turns, brings the segment to rest by increasing pressure without concussion. At the same time the other cam has lifted the hopper on the bell-crank clear of the gimbal ring and the gyroscope is free and in motion. We have now to see how its motion relative to the torpedo is transmitted to the vertical rudders. This is done by a steering engine or servo-motor and valve which lie close to the lower pivot of the outer gimbal ring.
The valve consists of an inner cylinder which turns in a casing, one direction opening air to one side of the piston of the servomotor and exhaust to the other, and the other direction reversing this.
Thus the piston moves back and forth according as the valve is turned, and is held in either position by the pressure of the air upon it. The tiller-rod is connected directly to the piston rod, and it can be easily seen how the rudders are worked. They have no midship section. They are either hard a-port or hard a-starboard all the time, and go from one to the other as the air is opened to one side or the other of the piston; that is, as the valve is turned one way or the other. This gives the torpedo a snaky course, which may be seen plainly by the air-bubbles in its wake, but as the gyroscope acts immediately upon the torpedo's head being turned from the straight line, the curves are of small size.
This is the case in ordinary running, but it may happen that the torpedo be broadly deflected from its course either by an obstacle, or, as is usually the case, by being fired from a boat at high speed. When this happens the gyroscope makes a large angle with the axis of the torpedo, but there is no necessity for turning the valve of the servo-motor to so great an extent. In fact, the sensitiveness of the device consists in its acting when turned through small angles. A very slight motion of the valve is all that it necessary to operate the servo-motor, the air-passages of which are no larger than coarse hairs, and only enough motion to open and close these is required. Thus in the cases mentioned, if the valve and gyroscope were rigidly connected, either the valve would be turned too far, or, if checked, the gyroscope would be deflected from its plane of rotation.
To prevent this, and at the same time to allow a large range of motion for the gyroscope, the following arrangement is adopted: On the pivot of the outer gimbal ring there is clamped a short arm terminating in a small stud. On the end of the valve stem there is at the same height as this stud a pair of jaws in which the stud fits. The shape of these jaws is such that after the arm has turned the valve stem through a small angle it goes clear. The rudders have thus been put hard over and the valve is left in that position until the arm returns as the torpedo answers her helm and puts the rudders over the other way. This arrangement also corrects any little deflection the work of turning the valve may have given to the plane of rotation of the gyroscope, but when the valve is clean and in good working order it has never been found that there is any deflection.
The short arm on the pivot of the gimbal ring is not a permanent fixture there, but may be clamped at any desired angle to the gyroscope. This permits a much broader field of use for the torpedo than heretofore. By properly adjusting this arm, a torpedo may be fired abeam from a bow tube, or ahead or astern from a broadside tube. Under-water tubes are also made feasible without any extra attachments.
Suppose it were desired to fire broad on the bow from a fixed broadside tube. The Obry gear would be removed, the short arm on the gimbal ring turned through an angle of forty-five degrees and the device then replaced in the torpedo. Now when the torpedo is fired the rudders will remain hard over until the torpedo has turned through an angle of forty-five degrees, which will bring the arm within the jaws of the valve stem, and the torpedo will then begin its snaky course forty-five degrees from the direction of the tube.
To fire right ahead or astern the arm should be turned through ninety degrees.
In regard to the servo-motor it may be said that such difficulties as have arisen with the device can be traced to this cause.
It must be made so sensitive that any foreign substance that gets in is enough to interfere with its efficient working.
The air does not come to it directly from the air-flask of the torpedo, but first passes through a reducing valve, which reduces the pressure from 1350 pounds in the flask to 65 pounds at the servo-motor. This expanding and consequent cooling of the air condenses any moisture it may contain, and even a single drop of water in the fine passages may cause trouble. Attempts have been made to fit filters or strainers to the air pipes to prevent water and dust from reaching the valve, but thus far without any great amount of success. If experience proves these to be necessary, I have no doubt something of the sort will be found to give satisfaction. The air after leaving the servo-motor exhausts directly into the balance chamber, which should be opened after each run to permit the excess of pressure to escape.
This little device, weighing in all about eight pounds and occupying an exceedingly small space, will almost revolutionize torpedo warfare, giving to a weapon whose chance of a successful shot hitherto has been almost as uncertain as the throw of a die, an accuracy but little less than that of a modern great gun.
It removes from the mind of the torpedo officer all thought of the allowances to be made for speed and rolling of his own vessel, combined with the state of sea and bearing of the enemy—a calculation containing so many variables as to put it almost beyond the reach of any one man's experience—and permits him to train his tube in the same manner that a gunner trains his gun, directly at the point he wishes to hit, making allowance only for the speed of the enemy. The allowance must be greater as the speed of the torpedo is less, but the principle remains the same.
The advantages are so many that even the excessive cost of the device—about one-third that of the whole torpedo—has not prevented nearly all the great powers from adopting it; for naval officers will welcome at any cost whatever will serve to relieve their minds of some of the strain of a torpedo attack.