carried in her she has become quite a favourite vessel. I have many times, year after year, brought the matter before this Institution, and I am almost ashamed to mention it again; but I think it ought to be generally known throughout the length and breadth of the land that the bilge pieces have proved themselves to be a perfect success.

Mr. SWAN: Do you know the depth of the bilge pieces to which you have referred?

Mr. LIGGINS: I am sorry I cannot give the exact dimensions; I think they are under 3 feet.

Mr. MARTELL: I think they are 10 or 12 inches.

Mr. SWAN: Bilge keels of 10 or 12 inches are very ordinary things.

Mr. JOHN A. WELCH: Allow me to say a word upon this subject. The object of bilge pieces is not to prevent rolling, but to diminish it, and they have been successfully applied to this ship; but still there is a right and a wrong way of fitting them, and they have been variously fitted according as the ship has been in different hands. The way to look at a bilge piece being fitted is to conceive that it shall be a plane in the diagonal direction of which the trace shall be parallel to the water line you fix on as the medium draught line, and if you could conceive a door on its hinges of which the hinge portion is in the direction of the length, that would give you the exact direction to fit the bilge pieces. In that direction they will act with perfect success, but in any other direction they will have a retarding influence on the speed of the ship; so that you successfully carry out your object with respect to diminishing the roll, but at the same time you do not diminish the speed the ship may have to any great extent. A gentleman has alluded to the strength of the bilge pieces. They are generally fitted as temporary side keels, and the angle irons of the inner edge are generally tap riveted, so that in the event of their coming in contact with anything they can be removed from the side of the ship without causing any danger to the ship. When you come to any great depth-even 2 feet 3 inches—the proper thing to do would be to fit half-round iron on the outer edges, and by that means you would get a perfectly strong bilge piece.

ON THREE-THROW CRANK ENGINES OF THE COMPOUND SYSTEM, CONSTRUCTED BY RAVENHILL, EASTONS & CO., LONDON, FOR H.M.S. ROVER.*

By JOHN R. RAVENHILL, Esq., Member of Council.

[Read at the Seventeenth Session of the Institution of Naval Architects, 7th April, 1876; Vice-Admiral Sir F. W. E. NICOLSON, Bart., C.B., Vice-President, in the Chair.]

AT the Fifteenth Session of the Institution of Naval Architects, in the month of March, 1874, a Paper was read by Mr. G. B. Rennie on engines of the above description, at that time under course of construction by his firm for H.M.S. Boadicea and Bacchante, in which he brought under our notice a series of diagrams, showing the theoretical force exerted by the three cylinders at all parts of the path of the crank pin's centres in four different ways of setting the relative angles of the cranks, which he described as follows:

No. 1.-With equal angles of 120° with each other.

No. 2.-The two low-pressure cranks, with 90° between them, and 135° between these and the high-pressure crank.

No. 3.-The low-pressure cranks, with the same angle between them, but at angles of 150° and 120° with the other one.

No. 4.—The low-pressure cranks placed opposite to each other, and at right angles to the high-pressure crank.

And he proceeded to give his reasons for concluding to adopt the position described in No. 4, the angles of the cranks of the two above-named ships.†

In the discussion that followed, I stated that in the three-cylinder compound engines then making by my late firm for the Rover, I had adopted the plan of placing the cranks at equal angles of 120° with each other (see Fig. 1, Plate XVI.), and promised to furnish the Institution with the practical result obtained, which I now propose to do; but prior to entering on this subject a short general description of the engines may prove interesting.

* Received 4th April, 1876. † See Diagrams Nos. 1 to 4, Fig. 1, Plate XX., Vol. XV. Transactions for 1874.

As may be gathered from the heading of my Paper the engines have three cylinders, and they belong to the type known as the "Horizontal Return Connecting Rod," having their cylinder cross-bars working in slipper guides, and by the contract were to indicate 4,750 indicated horse-power at the measured-mile trial.

In their general design I followed, as far as circumstances would permit, our old well-known form of double-piston rod engine, such as had been made at Glass House Fields over a long series of years. Their main parts are shown in plan (see Plate XVI.)

The diameter of the high-pressure cylinder is 72 inches, representing an area of 4071.51 square inches of piston.

The diameter of the low-pressure cylinders is 88 inches, each of them having an area of 6082-13 square inches of piston; the two low-pressure pistons combined giving an aggregate area of 12164-26 square inches, the high and low-pressure pistons thus being a multiple of 1 to 2, 98 in effective area.

The high-pressure cylinder, which is fitted with a working barrel made of Sir Joseph Whitworth's patent compressed metal, is placed between the two low-pressure ones, and each cylinder stands separately and distinctly by itself, securely attached to its own two main frames, these latter six in all carrying the connecting shaft.

The slide valve of the high-pressure cylinder is placed on the upper side of it, and lies on its face. The slide face of the cylinder is cast separately, and bolted on to its place, with a view to its ready repair or renewal in case of need.

The slide valves of the low-pressure cylinders are placed on the sides of the latter, the slide faces of these cylinders being also bolted on, and all the slide valves are fitted with the usual metallic packing rings on their backs, to relieve the pressure. Steam starting gear, in addition to the ordinary hand gear, is fitted for the facility of handling them, and the easy way in which these large engines can be handled is best described by the results obtained.

Whilst running full speed ahead, they were stopped in nine seconds, went astern at full speed in six seconds, and were reversed from that position to full speed ahead again in eight seconds.

The casing of the high-pressure slide valve, to which the expansion valve casing is attached, is connected with the low-pressure slide valve casings by four copper pipes, A A A A, which are fitted as shown, through which the steam passes after leaving the high-pressure cylinder; and these pipes and the passages connected with them form the only steam reservoirs between the cylinders.

The surface condensers are on Hall's system, the steam passing through the tubes standing on the opposite side of the connecting shaft to the cylinders, and each condenser is connected to its own low-pressure cylinder by an eduction pipe, through which the steam passes on leaving the latter, and have a total number of 7,724 brass tubes tinned inside and outside, giving a total surface of 9,500 square feet. The tubes are fitted in the tube plates with screwed glands and stuffing boxes, and the condensers are so arranged that they may be worked as common condensers if required.

The air pumps, double acting, are two in number, 23 inches diameter, with a stroke of 4 feet; and the circulating water through the condensers is driven by two centrifugal pumps worked by an independent pair of small engines, and the main suction pipes to these pumps are fitted with branch suctions leading into the bilges of the vessel, so as to become large bilge pumps in the event of any leak arising in the vessel.

There is the usual arrangement of feed and bilge pumps fitted on the main engines -two of each; they are of gun metal, and are 5 inches in diameter, with a stroke of 4 feet. In addition to these pumps there are two feed donkeys, a bilge donkey hand pump and fire engine, all fitted in accordance with the Admiralty specification.

The length of stroke of the engines is 4 feet, and the diameter of the connecting shafts is 18 inches, and of the crank pins 20 inches. The shaft is made in three pieces, having couplings forged on the ends, by which they are bolted securely together.

The diameter of the propeller shafting is 16 inches, and that of the stern shaft 18 inches, the latter running in the usual lignum vitæ bearings fitted in the stern tube, and the length of the connecting rod is twice the length of the stroke.

The screw-propeller is of gun-metal on the Hirsch principle, having a diameter of 21 feet, and is driven by an ordinary cheese coupling keyed on the stern shaft, and is fitted to lift in a banjo frame by means of sheer legs and tackle on deck.

The boilers which are ten in number stand athwartship in two groups. Each group of four and six respectively has its own separate chimney fitted on the telescopic principle, and the boilers carry a working pressure of 70 pounds to the square inch. They are about 11 feet 10 inches diameter, 9 feet 6 inches long, fitted with brass tubes of 3 inches outside diameter, and wrought-iron stay tubes. Each boiler has two furnaces, each 3 feet 10 inches in diameter, and 6 feet 8 inches long. The total heating surface of the boilers is 12,700 square feet, and the grate-bar surface is 510 square feet.

These engines were the first of their class to be tried, and as my connection with Messrs. Eastons & Anderson had ceased for some time prior to the 18th November, 1875, I had not the advantage of seeing them at work during the official trial, but in the Plate

will be seen on a reduced scale copies of three sets of indicator figures taken during the above trial, which represent the average working performance of the engines; and wethis Institution as well as myself—are indebted to Mr. Wright, Engineer in Chief of the Admiralty, for supplying me with tracings of the original diagrams, and without which I could not have prepared this Paper. These diagrams also show the force exerted by the sum of the pressures of the three cylinders at all parts of the path of the crank pin's centre, set out from the pressures indicated on the diagrams, the proportion of the length of the connecting rod-viz.: four times the length of the crank-being duly taken into account.

The mean indicated horse-power exerted by the high-pressure cylinder during the six runs on the measured mile was 2476.6 horses, and by the two low-pressure cylinders combined 2486.9 horses.

The forward low-pressure cylinder giving a power of 1343-2 horses, and the after low-pressure cylinder 1143-7 horses. The mean pressure on the square inch in the three cylinders being as follows:

The theoretical action of the slide valves had been carefully gone into, and although of course this showed a difference in the two latter pressures, I was not prepared for the great difference shown in practice, and I am confident these pressures might be more nearly assimilated. But now that, for a time at all events, I have ceased to be a manufacturer of the marine steam engine, I shall watch with interest the record of the performances of the other three-throw crank engines at present being fitted on board vessels in Her Majesty's service, where different angles for the cranks will be tried to those I adopted in the engines of the Rover.

The sudden drop in the curve at a, is due in my opinion to the high-pressure slide valve closing the exhaust too quickly at one end of the cylinder, thereby causing an undue amount of compression as shown in the indicator diagrams of the high-pressure cylinder, but this could be easily improved the first time the slide is taken out of its casing.

Recent experience has demonstrated most clearly that we have closely approached, if we have not actually reached, the point at which the crank pins of large high-speed engines will work satisfactorily, in consequence of the very limited amount of bearing surface that of necessity can be allotted to them; and unless phosphor bronze should come to the assistance of the marine engineer, as white metal and lignum vitæ did in days gone by, it may be necessary to reduce the load on the high-pressure crank pins.