The following text may have been generated by Optical Character Recognition, with varying degrees of accuracy. Reader beware!

12 3 MARINE REVIEW. [July 27, ADMIRAL MELVILLE'S REPLY.* THE ENGINEER-IN-OHIEF OF THE UNITED STATES NAVY MAKES ANSWER TO ORITICISM FROM ENGLISH ENGINEERS REGARDING HIS ADVOCAOY OF THE USE OF TRIPLE SCREWS IN VESSELS OF WAR. BY REAR ADMIRAL GEORGE W, MELVILLE I note that the principal desire expressed by the member of the In- stitution in regard to this paper is for more data, I fully appreciate the position of Sir William White regarding data which he is unable to pub- lish. In fact, much of the information used by me in the preparation of this article is such as I feel unwilling to make public. I am, however, at liberty to give the results obtained from a consideration of the data, and I have done so. I realize that a considerable amount of detailed informa- tion which was omitted might have been included in the article. This omission was caused largely by my desire that the paper should not be too long for ready presentation before the Institution of Naval Architects. It was caused also by the fact that the number of trials from which the conclusions were drawn, something over 200, was considered too small upon which to base any absolutely accurate figures, although establishing fully the general law. On that account such figures as have been included in my paper have been specifically stated to be tentative. They are sub- ject to alteration as a result of future experiment. I have attempted so carefully to avoid any exaggeration of the advantages of triple screws that I think the figures enumerating the economic gains due to their use, both for high and low powers, will be found greater rather than less than ; iven in my paper. Wea Sy Paes Be ea ee to the expressed desire of the institution, I submit such data regarding the performance of triple screw ships as may i $-43000 | f 2000' | | fee ase SPEED CURVES | at Pe of ah CRUISERS OF 11000 TONS. 7 ] os BE a Fours ano Cugve iw Fure LINE TAKEN F 7 CRUISERS, THOSE IN KUT 2 os---_| = Ses 3 OH i / 4000 0° i) / J (1009 i, 0900 9000 1900 $009 INbicarkp HORSE PpwerR SS f- =< : s 7009 " 3 6009 5009 4000 ZA e g A 000 ~000 Baal EA ia 1009 Paper ; i D BP 7 (HE ROE OES. CO a a a Ge a ee 22 23 be communicated without breach of a public trust. In this connection let me remark that I regret very much that Sir William White finds him- self unable to give us the benefits of such model experiments as he may have made with triple screws. While I believe that no tests of propellers are so valuable as those obtained from the propellers themselves in posi- tion on ships where the theoretical conditions are thus minimized and where we have to deal only with full loads and with service conditions, I am sure that it would be very interesting to compare these results with those obtained from tank experiments, It appears to me that such com- parison would be a good test of the accuracy of model trials. I append a speed and power curve of certain battleships reduced to 12,500 tons displacement by Froude's well known laws of comparison, giving a separate curve to those ships fitted with triple screws, All trials from which these curves were constructed are of ships of approximately the same size, none of them being of less than 10,000 tons displacement; all belong to the same navy; they have the same character of hull and the same general under-water body. They are, in fact, almost identical in all respects, with the exception of the propelling machinery and of the slight modifications in the under-water body necessary from the introduction of a central screw in the ships so fitted. Consideration of these curves will show at once that the power required is, in all cases, high in proportion to the speed. This is due to the peculiar character of the hulls in all of *Referring to a paper from Engineer-in-Chief Melville ("The Logical Ar- rangement of the Motive Power of War Ships'--Marine Review of 'April 20, 1899.) Fete oe metas oe une Institution of Naval Architects in London, and aan t ? 3 LN AERER ot es g was presented in support of the use of triple screws ee -- i idered. It is to be noted, however, that these character- the. ships aan the cases of the triple-screw battleships as well as in those fitted with a twin-screw system. The points marked by small circles are obtained from actual trials and include the power of the auxiliary rate ta also a speed and power curve of cruisers, reduced to 11,000 tons displacement by the same laws of comparison. _ While these vessels are not so homogeneous in type as the battleships cited, it will be noted that there is with them about the same gain for the triple-screw system in the power required to attain a given speed. This difference is somewhat greater than that which I have given in my paper as the probable advan- tage to be derived from the use of triple screws. . Each point is from the actual trial of some cruiser, although they are not all from the same navy. Only a part of these trials includes the power of the auxiliary machinery, It appears to me that these curves, drawn as they are to give the fullest advantage to the system of twin-screw propulsion, develop fully jigo0 | LH.P ARRANGEMENT OF ENGINES IN CRUISER. the superior efficiency of the triple-screw system for full power trials; or, indeed, wherever all engines are in use. However, as Sir William White says, the matter of the propulsive efficiency of the propellers is but one of the desiderata to be sought in the design of a naval vessel. I submit however, that the superior propulsive efficiency of triple screws is an element of considerable advantage in their favor, although it must by no means be considered the deciding one. The deciding point in this matter seems to me to be the fact that naval vessels do at least 90 per cent of their cruising at speeds below 16 knots. In vessels of the fast type now so universally prevalent, the condensation in the low pressure cylinders is enormous when the ships are making these low speeds. Mr. Barnaby has pointed out that the greater efficiency of propul- sion incident to the use of three screws and consequent upon the utiliza- | vi / ore | | | 0.000 7 Pad ia SIPEED CURVES ool as of I / 17,000 i BATTLE-SHIPS OF 12500 TONS. bs a (NTS ANDO Ww Fyer LIne we * / % sate B w E PL = E SHIPS i / ] 14.000 7 sku ai G / 42900 Wi / 11000 7 ; ; g vi ¢ 10000 < oe 7.000 § p 7 la e 9° y. 6000 t : g a 7.000 ; 4 "a : A ee 5000 i --T Z 4.000 eo ac __}| 7009 EE K600) _| 1400 WW am "9 74 Ses Soeee inn 78 79 "0 tion of the following wake would apparently imply that a single propeller is more efficient than are twin screws. I have stated that up to a certain speed I consider that a single screw is most economical. Beyond that speed it is necessary to increase the size of this screw so greatly that the efficiency of the propeller is considerably reduced, on account of the in- creased friction and also on account of the increased difficulty of securing a free run of water to the blades. I am not sure when the critical point is passed. In the light of experience, however, it would seem that for ships having a maximum speed of as much as 15 knots the advantages due to the division of the propelling instrument into two screws are more than equivalent to the loss due to working these screws in the less advan- tageous position under the quarters of the ship. This may explain why