- BOUGHT WITH THE INCOME FROM THE SAGE ENDOWMENT FUND THE GIFT OF 3Hetirg W. Sage 1891 ■■/'■ jr'?3, ' 3\TTj<>3 ENGINEERING LIBRARY 3513-1 All books are subject to recall after two weeks. Engineering Library DATE DUE '*itc«^^^* >l¥ U»:\ .mi net PRESENT-DAY SHIPBUILDING. Ninth Edition. Fully Illustrated. Cloth. #-i 60. KNOW YOUR OWN SHIP. LONDON: CHARLES GRIFFIN & CO., LTD., Exbter Strekt, Strand. PHILADELPHIA: J. B. LIPPINCOTT COM PANY. Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://www.archive.org/details/cu31924005016450 ■h *-^y> THE SCREW PROPELLER: AND OTHER COMPETING INSTRUMENTS FOR MARINE PROPULSION. BY A. E. SEA.TON, M.Inst.C.E., M.I.Mech.E, M. Council N. A., * FORMERLY LECTURER ON MARINE ENGINEERING TO THE ROYAL NAVAL COLLEGE, GREENWICH ; AUTHOR OF "A MANUAL OP MARINE ENGINEERING," ETC. TKflttb Jfronttepiece, 6 plates, 65 otber 3-Uustrations, an& 60 arables. LONDON: CHAKLES GKIFFIN & COMPANY, LIMITED. PHILADELPHIA : J. B. LIPPINCOTT COMPANY. 1909. \ 3 /tqt %"t -•V\ii 7: A.a v:n^\ PREFACE. Some thirty-two years ago, when engaged in putting forth the Manual of Marine Engineering, I could not find a single book on the Screw Propeller, or any text book containing such information on it as would enable a draughtsman to get out the leading dimensions, and much less to make a complete design of a screw suitable for any particular ship and conditions. At that time John Bourne's admirable book on the Screw Propeller was not only out of print, but out of date, and T fear no engineer could at any time have designed a screw which would give satisfactory results from what was contained in it. It was, however, many years after that I first saw his work, and my wonder then, as now, is that such a book was allowed to disappear, seeing how much of interest it contained. But perhaps the strongest comment on the knowledge of the screw propeller at that time is in the admission of the then Engineer-in-Chief of the British Navy, that he had settled the design of those for so very important a ship as the " Iris/' intended to be the fastest ship in the Navy, by copying that of H.M.S. "Himalaya," a ship built in 1854, with the result that the British Admiralty, with all its knowledge and opportunities, with all its records of tests and trials, perpetrated a blunder never equalled in the history of steam navigation, although in the mercantile marine there had been not a few mistakes. But even in modern times, notwithstanding the better knowledge and the aid of tank experiments, our best men do sometimes fail to achieve success, but the magnitude of the failure is inconsiderable compared with that of the " Iris." In the Manual of Marine Engineering I attempted to supply the wants of designers by giving a rule or formula for each important dimension, generally based on scientific reasons, and always capable of giving results agreeable with the best and most successful practice. VI PREFACE. Moreover, such rules were generally cast in such a form as to be quickly and easily used. From time to time I have added to them and modified such of them as further knowledge had shown to require it, so that they have become generally applicable to the design of a screw for an Atlantic liner or a torpedo boat. The object of the present work is to amplify and extend what was there mostly in skeleton or in rudimentary form ; to add to it much that is of interest, although old ; and to give all that is new and of importance and necessary to be known by the students, draughts- men, sea-going engineers, designers, and others who have found the Manual of assistance to them. The more abstruse and highly mathematical investigations connected with the theory of the resistance of ships and propellers have been left to be studied in the text-books of the schools and in the valuable papers contributed by Prof. Kankine, Prof. Cotterill, Mr E. E. Froude, Dr Froude, Prof. G-reenhill, and others, to the Transactions of our professional institutions and learned societies. Chapters I. and II. are devoted to the history of propellers in such a way as to give to the reader all the good inventions, their inventors, patent numbers, etc., and at the same time to revive and perpetuate the name of many a good pioneer among engineers. The young engineer of to-day is thus enabled to see what has been done by those gone before him, and to give the credit and praise to the right man. In connection with this it is curious to note how often the important invention has emanated from someone outside the pro- fession of engineering. For example, Francis P. Smith is described as a " farmer," Bennet Woodcroft as a " printer," Goldsworthy Gurney a "physician." The only "engineer" to whom it may be accorded that he did invent and patent a screw propeller as well as invent and make a locomotive engine long before anyone else, is the Cornishman Trevithick. But it is likewise a melancholy coincidence that except Gurney the above-named made no money by these inventions, and died poor men. It is true that poor Woodcroft had a crust given him by the Government in the form of an indifferently paid post in the Patent Office. I am indebted to Mr C. De Graves Sells of Genoa for the particulars of the various groups of most interesting experiments made by his father, as also for other important information respect- PREFACE. Vll ing the early history of the screw, and I take this opportunity of thanking him for these and some other acts of kindness I have experienced at his hands of a similar nature. I also desire to tender my thanks to several other friends who have been kind enough to furnish me with information such as T was lacking, and which was necessary to my purpose, especially when engaged in determining what was really good modern practice. In conclusion, I hope this book will have accorded to it the same kindly consideration as was experienced with the Manual, and that the readers will regard with favour the attempts to elucidate difficulties and to make known discoveries, rather than to scan too closely the many shortcomings it possesses , in fact, in the words of Prior — " Be to her virtues ever kind, And to her faults a little blind." A. K SEATON. Westminster, S.W., March 1909. CONTENTS. CHAP. PAGE I. Early History of Marine Propellers .... 1 II. Modern History or Propellers ..... 13 III. Resistance of Ships ...... 38 IV. Qn Slip — Real, Apparent, Positive and Negative. Cavita- tion. Racing . 54 V. Paddle Wheels . . 69 VI. Dimensions of Paddle Wheels . . 80 VII. Hydraulic Propulsion : Internal Propellers and Jet Propellers . ..... 91 VIII. The Screw Propeller : Leading Features and Character- istics ; Thrust and Efficiency. .... 104 IX. Various Forms of Screw Propeller . . . 130 X. The Number and Position of Screws .... 140 XL Screw Propeller Blades : their Number, Shape, and Proportions . .... 153 XII. Details of Screw Propellers, and their Dimensions . 169 XIII. Geometry of the Screw . .... 186 XIV. Materials used in the Construction of the Screw Pro- peller . ...... 194 XV. Trials of the S.S. "Archimedes" and H.M.S. "Rattler" . 202 XVI. Trials of H.M.S. "Dwarf" and other Ships, made from time to time by the Admiralty .... 20^ XVII. Analysis of Mr Sell's Experiments made in 1856 with Propellers Six Inches, varying in Pitch and Surface Ratio ....... 219 XVIII. Experiments made by Isherwood and others . . . 226 INDEX 248 EKEATUM. Fig. 6, page 22, for « 1891 " read « 1841." LIST OF ILLUSTRATIONS. Frontispiece, Stern of the s.s. " Lusitania." NO. PAGE 1. Savery's Engine applied to Ship Propulsion . . 4 2. Paddle Steamer "Charlotte Dundas," 1802 . . 9 3. Machinery and "Wheels of Paddle Steamer "Comet," 1812 . 11 Plate I. Various Propellers . . . face page 14 It *"*~ It It • ), 51 18 » HI- i, „ „ „ 20 ,, IV. ,, , ,, „ 22 ))*')» >j * » » -^4 4. Francis P. Smith's ScreAV as first tried . . 18 5. Ericsson's Double Screws . . 19 6. Napier's Double Screws, 1841 . . 22 7. Stern of H.M.S. "Rattler," 1843 . .23 8. Griffiths Early Patent Screw (self-adjusting) . 25 9. Woodcroft's Adjustable Blades . . 26 10. Roberts' Patent Boss, 1851 , . .27 11. H. Hirsch's Screw Propeller of 30 degrees . . 30 12. Thorny croft's Stern for Shallow Draught Screw Ships . 35 12a. Yarrow's Drop Flap for Shallow Draught Screw Ships 37 12b. Kirk's Block Model . . 42 13. Influence of Screw on Water beyond Tip of Blades . . 55 14. Flow of Water to Paddle Float . . .56 15. Passage of a Curved Screw Blade Section through Water 62 16. Screw working with Negative Apparent Slip . . 62 17. Flow of Water from Stern to Screw . . 63 Plate VI. Feathering Paddle Wheel in Motion . . face page 72 18. Paddle Wheel of Dublin R.M.S., Inside Feathering Gear . . 73 19. Paddle Wheel of "Normandy" (no outer rim) . . 75 20. Stern Wheel Steamer, Single Wheel . 77 21. Stern Wheel of the "Endeavour": Pair of Wheels, Feathering Floats 78 22. Details of Feathering Gear of a Paddle Wheel . . 87 23. Details of Float Bearing, etc., of a Feathering Wheel . 88 24. Locus of Float Centres ...... 89 xii LIST OF ILLUSTRATIONS. NO. PAGE 25. Ruthven's Hydraulic Propeller, as in H.M.S. " Waterwitch " . 9 4 26. Thornycroft's Hydraulic Motor in a Torpedo Boat 27. Bessemer's Hydraulic Propeller . • ' 28. Screw Blade on a True Helix . . 104 29. Increasing Pitch Helix (Woodcroft's Wheel) . • 106 30. Screw Blade bent forward, Griffiths' Patent - 106 31. Screw Blade thrown back by making central line Spiral on the Bed . 107 32. Screw Blade thrown back by Coning the Bed . . 107 33. Screw Blade Curved Back . . 1° 8 34. Curves of Friction, Resistance, etc., of a Common Screw Blade 111 35. Curves of Friction, Resistance, etc., of Griffiths' Screw . .112 36. Modern High Revolution Screw compared with that of H.M.S. "Rattler," 1845 ... 113 37. Fronde's Curve of Indicated Thrust . 122 38. Propeller, on Griffiths' Patent (1860), Adjustable Blades . . 131 39. Mangin's Double Screw . 132 40. Hirsuh's Screw of 60 degrees . . .135 41. Mercantile Four-bladed Propeller . . . 136 42. Oval Blade of Equal Surface for Different Diameters 137 43. Modern Bronze Naval Screw . . 138 44. Screw outside of Rudder . . . 148 45. Auxiliary Screw outside the Rudder . . 149 46. Phipps' Lowering Screw, 1850 . ... 150 47. Bevis' Feathering Screw . . 160 48. Maudslay's Feathering Screw and Banjo Frame . 161 49. Curve of Bending Moments 174 50. Typical Screw Blade Section . 174 51. Solid Cast Iron Propeller Blade . 178 52. Various Root Sections of a Screw Blade . 179 53. Longitudinal Section of Typical Screws of Equal Area of Blade . 179 54. Method of Delineating a True Screw Accurately . . 187 55. Simple Method of Delineating a Screw . . . 189 56. Increasing Pitch Blade ....... 190 57. Pitch Measuring Instrument ..... 192 58. Bent Propeller Blade of Parson's Manganese Bronze after Stranding 196 59. Bent Propeller Blade of Stone's Bronze .... 197 60. Bent Propeller Blade, Manganese Bronze Co. .... 200 61. Blades dovetailed into Forged Boss 201 62. Screws tried on H.M.S. "Iris". . . . 216 63. Screws used by Ieherwood in his Experiments . . . 226 MAEINE PROPELLEBS. CHAPTER L EARLY HISTORY OF MARINE PROPELLERS. It is more than probable that the first instrument whereby motion was imparted to a vessel floating in deep water was the human hand used as a paddle. It would soon dawn on the mind of the intelligent navigator of such early days that there was far too little resistance on the part of the hand, either with the fingers side by side, or ex- tended to get the best results from the muscular efforts of the arms ; and although no word would be found for the phenomenon, we should say now there was too much slip from want of sufficient acting surface. To remedy this defect, art would supplement Nature, and its first step would be to fit to the hand large shells or pieces of flat wood. No doubt the next step of art would be the substitution of a shank or handle for the forearm, which had no doubt suffered from continuous immer- sion in cold water j in this way a hand paddle would be evolved, which answered the purpose of those early mariners just as it does now for the modern barbarous nations on the rivers and lakes in various parts of the world, whose boats so propelled run side by side with the steam launches and steam stern-wheelers of the white man. All such paddles so impelled impart motion to some of the water with respect to the still water through which the boat has to pass, and the reaction from this disturbed water is taken at the shoulders of the man and imparted through the body to the boat so as to produce the motion of it in the opposite direction to that taken by the disturbed water. In course of time art advanced the propeller another step by lengthening the handle so as to permit of it resting on the gunwale or edge of the boat, and of acting against a peg or notch on it, thus permitting of the muscular effort of the rower to be extended beyond 2 MARINE PROPELLERS. that of his arms ; and although the reaction from his body is then in the opposite direction from that in which the boat is intended to travel, the action at the gunwale is in excess of this, due to the leverage of the paddle or oar, so that the resultant pressure produces motion in the right direction. Propellers on this principle (viz. the paddle or oar) of various forms and dimensions are now used throughout the civilised world for all kinds of boats and barges, and in the past for many centuries were employed, especially in the Mediterranean, for the propulsion of all ships, even those of very large size, when there were two, or some- times three, rows of rowers on each side. How, or when, or by whom it was discovered that the motion could be induced by the oscillating of a single oar placed in a notch at the stern of the boat so that the blade moved obliquely to the water stream somewhat as a blade of a screw propeller does, is not known. It is certainly a very convenient form of propulsion for a single man to adopt, and may possibly have led more than one mind to the idea of the screw propeller, in much the same way as the paddle wheel was no doubt evolved from the idea of getting con- tinuous instead of reciprocating motion with paddles by mechanical means. All that is necessary for a book of this kind in tracing the early development of propellers is to show that the principle involved is in every case the same, and that, while improvements have gone on and do go on from time to time, there is no departure from the principle that the motion of every self-propelled ship is due to the projection by its propeller of a stream of water in a direction opposite to that in which it is intended the ship shall move. The only exception that can be cited is that of the ferry boats and river craft which are moved by means of a chain or rope submerged and operated upon by a winch arrangement in the ship itself. There is every reason to believe that paddle wheels were used as a means of propulsion in quite early times, even as far back as the time of the Consul App. Claudius, who obtained the cognomen of " Caudex" because he employed boats propelled by paddle wheels to transport his troops into Sicily (B.C. 264). The Chinese as well as the ancient Egyptians are more than suspected of having a knowledge of the use of a paddle wheel for ship propulsion. In a.d. 1472 there is the evidence of E. Valturius that paddle EARLY HISTORY OP MARINE PROPELLERS. 3 wheels were in use instead of oars, inasmuch as he shows a view of two galleys having wheels (five pairs) on each side, each pair running on one common axle with a crank in the middle and the cranks connected together. In a.d. 1543 it is stated that one Blasco de Garay, a Spaniard resident at Barcelona, propelled a vessel by an engine " consisting of a large caldron or vessel of boiling water and a moveable wheel attached to each side of the ship." This is the first record of a steam ship, and it is possible that the engine may have been on Hero's principle, and therefore of the nature of a turbine. In a.d. 1578 it is related by one W. Bourne that " you may make a boat to go without oars or sail by the placing of certain wheels on the outside of the boats in that sort that the arms of the wheels may go into the water and so turning the wheels by some provision, the wheels shall make the boat go." In a.d. 1597 Eoger Bacon, writing, said that " We have seen and used in London, a warlike machine driven by internal machinery, either on land or water." " Succeeding years have shown us a vessel which being almost wholly submerged would run through the water against waves and wind, with a speed greater than that attained by the fastest London pinnaces." In a.d. 1663 the Marquis of Worcester claims, among other matters, to have invented a boat with paddle wheels on an axle across it, which axle is turned by the action of the stream on the paddles. In A.D. 1681 the celebrated philosopher Robert Hook describes certain windmills which, as stated by Mr Bourne, u had all the main features of the screw propeller and feathering wheel." In a.d. 1682 it is related that a " tow vessel " was used at Chatham Dockyard for moving the ships in the river by means of paddle wheels on an axle turned in some way by a horse, probably by means of a gin. In a.d. 1690 D. Papin described his steam cylinder containing a piston which was forced downwards by the pressure of the atmos- phere when the steam below it had been condensed, and stated that it might be applied, among other things, to the propulsion of ships by paddle wheels, such as he had seen made in London for Prince Rupert, to be turned by horses. He proposed to use two or three of his cylinders with a rack arrangement. In a.d. 1693 M. Du Quet tried as a propeller four oar- 4 MARINE PROPELLERS. shaped blades set in a wheel so that they could be moved through 90 degrees, and so that each one after propelling could be made to feather and return edgeways. In a.d. 1698 Thomas Savery, in his patent No. 356, relating to the raising of water and occasioning motion to all kinds of mills by the impelling force of the steam engine, states : " I believe it may be very useful to ships, but I dare not meddle with that matter, and leave it to the judgment of those who are the best judges of maritime affairs." And again : " As for fixing the engines in ships when they may be thought probably useful, I question not but we may find conveniences for fixing them." Fig. 1.— Savery's Engine applied to Ship Propulsion. In a.d. 1707 it is stated in letters addressed to Leibnitz that Dennis Papin had used one of Savery's engines to propel a ship or boat on the river Fulda. In a.d. 1721 experiments were made in France by M. Du Quet with paddle wheels of a sort, as much as 18 feet in diameter, three of which were fitted on each side of a galley and operated by as many as two hundred men. In a.d. 1729 John Allen took out a patent claiming, among other things, to navigate a ship in a calm. "My method will be effected by forcing water or some other fluid through the stern or hinder part of a ship, at a convenient distance under the surface of the water, into the sea, by proper engines placed within the ship. Amongst the several and peculiar engines I have invented for this purpose is one of a very extraordinary nature, whose operation is owing to the explosion of gunpowder, I having found out a method of firing gunpowder in clauso, or in a confined place, whereby I can EARLY HISTORY OP MARINE PROPELLERS. 5 apply the whole force of it, which is inconceivably great, so as to communicate motion to a great variety of engines ; which may also be applied for the draining of mines and other purposes." This is most interesting as being the first instance of a pro- posal to propel a ship by jet or stream of water, and further, that the power was to be by means of what is known as an internal combustion engine. In the same year, a.d. 1729, Du Quet, who seems to have been very interested in marine propulsion, called attention to a method for towing vessels by submerging a screw or helical frame having eight vanes fixed to the end of as many spokes and inclined at an angle of 54 degrees. The screw was caused to revolve by the stream of water, and by gearing to impart motion to a windlass around which the tow rope was wound. In a.d. 1730 Mr Allen makes a further suggestion that application of the fire-engine should be made to work pistons for propelling a vessel by forcing out air or water under the surface of the internal water. In a.d. 1734 Duviver gives a description of a ship having paddles fixed to a frame, very much like that which is now known as the lazy-tongs, on each side and worked by the oscillating of a heavy pendulum. In a.d. 1736 Jonathan Hulls took out a patent which is interesting, inasmuch as it is claimed for him by his descendants that he was the first to produce an actual steam-propelled ship. His own claim was for a machine " for carrying ships and vessels out of or into a river or harbour against wind and tide in a calm." In his practice, it appears from such records as there are that his vessel was what is now known as a " stern- wheeler,'' the paddle wheel being carried in a framework at the stern of the vessel. The circular motion of the wheel was given by means of ratchet wheels operated on by ropes, one of which was connected with the piston of a condensing engine and the other to a falling weight, the weight being raised by the excess of the power of the steam piston. In a.d. 1738 D. Bernouilli proposed to propel ships by forcing water through orifices so as to make jets to flow towards the stern, somewhat as Aljen had suggested in 1729, and as shown in fig. 1. In a.d. 1752 D. Bernouilli described a method of propelling vessels by wheels with vanes " set at an angle of sixty degrees, both with the arbor and keel of the vessel, to which the arbor is placed 6 MARINE PROPELLERS. parallel. To sustain this arbor and the wheels, two strong bars of iron, of between two and three inches thick, proceed from the side of the vessel at right angles to it, about two feet and a half below the surface of the water." From this it would appear that practically a submerged screw propeller was intended. It is also noticeable in a.d. 1746, Bouger speaks of revolving arms like the vanes of a windmill as having been tried for the propulsion of ships. In a.d. 1760 J. A. Genevois proposed to employ a propeller having hinged blades so as to open and shut like the back of a book, which would be full open on making the effective stroke, and closed again on the return so as to offer little obstruction. No doubt this idea was suggested by observing the foot of a duck or other water bird. This was followed years afterwards by a whole series of patents, none of which were successful, and all proved a source of extreme disappointment to their promoters. It is interesting to note in passing that this man was the first to suggest the use of water distilled from the sea for drinking and other purposes on board ship. In a.d. 1770 James Watt made a suggestion for a screw pro- peller turned by one of his engines, but he did nothing further ; in fact, he was much opposed to the use of his steam engines on board a ship. In A.D. 1776 John Barber claimed, among other things, in his patent No. 1118, to apply the turbine on Branca's type to the pro- pelling of ships, and provided for reversing the motion by making the nozzles reversible. In a.d. 1778 a Jesuit missionary described a paddle-wheel boat he had seen at Pekin, China, as being 42 feet long by 13 feet broad, propelled by a paddle wheel on each side having flat arms whose ends dipped in the water to the extent of a foot ; it was moved round by men. In a.d. 1780 James Pickard of Birmingham took out his patent for a crank as a means of transforming the lineal motion of a steam engine piston into circular motion as in the turning of wheels. In the same year Jouffroy used an engine with two cylinders to propel a boat having duck-foot-shaped propellers ; a year later he tried a steamboat on the Rhone of a similar kind, but it was worked with paddle wheels. In a.d. 1783 Jouffroy is said to have built a steamship 140 feet long, having a pair of wheels on one shaft moved by a single cylinder engine. In a.d. 1785 Joseph Bramah, in his patent No. 1478, claimed to t/ EARLY HISTORY OF MARINE PROPELLERS. J fit a wheel with inclined fans or wings, similar to the fly of a smoke jack or the vertical sails of a windmill ; from this and other descrip- tions in the patent there is no doubt that Bramah had the complete idea of the screw propeller working at the end of a shaft projecting from the stern of a vessel and wholly submerged. ( Vide fig. 1, Plate I.) In A.D. 1786 John Fitch of Philadelphia, U.S.A., had built by Messrs Brookes & Wilson a boat 45 feet long and 12 feet beam, fitted with a steam engine having a cylinder 12 inches in diameter and 3 feet stroke, running at forty revolutions, operating on twelve oars or paddles so arranged that half of them were acting when the other half were out of action. It was tried in August 1788, but failed to make a higher speed than three miles per hour. In a.d. 1787 Pitch had another boat built, 60 feet long, 8 feet beam and 4 feet deep, and fitted with the machinery of the above described boat, but a better system of paddles was fitted, they being so arranged as to have the same motion imparted to each of them as to a paddle worked by hand. The speed of this ship was somewhat better than the former, but still low. In a.d. 1787 William Symington patented a means of obtaining rotary motion. In the same year Patrick Miller of Dalswinton, KB., published his pamphlet on propelling ships by paddle wheels turned by men ; it gave also the results of his experiments. In a.d. 1788 Robert Fourness and James Ashworth took out their patent for raising the paddle wheels out of the water when not required. a.d. 1788 is chiefly famous, however, for the trial of the first practicable steamship of which we have a definite record and accurate particulars. William Symington in that year was introduced to Patrick Miller's notice by Mr James Taylor, and his steam engine, which could be worked without an air pump, was chosen as suitable for the purpose of propelling Mr Miller's double-hulled pleasure boat, 25 feet long by 7 feet beam, which he had built at Dalswinton the preceding winter, by means of a paddle wheel working between the hulls. The engine was not much more than a toy, as the diameter of its two cylinders was only 4 inches, and it was mounted on a portable frame on the deck. It was, however, sufficiently powerful to move the ship at the rate of five miles an hour. In this pioneer work honour is due to Symington, the engineer who designed and superintended the construction of the machinery; to Patrick Miller, who found the capital and built the ship ; ,and lastly, to James Taylor, who brought O MARINE PROPELLERS. them together and acted the part of "amicus curiae" throughout. Strange to say, he is the only one who had a reward for .this valuable experiment, for his wife enjoyed a Government pension of £50 per annum. Patrick Miller having spent a fortune on it and other experi- ments, let the question of marine propulsion drop. In the same year James Rumsey proposed to propel ships by drawing water through an orifice near the bow into an ordinary vertical pump and expelling it by a tube through the stern ; but his ideas on propulsion were vague as well as numerous. He succeeded, however, in getting a wealthy American residing in London to pay for the construction of such a boat, and in 1793 she was tried on the Thames and attained a speed of only four knots, for although the pump was 24 inches diameter, the discharge orifice was no more than 6 inches square. In a.d. 1789 Fitch had another boat built, named the " Thornton," fitted with more powerful machinery than supplied to the previous ones, so that the speed attained by it was at the rate of eight miles an hour. In a.d. 1790 Fitch applied for a patent for forcing air and water by means of steam, through trunks, which was therefore a mode of hydraulic propulsion. It is also said that about 1791 he applied for a patent for paddles, both as side wheels and stern wheels. He appears to have been ingenious, persevering, and a good mechanic, but a dissolute man. In 1798 he committed suicide. In A.D. 1792 James Rumsay took out another patent, No. 1903, in which he claims a centrifugal pump, etc. ; he proposed to place a screw in a frame between the hulls of two boats connected together "whose axis being moved by horses, by steam, or by men, or any other power applied to the cog wheel, when the boat will take motion." The axle and screw may be used " before or behind a single boat or in the bottom." In a.d. 1793 a paddle boat worked by steam was actually run by one John Smith on the Bridgewater Canal from Manchester to Runcorn. The engine was one of Newcomen type working with a beam, a connecting rod and crank ; each paddle wheel had seven arms or floats. The speed was, however, only two miles an hour. In a.d. 1794 William Lyttleton patented (No. 2000) an arrange- ment of three helical strips or threads projecting out of a cylinder hung in a frame and submerged either at the bow or stern of a ship. This propeller, which was of course a screw, was rotated by an end- EARLY HISTORY OF MARINE PROPELLERS. 9 less rope from on deck. Here, again, when tried, the speed obtained by it was disappointing, being only two miles per hour. (See fig. 2, Plate I.) In a.d. 1798 Robert Fulton stated he had, in this year, tried a four-bladed screw as a propeller on a boat. In a.d. 1800 Edward Shorter patented, as a means of propelling ships, the fitting of a perpetual sculling machine to the stern of a ship, consisting of a two-bladed screw at the end of a revolving shaft set at an angle like an oar in sculling, and having a universal joint connection with a horizontal shaft on the deck, so that the screw could be raised and lowered to suit the trim of the ship. The screw shaft end was supported from a float and steadied in place by guy ropes. In a.d. 1801 William Symington patented the fitting of a con- necting rod from the piston rod end to the crank pin of a paddle Fig. 2.— Paddle Steamer "Charlotte Dundas," 1802. shaft, as seen always in stern-wheelers to-day. He applied this to the " Charlotte Dundas " in 1802. In a.d. 1802 Snorter's screw arrangement was tried on board H.M.S. <( Dragon" and " Superb," and to the transport " Doncaster," which latter ship attained a speed of 1J miles per hour when deeply laden, with eight men only at the capstan working it. In A.D. 1802 William Symington constructed the tow boat i£ Charlotte Dundas " for Lord Dundas of Kerse, N.B. This may be said to be the first steamship to be used for practical purposes as well as experiment ; and but for the fears of damage to their canal by the proprietors of it, this ship might have been regularly employed, and the general use of other steamships would probably have followed at once. As it was, this little ship, after she had demonstrated her power by towing two barges, each of 70 tons burden, 19J miles in six hours against a strong wind, was laid up and made no use of. IO MARINE PROPELLERS. The " Charlotte Dundas " was what we now call a stern-wheeler, having one paddle wheel at the stern turned by a horizontal double- acting engine having one cylinder 22 inches in diameter and 4 feet stroke, directly connected by a rod as before described ; in fact, quite like a present-day engine for the same purpose. She seems to have been about 44 feet long over all. In a.d. 1804 John Stevens in America made and tried a steamboat having a screw propeller on Bramah's plan worked by a rotatory engine. The latter, however, was not a success, and was replaced by a Watt engine, when the speed attained was four miles an hour. The enterprise, however, proved a failure, owing to boiler troubles consequent on its being tubular and novel in design and too small for the engine. In A.D. 1807 Kobert Fulton of New York produced his famous steamer the "Clermont," of 160 tons burden, 130 feet long, 16J feet beam, and 7 feet deep. She was propelled by a pair of side wheels 15 feet diameter, having floats 4 feet long dipping 2 feet into the water, operated by an engine made by Bolton & Watt of Birmingham, England, having one cylinder 24 inches diameter and 4 feet stroke supplied with steam by a boiler 20 feet long, 8 feet broad, and 7 feet high. On trial her speed over a run of 110 miles was at the rate of 4*6 miles per hour ; the following day she did forty miles in five hours. The following year she was lengthened to 140 feet of keel, and then attained a mean speed of quite live miles per hour. Mr Fulton built other equally successful steamers, the largest being the " Paragon/' of 331 tons burden and as much as 173 feet long. This was in 1811, when the "Comet" was being projected by H. Bell in Scotland. Very many years elapsed before a steamer of this length was built in Great Britain. Fulton, like Symington, Taylor, Bell, and so many other pioneer engineers, ended his days, in 1815, in penury. In a.d. 1811 Henry Bell of Helensburgh, N.B., had built by John Wood & Co., Port-Glasgow, a little vessel of 30 tons burden, 40 feet long and 10 J feet beam, named the " Comet." He fitted her with a side lever engine having a single cylinder 11 inches diameter and 16 inches stroke, whose crankshaft was geared to two shafts, one before and one abaft the engine ; each had a paddle wheel at each end, and was therefore what a locomotive engineer would call " a two- pair coupled " job. She attained a speed of five knots, and traded between Greenock and Glasgow ; after being lengthened 20 feet and fitted with one pair of complete wheels and a new cylinder 12£ inches EARLY HISTORY OF MARINE PROPELLERS. I I j3 5= [2 MARINE PROPELLERS. diameter her speed seems to have been 7*8 miles per hour or 6 J knots. She was wrecked at Crinan, KB., in 1820 in the tide race. In a.d. 1815 the steamship "Thames" performed the voyage from Glasgow, where apparently she was built, to London, success- fully. She called at Dublin on the road, and had Dr Dodd as a passenger. The construction of steamers for commercial purposes now became general in various parts of the kingdom, even at those remote from the sea ; for example, the paddle steamer " Britannia," of 50 tons, 65 feet long and 13 feet beam, was built at Gainsborough, Lincoln- shire, in 1816, and sold to Portsmouth the following year. In 1817, at the same place, another builder produced the " Prince of Coburg," of 71 tons, 76-5 feet long, 14*4 feet beam, with engines by Aaron Manby of Staffordshire, and sent her to the Solent. Three years later Richard Pearson built at Thorne, near Doncaster, the paddle steamer "Kingston," 1201- tons, 106 feet long and 20 feet beam, followed by the " Yorkshire man *' a year later, of 164| tons, 120 feet long, 21 feet beam. The latter had geared engines made by the Butterly Company, Derbyshire, wherein the engine shaft ran three times as many revolutions as the paddle shaft. In A.D. 1822 Marc Isambard Brunei took out a patent for two inclined cylinder engines, the cranks at right angles, the piston rods fitted with roller guides, and the weight of the piston relieved by " spring supporters " on the extremities of the " head beam," the engine to be governed by means of a stream of water pumped through an orifice. The condenser was to be formed of an assemblage of pipes which collectively formed a spacious chamber. They were to be connected together with a set of smaller pipes, and the whole placed in an iron reservoir, thus forming a surface condenser. It was not till 1822 that the Admiralty indulged in a steamship of their own, when Oliver Lang, the master shipwright of Deptford, built for the navy a small tug or tender, which, strange to say, their Lordships named the u Comet," as another mysterious visitor had crossed the heavens since the one from which Bell named his little pioneer ten years before. Ten years afterwards their Lordships ventured to build the " Salamander " paddle steamer, and fit her with guns as a warship. CHAPTEE II. MODERN HISTORY OP PROPELLERS. Many years were to elapse from the time of the trials of Snorter's screw in H.M.S. " Dragon " before the screw was again put to a real practical test as a propeller ; to the paddle wheel, having been proved successful for steam navigation both in smooth water and rough seas, men's minds and money were directed rather than to the out-of-sight screw, as is shown by the huge number of patents taken out for new and sometimes improved wheels or accessories. The most important of them was the Morgan wheel, applied for by Elijah Galloway and sealed in July 1829 (No. 5805), which after running the full fourteen years, was renewed for a further five years, and consequently held good until 1848. The object of the inventor was to so manipulate the floats that at entry and emersion they were at the angle or inclination they would have occupied on a radial wheel of much larger diameter. This will be described in detail in another chapter ; suffice it to say here that Hook had anticipated the idea so far back as 1683. A.D. 1815 — Kichard Trevithick, the ingenious Cornishman, sug- gested the screw as a fit instrument for marine propulsion in connec- tion with his high-pressure engines, and the form he chose was that of leaves or blades placed obliquely on a cylindrical axle, and in some cases he would have it to revolve in a fixed cylinder ; in others, the cylinder revolved with it ; but generally the screw revolved without any cylinder surrounding it. " It may revolve at the head or the stern of the vessel; or one or more such worms may work on each side of the vessel," he said. A.D. 1816. — John Millington of Hammersmith took out a patent for the use of a screw as a good method of propelling vessels having " two vanes, each extending to about a quadrant of a circle, as they produce a greater effect than any other number." He claims " to fit 13 T4 MARINE PROPELLERS. a propeller either at the head or stern of vessels, or at both of them at the same time." Millington's propeller shaft was set at an angle so that its inner end was above water, and attached to a horizontal shaft going inboard by means of a Hook's universal joint. The outer end was suspended from a spar on the ship's end, and arranged to be raised and lowered to suit the immersion. He also proposed to fit guy ropes on either side so that the spar might have motion horizon- tally, and so cause the propeller to steer the ship. (See fig. 3, Plate I.) a.d. 1816. — In this year William Church of Birmingham took out a patent for a propeller consisting of two wheels revolving in opposite directions like Perkins'. Church, however, proposed a number of bent paddles placed upon cylindrical rings, set so as to work in opposite directions, and they might be placed within a fixed cylinder. (See fig. 6, Plate I.) a.d. 1816. — Richard Wright took out his patent No. 4088, in which he claims, among other things, to make a feathering paddle wheel with a somewhat complicated apparatus, which he describes, and claims that it will cause the floats to enter the water vertically, which, of course, is not what is best with the ship when under way. He also claimed to fit a two-cylinder two crank (at right angles) compound engine with a receiver between the cylinders. a.d. 1817. — Joseph Claude Niepce claims, in his patent No. 4179, to propel a ship by expelling water alternately from two reservoirs in a stream through the stern of the ship, and to use as his expelling force "expanded air produced by the combustible matter in a receiver which escapes through apertures closed with valves, etc. , which will be like the force of steam by pressing upon the surface of the water enclosed in the receivers." In other words, an internal combustion apparatus used direct as a means of propulsion ; he proposed using volatile oils for the purpose. a.d. 1818. — A regular line of steamers was established by Mr David Napier for service between Glasgow and Belfast, and Mr Dawson established a similar service of steamers on the Thames from London to Gravesend. a.d. 1819.— The first steamship, the " Savannah/' 350 tons, fitted with lifting paddle wheels, crossed the Atlantic from New York to Liverpool, but the voyage was done mostly under sail; the engine being used only eighteen days out of the thirty-five taken in crossing • it consisted of one cylinder 40 inches diameter and 72 inches stroke. MODERN HISTORY OP PROPELLERS. I 5 a.d. 1822.— H.M.S. " Comet/' paddle steamship, built at Deptford for the Admiralty. a.d. 1824. — Jacob Perkins patented No. 4998, an arrangement of screw propeller by which two blades were fixed on the end of a hollow shaft through which a second shaft passed, also having a pair of blades on its end, but set to the opposite " hand " of the former, so that when revolving in different directions they acted together in project- ing a stream of water. (Fig. 4, Plate I.) He is also the first to prescribe a varying pitch, as he proposed to make the blades with an inclination of 45 degrees at the boss and 22J degrees at the tips. Since his day this idea of a right-handed and left-handed screw combination has been patented many times, notably by Ericsson in 1836, George Smith in 1838, and others until 1853, when a similar application from John Pym was refused protection by the Patent Office. It is recorded that in this year John Swan tried double-bladed screws, one on each side of the ship, and fully immersed. This would appear to be the first attempt at twin screw propulsion. a.d. 1825.— Samuel Brown patented (No. 5126) the idea of pro- pelling such a ship as a ferry boat by a chain secured at the ends lying in the water and passing round a wheel on the ship, which is turned by machinery. In the same year Samuel Brown gained a reward of a hundred guineas for the best suggestion for propelling ships without paddle wheels, by proposing a ship with a screw at the bow. Such a ship was built and tried on the Thames, attaining a speed of six to seven miles per hour ; the screw, however, was rotated by a Brown " gas vacuum " engine, which gave trouble to such an extent as to discredit the whole undertaking and bring about the bankruptcy of the Company. This failure was, no doubt, also the means of postponing the adoption of the screw as a marine propeller for many years. In A.D. 1825 also was witnessed the successful application of steam to ocean navigation, the steamer " Enterprise ' J having made the voyage from England to Calcutta in 113 days. This little ship was only 470 tons burden, 122 feet long, and 27 feet beam. a.d. 1827. — William Hale patented No. 5594, an arrangement of screw having one or more threads turning on its axis in a vertical cylinder, drawing water through the bottom of a ship and expelling I 6 MARINE PROPELLERS. it through the stern to obtain motion. He says he found " two threads in the screw are better than one.'' A.D. 1828.— Charles Cummerow, in his patent No. 5730, manifests a grasp of the practical problems involved in screw propulsion. He prescribed a screw of a single convolution with a pitch equal to half the diameter ; he proposed to tit it in the deadwood of the ship and to support it at the outer end of the after-stern post on which the rudder is to hang ; the shaft is to pass inboard through a tube and stuffing-box, and finally proposed to gear it to the engine shaft so that the screw should turn three times per second — that is, 180 revolutions per minute. (See tig. 5, Plate I.) a.d. 1829. — Archibald Eobertson patented, No. 5749, the idea of fixing the floats of a paddle wheel obliquely " at an angle of from 40 to 70 degrees to the plane of the wheel's motion " and parallel to each other. Also to place the paddle shafts obliquely. Jacob Perkins in the same year proposed, in his patent No. 5806, to place the floats of a paddle wheel at an angle of 45 degrees to the plane motion, and the shafts at the same angle, so that when working the immersed floats would be perpendicular to the ship's keel. In this year Elijah Galloway patented the well-known feathering wheel generally called the Morgan. His method of feathering remains in use to-day, while all others have disappeared. A full description of it is given in Chapter IV. In the same year Julius Pumphrey claimed, in patent No. 5765, " to use two spirals at the stern, one on each side of the rudder, in the direction of the ship's length, wholly under water, so that the rudder may act freely between the shafts of the spirals ; and cutting the shafts off opposite the stern posts or hangings of the rudder, to connect them again by Hook's universal joint, and carry the spirals with a strong frame which is hung to turn in concert with the rudder. The spirals are to have a rotatory motion given them by a steam engine, the shafts of the spirals passing through a water-tight packing into the vessel." a.d. 1830. — William Church also claims (No. 6041) the applica- tion of oscillating cylinders with hollow trunnions to driving paddle wheel shafts and to work expansively by means of a throttle valve operated by means of tappets. Joseph Maudslay, however, in 1827 had also patented an oscillating cylinder engine. Goldsworthy Gurney, in the same year, used oscillating cylinders to drive his steam coach. Plate I. William Lyttleton, 1794. Jacob Perkins, 1824. Charles Cummerow, 1828. William| Church, 1829. Bennei Woodcroft, 1832. § vl Francis Pettit Smith, 1836. Plate III. George Rennie, 1839. Captain Carpenter, 1840. Miles Berry, 1840. (J George Blaxland, 1840. William Joest, 1841. Benjamin Biram, 1842. 21 Earl of Dundonald, 1843. Thomas Sunderland, 1843. MODERN HISTORY OF PROPELLERS. 2 1 To turn the vessel a valve causes the water to be expelled at one side in direction of the bows. The Admiralty built H.M.S. " Waterwitch " in 1866 to try Eutbven's invention. The results as to speed were very disappointing. A.D. 1840. — Captain Edward J. Carpenter proposed a patent No. 8545 (see fig. 18, Plate III.) to fit twin screws with shafts having Hook's universal joint to permit of the screw being lifted out of water when under sail, and the dividing of the rudder into two parts con- nected by a hasp forging when a single screw is used projecting beyond the rudder. George Blaxland in this year patented the idea of making a screw blade in a series of strips set at different angles so as to constitute steps. (See fig. 20, Plate III.) In this year the ''Archimedes," after competing with the Cross- Channel Dover steamer, made a voyage round Great Britain, calling at all principal ports. (Vide Chapter XV.) a.d. 1841. — David Napier, in patent No. 8893, proposed two paddle wheels with oblique floats or blades like a screw, having their shafts parallel to one another in a fore and aft direction, but above water at the stern, so that the arms were about half immersed. They turned in opposite directions, and had their centres so close that one half masked the other. They were, in fact, overlapping twin screws with their shaft centres a quarter their diameter above water, and so behaved much as a twin screw bluff-sterned ship would to-day under similar circumstances. The screws were in this case, however, in a very bad position for getting a good water supply, and consequently the experiment failed badly. (See fig. 6.) A.r>. 1843. — John Laird took out patent No. 9830. The screw shaft is surrounded by a watertight trunk, "by which arrangement, cargo may be stowed around and about the trunk, and the steam engine and machinery may be amidships." The screw shaft works through the trunk watertight at both ends. A.D. 1843. — James Hamer proposed, in patent No. 9592, to drive a screw propeller by means of a turbine on Brancas principle and fit it to a shaft connected to the driving shaft by a Hook's joint. Joseph Maudslay in the same year (No. 9833) suggested fitting a rudder on each side of a single screw projecting downwards from the quarter without external bearings and having their tillers con- nected, very much as done in later days by Sir John Thornycroft. H.M.S. " Rattler," the first screw steamer in H. M. Navy, was 22 MARINE PROPELLERS. completed this year ; she was 176*5 feet long, 32*7 feet beam, and 135 feet draught of water, 1140 tons displacement, propelled by a screw 10 feet in diameter and 11 feet pitch, driven by Maudslay's twin-cylinder engines of 200 KELP, with four cylinders 40 inches diameter and 4 feet stroke. Her speed was 10 knots. (See fig. 7, also Chapter XV.) a.d. 1844.— Bennet Woodcroft claims to fit a propeller with blades capable of turning round on the axis of their shanks and moved by means of a sliding sleeve within the boss, having on its surface a spiral groove into which the pinion, the arms, or levers of the shanks are fitted, and free to slide and have an angular movement as the sleeve slides axially on the shaft. Fhj, 6.— Napier's Double Screws, 1891. a.d. 1845. — The screw steamer "Great Britain," 3270 tons, crossed the Atlantic. She was 322 feet long, 48 feet beam, and 3T5 feet deep, and had engines of 1500 I.H.P. geared to the screw shaft. The four cylinders were 80 inches diameter and 72 inches stroke. This was the first screw steamer and first iron ship placed on this station. She was designed as a paddle-wheel ship, but altered when partly built to a screw by Brunei. She did not, however, remain long on the New York and Liverpool station. A.D. 1846. — Joseph Maudslay proposed to fit screws that they might be lifted on deck through an aperture. The shaft end is conical, and fits into a conical recess in the boss and tightened up by end pressure. Plate IV. Frederick Rosenburg, 1845. 23 Captain George Beadon, 1845. John Samuel Templeton, 1846. .John Buchanan, 1846. Conrad H. Greenbow, 1847. John Macintosh, 1847. Gardiner Stow, 1848. MODERN HISTORY OF PROPELLERS. 23 ad 1846 -Henry Bessemer claimed (No. 11,352) to use the exhaust steam from a reciprocating engine to turn a hollow shaft, through which it is caused to pass and discharge through two axes having tangential openings; these axes may revolve insrde the condenser. This undoubtedly is the rudiment of the low-pressme -Stem of H. M.S. "Rattler, turbine- and it is singular that Mr Bessemer with his scientific know dg did not appreciate the true value of his invents, or rath" the adaptation of Hero's reaction engme-the first form of steam engine-when worked in conjunction with a condenser. Td 1847 -John Macintosh (No. 11,763) proposed to make the blades of a screw (see fig. 31, Plate IV.) so that they could be turned 24 MARINE PROPELLERS. sufficiently to act either " ahead " or " astern." Such screws are now often used in small craft propelled by oil and gas engines. Johann Gottlob Seyrig in this year claimed (No. 11,695) "the use of a turbine or centrifugal wheel and a screw propeller in such conjunction that the currents produced shall simultaneously help both machines," much as Sir J. F. Thomycroft did with the " Lightning " and other ships in 1882. a.d. 1848. — Joseph Maudslay claimed (No. 12,088) to make screws with the shanks of opposite blades overlapping in the boss and geared together by sectors, so that as the shaft rotates the blades turn into position, and when it stops the water places them fore and aft. a.d. 1849. — Wakefield Pirn proposed (No. 12,440) to fit a screw at the bow as well as stern to work in conjunction. John Dugdale and Edward Birch claimed (No. 12,625) to fit one or more propellers on one shaft, each being larger as it is nearer the stern. John Euthven took out in this year his celebrated patent (No. 12,739) in which he claims to use a centrifugal wheel with curved blades supplied with water admitted through apertures below the ship, and flowing through passages and forced through pipes terminating in nozzles outside the vessel (see fig. 25), the nozzles to be jointed so as to be turned in any direction and the motion of the vessel regulated and directed without stopping the engines. Robert Griffiths, also in this year, claimed to fit his propellers so that each blade may turn on its axis in a socket by the pressure of the water on the leading side of it, and thus act against a spring (see fig. 8). He expected in this way that the pitch w T ould be increased as the velocity increased, and vice versd, a most valuable feature in those days of auxiliary steamers, as it enabled the engines to be used advantageously when the ship was under sail. Otherwise by a passing squall or other sudden quickening of the ship the engines were liable to " race " to a dangerous extent. He also specified some distinct forms of blade, one of which is shown fig. 37, Plate V., and claimed to make the blades with their outer ends curved towards the bow, as he made them all eventually. But he also proposed to bend two sternwards and two forwards, as if uncertain as to which was the best direction. George Callaway and R. A. Purkiss took out a patent this year Platk V. Moses Poole, 18-48. Joshua Beale, 1848. Hick Gaitiix, 1849. C~H 11 Alexander Campbell, 1849. 'l- ^ ^ ™* m - l rA--..^y Robert Griffiths, 1849. Buckwell and Apsey, 1849. m Florid Heindryckx, 1850. — Oj §MuJ^ Gaspard Malo, 1850. Captain George Beadon, 1851. Ethan Baldwin, 1850. MODERN HISTORY OF PROPELLERS. 25 on lines almost identical with those of Kuthven, but three months after him. a.d. 1850. — Henry Wimshurst proposed, among other things, in his patent No. 13,340, an apparatus for measuring the power exerted m turning the shaft of a screw propeller which would be applicable now to propellers driven by turbines, and was the forerunner of the Fig. 8.— Griffiths' Early Patent Screw (Self-adjusting). Torque instruments at present used to determine the power trans- mitted by shafts. A.D. 1850. — The screw steamer " City of Glasgow/' 1600 tons, 237 feet long, 34 feet beam, with engines of 350 JNT.H.P., built of iron by Tod & M'Gregor, Glasgow, was the first of the famous Inman line of steamers. It was placed on the New York station to run against the paddle steamers of the Cunard Company in this year. a.d. 1851. — Bennet Woodcroft claimed (No. 13,476) to set the 26 MARINE PROPELLERS. blades of a propeller at any angle by means of toothed wheels and worms, as shown on fig. 9, by a sliding sleeve with spiral groove, as described in his 1844 patent ; and further, to move the blades, even to the extent of reversing them, by controlling the apparatus by means of rods fitted in grooves in the shaft and thus carried inboard and worked in the tunnel or engine-room. Griistav A. Buckholz proposed (No. 13,515) in this year to fit ships with three screw propellers, the middle one being nearer to the stern than the others, all geared together, however, so as to be worked by Fig. 9. — Woodcroft's Adjustable Blades. one engine. In 1855 H.M.S. "Meteor" was fitted in this way, but without success. a.d. 1851. — Edward J. Carpenter proposed to form twin screw ships with two submerged after bodies, each with aperture, rudder, etc. H.M.S. "Penelope" was built on this plan in 1867, as were also H.M. gunboats "Viper" and " Vixen" by Sir E. J. Eeed. "Richard Eoberts in this year claimed (No. 13,779) improvements in screw propellers, " making the boss much larger than usual in order that the vanes may act more effectively on the water, and in extending the bosses backwards far enough to admit of their being tapered or otherwise formed so as to allow the water to close upon them without a counter-current being produced." The boss was to be at least one- MODERN HISTORY OF PROPELLERS. 27 third the diameter and to have at its after end a curved conical point, and forward the boss was to be " softened off with the body of the vessel." A most valuable improvement, and one much appreciated thirty years after, and since, where high efficiency is requisite with high speeds. (See fig. 10.) a.d. 1852. — ■William Clark propounded the making of a screw with the pitch near the tip greater than that at the boss, the latter being u equal only to the speed of the vessel. 5 ' He did not complete his patent. Donald Beatson and Thos. Hall in this year patented a screw with a flange at each blade tip so as to diminish the slip by retaining the water. a.d. 1853. — ■Beatson claimed (No. 175) to make propeller blades corrugated, ribbed, fluted, or ridged in lines. Robert Griffiths took out another patent in this year (No. 492), Fig. 10.— Roberts' Patent Boss, 1851. and includes the spherical boss with gear inside it setting the blades (see fig. 8), but consisting of a pinion fixed to each blade shank geared to a bevelled segment, etc. Joseph Maudslay claimed, in patent No. 646 of this year, to turn round propeller blades by a lever on the side of the shank worked by a link taken into a grooved sliding collar, etc. (See fig. 47.) John Fisher proposed making blades with various openings through them ; to form flanges or ridges at the back, and to enamel the surface. James M'Connell claimed, in patent No. 1775, to make screw propeller shafts hollow or tubular, as is now always done for H.M. Navy. James Mackay proposed to shape the " deadwood " of the ship near the screw into a cylindrical enlargement so as to cause the water to run along without closing in and be again disturbed by the boss. This idea was improved upon many years after by Sir John Thornycroft. 28 MARINE PROPELLERS. a.d. 1854. — Eennie of London built a twin-screw steamer for the Khedive of Egypt, 60 feet long, 6 feet beam, and 21 inches draught of water. The screws were 24 inches diameter and made 310 revolutions per minute when the boat was travelling ten knots per hour, driven by disc engines 13 inches in diameter supplied with steam at 45 lbs. ; with 60 lbs. it was said that a speed of twelve knots was attained. The following year a similar, but somewhat larger, boat, 70 feet long and 7 5 feet beam, did ten knots at 260 revolutions and consumed 100 lbs. of coal per hour. In this same year was built the screw steamer " Brandon," 210 feet long and 20 feet beam, and was the first ship fitted with compound engines. They were of Randolph & Elder's patent design, having two H.P. cylinders 41 inches diameter and two L.P. 64 inches diameter, working two crank shafts and geared to the screw shaft. The boiler pressure was only 22 lbs. and the cut-off in H.P. cylinder four-tenths. With a cargo of 650 tons on board, she steamed the voyage Glasgow to Limerick at eleven knots on a consumption of 14 cwt. of Scotch coal per hour. John Penn in this year took out a patent, No. 2114, for fitting fillets of lignum vitae or some other hard wood in dovetail grooves in the underwater bearings of propeller shafting, now almost the universal practice. William Wain proposed to operate on the blades of a screw so as to change their angle by a rod down the centre of the shaft, "either by sliding or turning with suitable gear.'' a.d. 1855. — George Peacock claims to make his screw of wrought iron with blades shaped like a bee's wing or parabolic in their curvature. Henry Bessemer, on the other hand, claimed (No. 1382) to make screw propellers, their shafts and cranks, of "cast steel and pig iron." Casimir Deschamps and Charles Vilcoq patented (No. 1646) a " free diving boat," what would be called now a " submersible boat," propelled by a screw, etc., and having an electric light at the top. William E. Kenworthy and Henry Greenwood protected the invention of fixing blades to a boss by dovetailing their ends into grooves. This method was years afterwards adopted both by Yarrow and Thornycroft for fixing the blades to torpedo boats and destroyers' propellers. The patent was never completed. (Vide fig. 61.) Christian Schiele proposed a rotary engine for driving screw pro- pellers that was really a turbine, inasmuch as he proposed (No. 1693) MODERN HISTORY OF PROPELLERS. 29 to turn the shaft by the action of steam directed against incisions in the circumference of a wheel upon the shaft. The wheel is in a case connected with the condenser. The steam is said to impinge on one side of each of the curved vanes round the periphery, and escape by the open sides into the case. a.d. 1857. — John Bourne (No. 935) claimed to propel ships by injecting and burning fuel in a fine state of dust in a closed chamber containing air. He said : " The hot air may be used for propelling vessels by being used to move machines resembling a Barker's mill, a smoke jack, or a turbine." a.d. 1858. — Robert Griffiths (No. 319) proposed to make his screws " so that if a straight edge is held against the blade and perpendicularly to the length of the axis it will be found that at a point at a distance from the axis (generally about half the radius of the screw) the straight edge and the blade part company, the blade falling forward towards the ship, or the screw may be so constructed that the blade shall continually fall away from the straight edge. " A.D. 1858. — The steamer " Great Eastern " was built by John Scott Russell from the general designs of Brunei. She was 680 feet long, 82*8 feet beam, and 48*2 feet deep. Her gross register tonnage was 18,915 tons, and the displacement 32,000 tons. She was propelled by a single screw and a pair of side wheels. The screw engines had four cylinders 84 inches diameter and 48 inches stroke. The paddle engines had four cylinders 74 inches diameter and 174 inches stroke. The only other instance of a ship having paddles as well as a screw was the " Bee/' a small ship used for instructional purposes at Portsmouth College in the early fifties of the ninteenth century. A.D. 1859. — Thomas Symons proposed'to fit two propellers, " one above the other, each of them being of less diameter than the solitary one in present use, and may be driven at a greater speed without producing the like vibration ; and the risk of their fouling by floating materials or being struck by shot will be greatly diminished." a.d. 1860. — Herman Hirsch took out patent No. 2930, in which he states: "The improvements apply to the form of the blades of propellers for vessels, the surfaces of which are made of such a curvature that if sections were made by cylindrical surfaces concen- tric with the axis of the propeller, the profiles of the sections on these cylindrical surfaces would show a pitch gradually increasing from the entering edge in such a manner that every successive portion of 3o MARINE PROPELLERS. the surface, reckoned from the entering edge, gives the water an additional impulse backwards as it revolves through it ; and if sections were made by plaues perpendicular to the axis of the propeller the profiles of the surfaces on these planes would be spiral lines concave towards the water, acted on in such a manner that every portion of the surface in revolving through the water impels it to some extent towards the axis, and thus overcomes a portion of its centrifugal tendency. That surface of the blade which acts on the water is therefore made of such a curva- ture as to combine the two curva- tures above described, namely, the curvature of increasing pitch as projected on cylindrical sections, and the spiral curvature as pro- jected on planes perpendicular to the axis." (Fig. 11.) a.d. 1860. — Eobert Griffiths took out a fresh patent, No. 2976, in which he states: "This invention has for its object improvements in screw propeller blades which de- crease in their width of surface as they become more and more distant from the propeller shaft. It is preferred that each screw propeller blade should be a portion of a true screw of the pitch desired, excepting at the further edges of the blades, which after edges are each composed of an angular surface which is in its whole length at the same or nearly the same angle to the propeller shaft as that at which the widest part of the screw propeller blade stands to the shaft." The widest part of the blade should be "at a point about one-half the radius of the screw from the centre of the propeller shaft," but its position may be varied. (See fig. 38.) "The angular surface at the edge of the blade commences at or springs from that part of the screw propeller blade which is widest, and it becomes wider and wider as it proceeds outwards to the Fig. n.- -H. Hirsch's Screw Propeller of 30 Degrees. MODERN HISTORY OF PROPELLERS. 3 1 periphery or circumference of the propeller blade; the angular surface stands at an inclination to the after face of the propeller blade, and consequently, as the propeller blade rotates, the water which has been acted on and put in motion by the fore part of the blade is again struck by this after portion of the blade." " Screw propellers of smaller diameter than those heretofore used may be employed without decreasing the propelling effect." The last paragraph is very interesting, especially in the light of modern experience. A.r>. 1861. — Kobert Wilson patented, among other things, the idea of making the section of a propeller ct a long and narrow ellipse with pointed extremities so as to cause little disturbance in passing through the water." A.d. 1861. — William Holland Furlonge claimed " in screw steamers, the propeller can, if desired, be made to draw the water through the condenser and deliver it near the stern post." A.D. 1862. — Alfred Krupp, in his patent No. 1116, says the "invention consists in forming screw propellers in one piece or in two or more pieces from a solid block or blocks of cast steel, and forging the said block or blocks into the necessary shape." a.d. 1862. — Eobert Griffiths extended his patent by taking out No. 1618, and containing the following: — "The improvements consist, first, in constructing screw propellers for steam ships and boats with blades and centre boss of similar form and construction (or the blades may be cast on the boss) to those described in the Specification of a Patent granted to me the 20th February 1858, No. 319, but having four blades (or two sets of blades) which are to be fixed either ' to the same boss or to separate bosses on to the screw shaft, and so fixed that one set (or pair of blades) is placed before the other set, the first set or pair of blades next the shaft to be of larger diameter than the after set, so as to get greater hold on the water for propelling the ship." a.d. 1862. — Thomas Carvin patented No. 2301, and claimed to make the propeller blades in the shape " of an elongated irregular oval." a.d. 1862. — The last large paddle steamer for the Atlantic trade, the " Scotia," was launched and owned by the Cunard Company. This ship in 1879 was converted to a twin screw, and employed after as a cable-laying and repairing ship. She was 379 feet long, 47'8 feet beam, and 37*7 feet deep. 32 MARINE PROPELLERS. A.D. 1863. — Arthur Rigg, junior, claimed " to surround a screw propeller by a cylinder and to have a grating in front of it to prevent weeds entering." A.D. 1863. — F. E. Sickels proposed to make " propeller blades of vulcanite strengthened with iron bands." A.D. 1864. — William B. Adams took out protection for the use of (( liquid fuel, such as oil, melted grease of any kind," but he pre- ferred coal oil, petroleum, or shale oil for the propulsion of vessels, and described the burners by which it might be used. A.D. 1866. — Herman Hirsch obtained another patent, No. 17, in which he claimed : " In the improved propeller the front or enter- ing edge is so inclined to the axis that it cuts the unbroken water with little or no resistance, and the rest of the blade gently curves backwards, being more and more inclined so as to give the water a gradually increased backward motion, thus not only avoiding the excessive resistance caused by a sudden impulse on the water, but also maintaining an uniform reactive pressure from the unbroken water over the whole breadth of the revolving blade/' " The two extreme lines and the whole of the intermediary lines of the surface are spirals comprised in an angle of 60 degrees, thus generating from the axis to the circumference a hollow curved or spoon-shaped form of blades/' " The points of the spirals may be rounded off," and their curvatures modified near the axis, to give additional strength. (See fig. 40.) a.d. 1866. — John Henry Johnson patented No. 256, and pro- posed to use two distinct sets of paddle wheels, the after pair about one-eighth larger than the forward pair ; and he goes on to say that in a ship 500 feet long the two sets of wheels are to be about 150* feet apart. a.d. 1866. — William Dudgeon, in his patent No. 2068, claimed among other things that " as regards ships, the object sought is to provide support for the two screw shafts, which project astern in parallel lines with the keel from under the ships' quarters, and respectively are encased and carry the screws. For this purpose a flat horizontal chamber is constructed on each side and fixed to the skin of the vessel, projecting laterally therefrom so as to enclose and support the screw shafts; and inside the vessel, extending from side to side, is fixed a horizontal plate framing to correspond with the level of the shafts, so as to form a substantial lateral support between the chambers, which are rendered more firm by a transverse vertical MODERN HISTORY OF PROPELLERS. $$ bulkhead. The extreme ends of the chambers, wherein the shaft bearings and stuffing-boxes are fitted, are further supported by wrought-iron diagonal stays or brackets, which are fixed to the sides of the vessel by riveting through the skin and ribs." A.D. 1866. — Dr A. C. Kirk took out provisional protection for the following: "This invention has principally for its object to render steam dredgers capable of being more easily manoeuvred than hitherto ; and it consists in employing for that purpose centrifugal or other pumping apparatus to be worked by the main or separate engines, and to cause the projection by suitable passages or orifices at the stern of one or more streams of water." " As it is of great importance to have the power of turning and generally manoeuvring such vessels independently of the tug usually in attendance, deflectors or rudders are attached to the stern orifices through which the water is projected ; whilst to permit of reversing the direction of propulsion, either rotatory reversible pumps are used, or reversing valves are fitted to the water passages communicating with the pumping apparatus." A.D. 1866. — J. S. Martin and J. F. Droop claimed to propel and steer ships by means of a stream of water ejected by means of a jet of steam, as in the case of the Giffard injector. A.D. 1867. — Robert Atkin claimed to fit three, four, five, and six screw propellers ; namely, two at each end and at any suitable distance from the stem and stern. a.d. 1873.— James Howden took out patent No. 3278, in which the steamer is fitted with a large propeller at each end, "the pro- peller being of such a size that the blade thereof extends below the keel." The keel may be curved downwards at the bow and stern so as to pass under the propeller. a.d. 1873. — Samuel Osborne and Stephen Alley took out patent No. 3379. The invention consists in forming propellers of thin sheets of metal- — steel, gun-metal, etc. Each blade may be made of a single sheet, but it is preferred to be made of two thicknesses with a space in the middle. The blades thus formed are fixed on the boss in various ways. They may be let into a groove in the boss and fixed by wedges, or flanges on the blade may be bolted to the boss, or the blades may be fitted to flanged pieces in sockets in the boss, or the boss may be made up of four quadrants the flat surfaces of which are shaped so as to receive the blades between them. 3 34 MARINE PROPELLERS. a.d. 1873. — John Isaac Thornycroft took out patent No. 3551. The object of the invention is to cause the water to be driven back- ward by the screw, " in what may be termed hollow cylinders, or concentric annular volumes/' For this purpose each blade is curved in the direction of its length, "in such manner that assuming the blade to be cut in a plane passing through the axis of the propeller, and through the centre of the blade in a direction parallel to the said axis, the section of the blade will be convex on its driving face, that is, on the outer, or, in other words, the after side or surface," the pitch of the blades increases gradually from the boss towards the centre, and then decreases towards the outer edge in the direction of the screw's diameter. In the direction of the axis the pitch increases towards the after end of the screw. a.d. 1873. — Robert Griffiths took out patent No. 3817, relating to methods of affording a larger supply of water to screws working in tunnels at the bows and stern. (1) The tunnel for a stern propeller is made with a " lip, scoop, or projection," extending down from the after side of the mouth of the tunnel. The projecting piece may be hinged, so that it can be raised up when not required, as in the case of vessels driven by screws both at the bows and stern, in which case it is stated that the additional speed given by the bow propeller ensures a sufficient supply of water to the propeller at the stern. (2) The tunnel for the bow propeller is made with a flaring mouth for the above-mentioned purpose. (3) Similar arrangements may be made when twin screws are employed. (4) When one screw only at the stern is used, it is preferred to make it " of parallel form and of increasing pitch from end to end." With such a screw the " lip " is not required. (5) A propeller "of increasing pitch and taper form" is used in the stern tunnel. (6) A screw of increasing pitch is mounted on a taper boss, the screw itself being " parallel from end to end." (7) The outlet end of the stern tunnels may be widened to allow the water to escape freely. a.d. 1874.— Sir J. I. Thornycroft took out patent No. 382. A propeller is placed " within an external recess or recesses so formed in the under part of the hull that when the vessel is afloat the mouth or opening, or mouths or openings, of such recess or recesses shall be MODERN HISTORY OF PROPELLERS. 35 below the surface or level of the water in which the vessel is floating, and the crown or crowns of the recess or recesses considerably above the said surface or level." The action of the propeller empties this chamber of air and fills it with water which rises above the level of the external water, or an air pump may be employed for the purpose. (See fig. 12.) a.d. 1874.— Sir E. J. Eeed took out patent No. 2565. The inven- tion consists in using two screw propellers in the same axial line, one being on a sleeve and the other on a shaft within the sleeve. One of the propellers is larger than the other. In deep water both are to be employed; in shallow water the smaller propeller only, the larger Fig. 12. — Thornycroft's Stern for Shallow Draught Screw Ships. one being two-bladed and placed horizontally so as to be clear of the bottom. a.d. 1874. — James Howden took out patent No. 3246 for com- posite screw blades. The blade is formed of one or more plates of sheet iron riveted to a stem formed with cross pieces, this stem occupying the middle line of the blade. One plate may form the front and the other the back of the blade, or one plate alone may be employed, filling pieces being fitted to the back. " The propelling face of the blade may also have one or more parallel bars of a limited width riveted or otherwise fastened through the plate, stock, and filling pieces, so as to further stiffen the blades, these bars being curved and fixed on the face of the blades to the circle of the revolution of 36 MARINE PROPELLERS. the propeller. The central bar of the stock may also be increased across the back of the plate or sheet forming the face of the blade, so as to take the place of the filling pieces." a.d. 1875. — Joseph Hirsch took out patent No. 576, being im- provements on No. 2930, a.d. I860, and No. 17, a.d. 1866. The invention refers to a special shape to be given to the propeller blade. The entering edge of the blade is slightly curved and the opposite edge has " a considerably greater curvature/' the lines of the intermediate sections of the blade "being more and more curved as they are nearer to the leaving edge. This increase in forward curva- ture is determined in such a manner that the blade has a pitch increasing from the front edge backwards ; but this increase of pitch is greatest at the tip of the blade and becomes less and less towards the root. It is preferred that the pitch along the entering edge should be that due to the forward motion of the vessel, so that the blade has there an obliquity to the axis determined by compounding the velocity of rotation with the velocity of advance, and the increase of pitch at the tip of the blade may be from 20 to 30 per cent." The blade can be made with " a uniform width from the root to the tip," or it may be made " to taper in width from the root outwards, the intersections on its face being in that case not planes perpendicular to the axis, but conical or conoidal surfaces having an obliquity to the axis which becomes less towards the middle of the blade's width." a.d. 1875. — Hermann Hirsch took out patent No. 4479 for " Improvements in screw propellers." The invention relates to making the propeller blade of one of two forms. In the first form the blade has " a double curve, so that seen in edge view in a longi- tudinal direction, it will present the figure of an elongated S." In the second form it has " two such double curves in juxtaposition, so that the profile formed by its opposite edges will, when seen in an oblique direction together, present the figure of 8 more or less elongated." Yarrow's Shallow Draught Screw Arrangement. — Some very interesting experiments were made by Mr Yarrow with a view to adopting the propeller as a means of driving shallow draught ships whose draught of water would be considerably less than the diameter of the screw. Griffiths had suggested a method of doing this, and Mr, now Sir John, Thornycroft, carried out the idea most successfully on the plan shown in fig. 12. But Mr Yarrow wished to go a step further, and to see how far it would be of advantage in such ships if MODERN HISTORY OF PROPELLERS. 37 the obstruction abaft the screw were removed when the ship was in deeper water, and did not require the compulsory submersion of the screw. For this purpose he fitted a flap or shutter hinged to the roof of the waterway in wake of the screw, and arranged to be raised or Flap-down Shallow Draught. Flap-up Deep Draught. 30,000 Speed I Statute Miles per Hour/ Speed (Statute Miles perHourJ Fig, 12a. — 'Yarrow's Drop Flap for Shallow Draught Screw Ships. lowered to suit the draught of water of the boat. That is to say, the flap could always be lowered down so as to touch the water and thereby seal the screw race from the air. The results of the trials made by Yarrow with a boat fitted with this arrangement at the stern is clearly set forth in fig. 12a. CHAPTER III. RESISTANCE OF SHIPS. Ef a plane or thin sheet of material is pushed or drawn through the water in a direction at right angles to its surface, it resists very vigorously ; if it is moved in the direction of its surface the resistance, though very slight, is appreciable, especially if its surface is rough. It will be observed in the former experiment that the water in front has to swell up against the plane and pass around it on either side ; behind the plane there is a tendency to form a depression or space free of water, especially if the movement is accelerated ; and further, that the water in flowing around the sides of the plane and from the bottom fills up the space behind with confused motion and the formation of eddies. If the speed be further quickened it will be observed that there is a flow of water from behind approaching the plane in the direction in which it is moving, due to gravity acting on the water to make it fill the now increasing cavity behind the plate. So far back as 1798 Colonel Beaufoy made some experiments with such planes deeply immersed with a view to ascertain the exact amount of resistance at various speeds and positions, and found The resistance R =f Z- X Au 2 , where A is the area in square feet, v the velocity in feet per second, o- the weight of a cubic foot of liquid, g gravity ( = 32). Now /is the factor which Beaufoy set out to determine, and which he finally concluded had a value of 1*1. Hence for sea water with o- =64, resistance = 1*1 x Av 2 . Many years after, however, Dr AV. Froude was led to put quite a 38 RESISTANCE OF SHIPS. 39 different value, viz. 1*7; and later Dubuat estimated it at 1'43, while another experimenter made it 1*6. It is now very interesting to know that Mr R E. Froude, after most careful and exhaustive experiment- ing, has come to the conclusion that Beaufoy was right in putting the value of /at 1*1. While the value of / has been confirmed, doubts have been cast on the resistance varying exactly with the area, or exactly as the square of the velocity. It will be seen later on that the resistance of ships in practice varies with a lower velocity index than 2, except when the bottom is rough as with fine sand cast on the paint. It is, however, sufficient for all purposes here to assume generally that K does vary with A and v 2 . If the plane be bent into a curve and towed with the convex face leading, the resistance will be considerably reduced ; if the plane be bent so as to form an angle, the resistance will also be comparatively small, and be somewhere between K and the skin resistance R. Now if the plane move in a direction inclined to its surface at an angle 6, then, in sea water it is found by experiment that Eesistance E, - 1'622 nn E^ ^-.- x Av* ; 1 039 + 0-61 sing and the resistance at right angles to the line of motion t, 0-39 + 0-61 sin 6 m Es= — sin^e R - Then K^ K 2 = sin a + 0-39 + 0-61 sin 9. All ships and bodies with ship-like form in moving through the water are subject to resistance from the friction of the water on the skin or surface submerged, however smooth it may be, and a head resistance, due to the pressure of the body on the water in front ; at the stern or following end there is a further cause of loss or increase of resistance, due to the imperfect action of the water in filling in the void space behind, so that it fails to follow up the body without " loss of head " — that is, without any decrease of hydraulic pressure. A " balk " or log of timber being towed through the water may be observed with advantage, as all these phenomena can be clearly seen and fully appreciated. The pressure at the bow 4 o MARINE PROPELLERS. sets up waves by raising a mass in front of it, which only disperses to allow more to form. The friction of the skin at the bow and sides sets in motion other portions of the surrounding water, and from ripples at the sides waves of another kind are formed and spread out fanwise. At the stern further waves are caused by the replacement of the water displaced, and eddies set up in the wake. Experience has taught raftsmen to tow balk timber with the big end leading, as thereby the pressure is relieved from the sides and the speed at which it moves cannot cause much end resistance or wave-making ; also the stern replacement, etc., is easier and effected with less loss of energy. A ship has not a flat end like a balk ; hers are more or less of wedge form ; but however fine the entrance may be, a speed is arrived at when definite waves are formed near the bow and flow away from it ; that is, the water displaced has not time to spread gently and so be unobservable, but is heaped up more or less : they are called the waves of displacement, just as those near the stern caused by the inflow in wake of the ship are called waves of replacement, and both are toaves of translation, caused by the ship being translated from one spot to another. Skin resistance, or, as some prefer to designate it, " tangential resistance," R =j xAp", A being, as before, the area in square feet; v the velocity in feet per second ; j a factor deduced from experiments by W. Froude, at the same time that he found the value of the index n as follows : — Table I. — Froude's Values of j t with Index for Variation 1"825. Length. Coefficient Resistance. Length. Coefficient Resistance. Length. Feet. 80 90 100 120 •00933 ■00928 ■00923 ■00916 Feet. 140 160 180 200 ■00911 ■00907 "00904 "00902 Feet. 250 300 350 400 Coefficient Length. Coefficient Resistance. Resistance. Feet. •00897 450 ■00883 •00892 500 •00880 ■00889 550 •00877 •00886 600 ■00874 RESISTANCE OF SHIPS. 41 Table II. — Froude's Values of n and (j) from Experiments with Surfaces of Different Material. Nature Length of Surface, or Distance from Cutwater, ir Feet. of 2 Feet. 8 Feet. 20 Feet. 50 Feet Surface. A P> 1-83 ! '250 C •226 A 2-00 B •41 C A B C A B •278 C Varnish . . •390 1"85 •325 •264 1*85 ■240 Paraffin . . 1*95 •38 •370 1"94 -314 i -260 1'93 •271 •237 ... Tinfoil . . 2-16 ■30 •295 1 1-99 *278 ■263 1*90 ■262 •244 1*83 *246 •232 Calico . 1-93 •87 •725 l'S>2 -626 •504 1-89 ■531 ■447 1-87 ! "474 ■423 Fine sand 2-00 •81 •690 ! 2-00 "583 ■450 2-00 ■480 ■384 2-06 -405 ■337 Medium , , . 2'00 •90 •730 2-00 '625 '488 2-00 534 •465 2-00 -488 •456 Coarse ,, , 2-00 1-10 ■880 2'00 '714 ■520 2"00 •588 ■490 ... Column A is the index or value of n. Column B gives the mean resistance in pounds per square foot of the whole surface to the point named at a speed of 10 feet per second. Column C gives the actual resistance per square foot at the distance named for the same speed. The resistance decreases as the distance from the bow increases, for in the case of varnish, while 2 feet from the bow, it is *390 ; 8 feet from the bow, '264 ; 20 feet from the bow, '240 ; and 50 feet from it is only '226 lb. per square foot at 10 feet per second. It was usual formerly to assume that the resistance of a ship varied throughout as the square of the speed, and to state that the I.H.P. varied with the cube of the speed. One result of the early pro- gressive trials was the dissipation of that idea. No doubt, theoretically, it was not far wrong to say that power varied as the cube, but what power was meant is another matter ; one thing is certain, it could not be the gross I.H.P. With modern marine engines at full speed the efficiency is from 0*90 to 0'95, so that the amount absorbed in over- coming friction working the pumps, etc., is from 5 to 10 per cent. Of this amount §ths vary with the revolutions and fths with the square of the revolutions. Hence an engine whose efficiency is 90 at full speed at half the revolutions will absorb only 3-5 per cent, of the original I.H.P., while the I.H.P. at half revolutions will be probably one-seventh the gross ; the efficiency then will be 0755. With turbo motors the mechanical efficiency is still higher, and may be taken at 98 per cent, at full speed. 42 MARINE PROPELLERS. The wetted skin is, to those who have no model and tank apparatus wherewith to experiment, still the criterion of quantity with which they have to deal and guide them in estimating the power required to drive a ship or the resistance encountered in towing a ship at a certain speed. For, although the wave-making absorbs a large amount of the developed power, and especially so at high speeds, it is trivial with ships when moving at speeds well under the designed full speed. With a tug-boat running free at full power, the waste in wave-making is enormous ; with a 20-knot ship moving at 16 knots the waves are inconsiderable and at 10 knots are negligible. At a speed of 66 per cent, of the designed full speed, it is sufficient for practical purposes to assume the skin resistance with a liberal allowance of coefficient of friction as the gross resistance of the ship when towed. The area of the surface of the ship immersed can of course be Fig. 12b.— Kirk's Block Model. measured and computed; but it is a long and troublesome task compared with other methods which give a close approximation to the true amount. Of these the earliest and best known was : — 1. The general idea proposed by Dr Kirk is to reduce all ships to so definite and simple a form that they may be easily compared, and that the magnitude of certain features of this form shall determine the suitability of the ship for speed, etc. As rectangles and triangles are the simplest forms of figure, and more easily compared than surfaces enclosed by curves, so the form chosen is bounded by triangles and rectangles. The form consists of a middle body, which is a rectangular par- allelepiped, and the fore body and after body prisms having isosceles triangles for bases ; in other words, it is a vessel having a rectangular midship section, parallel middle body, and wedge-shaped ends, as shown in fig. 12b. This is called a block model, and is such that its length is equal to that of the ship, the depth is equal to the mean draught of water, RESISTANCE OF SHIPS. 43 the capacity equal to the displacement, and its area of section equal to the area of immersed midship section of the ship. The dimensions of the block model may be obtained by the following methods : — Since AG is supposed equal to HB, and DF equals EK, the triangle ADF equals the triangle EBK, and they together will equal the rectangle whose base is DF and height AG. Therefore, the area ADEBKF equals EK x AH. The volume of the figure is this area multiplied by the height KL. Then the volume of the block is equal to KL x EK x AH. But KL x EK is equal to the area of mid section, which is by supposition equal to the area of immersed midship section of the ship, and the volume of the block is equal to the volume displaced by the ship. Hence Displacement x 35 = immersed midship section x AH ; or, AH = displacement x 35 -~ immersed midship section. Now HB = AB-AH, and AB = the length of the ship. Therefore, the length of fore-body of block model is equal to the length of the ship, less the value of AH as found above. Again, the area of section KL x EK is equal to the area of immersed midship section, and KL is equal to the mean draught of water. Therefore EK = immersed midship section ~ mean draught of water. Dr Kirk also found that the wetted surface of this block model is very nearly equal to that of the ship ; and as this area is easily calculated from the model, it is a very convenient and simple way of obtaining the wetted skin. In actual practice, the wetted skin of the model is from 2 to 5 per cent, in excess of the ship ; for all purposes of comparison and general calculation, it is sufficient to take the surface of the model. The area of bottom of this model = EK x AH. The area of sides = 2 x FK x KL = 2 (AB - 2 HB) x KL = 2 (length of ship — 2, length of fore-body) x mean draught of water. The area of sides of ends = 4xKBxKL = 4 VHB 2 + HK 2 xKL = 4 J Length fore-body 2 + half breadth of model 2 x mean draught of water. 44 MARINE PROPELLERS. The angle of entrance is EBL ; EBH is half that angle ; and the tangent EBH = EH-^HB. Or, tangent of half the angle of entrance = half the breadth of model — length of fore-body. From this, by means of a table of natural tangents, the angle of entrance may be obtained. The block model for ocean-going merchant steamers whose speed is from 15 knots upwards has an angle of entrance from 15 to 24 degrees, and a length of fore-body from 0'3 to 036 of the length. That of ocean-going steamers whose speed is from 12 to 15 knots has an angle of entrance from 24 to 30 degrees, and fore-body from 0*26 to 0-3. Beaton's Rule for angle of entrance of " block model " : Angle in degrees = 70^^. L is the length of ship in feet ; S is the speed in knots. Dr Kirk measured the length from the fore side of stem to the aft side of body-post on the waterline. This is an unnecessary re- refinement when screw steamers alone are being compared, as then the length may be taken as that " between perpendiculars." However, when small or moderate size screw steamers are being compared with paddle-wheel steamers, it may be necessary to measure in this way. 2. Mumford's method is a simple one, and gives results much nearer the actual surfaces than Kirk's. L is the length of the ship between perpendiculars in feet. B is the greatest beam ; H the depth of immersed midship section. b is the block coefficient. D is the displacement in tons of 35 cubic feet. Then Wetted skin = L (17 H + 6B). 3. Seaton's Rule. — It is, however, not always easy to ascertain the moulded draught of every ship, for some have flat keels and some bar, the latter of uncertain depth ; moreover, while the displacement of a ship on trial is given, and from Lloyd's Register and other sources the length and beam can be always obtained when it is not stated, the draught of water is seldom given. For these reasons, and the fact that quite as accurate results are given by it in even less RESISTANCE OF SHIPS. 45 time than by Mumford's, the author has devised and recommends for use the following formula : — D being the displacement in tons and K = L-^B. F = 42x */K. Wetted skin = FxD 2 A For ships 8 to 10 beams F = 71 to 74. 6 to 8 beams F = 67 to 71. 4*5 to 6 beams F = 62 to 67. The following table gives the actual values of F for the variations in K. K being the ratio of a ship's length to her beam. F the factor with which to multiply D% being 42 x 4 JK. Table III. K. F. K. F. K. F. K. F. 74 7 4*0 59'40 6-0 657 8-0 70*6 10-0 4'1 59'85 6-1 65-98 8*1 70-82 10-1 74-89 4'2 60-25 6-2 66*24 8-2 71*05 10'2 75'08 4-3 60-6 6*3 66-5 8-3 71*27 10-3 75*24 4'4 61-0 6-4 66-75 8-4 71-48 10-4 75-42 4*5 61-3 6-5 67 '0 8-5 71'69 10-5 75-58 4'6 61*65 6-6 67-25 8*6 71-90 10-6 75-85 47 61*95 67 67-5 8-7 72-11 10-7 76-05 4'8 62'3 6-8 67'75 8-8 72-33 10-8 76-0 4-9 62'6 6*9 68-0 8-9 72-55 10-9 76-18 5*0 62 '9 7*0 68-25 9-0 72-76 11-0 76-35 5*1 63*2 7*1 68*5 9-1 72-97 11-1 76-53 5*2 63-5 7-2 6874 9*2 73-18 11-2 76-68 5-3 6378 7-3 68-98 9-3 73-39 11-3 76*85 5-4 64-1 7-4 69*23 9-4 73-59 11-4 77-03 5-5 64-35 7-5 69-45 9-5 7377 11-5 77'2 5*6 64-65 7-6 697 9-6 73-96 11-6 77-35 57 64-9 7-7 69-93 9-7 74-15 117 775 5-8 65-18 7-8 70-15 9-8 74-34 11-8 77-65 5'9 65-46 7-9 70-4 9-9 74-53 11-9 77*82 Kirk's Method, when applied to ships of the older type having a good " rise of floor " and moderately fine lines, gave results within a very close percentage of the actual surface, but with flat floors and very fine lines the error becomes serious, and as much as 10 per cent, in excess of the truth in extreme cases. MumforcVs Bide suits every class of ship, and the error is seldom over 5 per cent., and generally much less. Section's Rule suits all classes of ship, and the error is generally even less than that of Mumford's when the draught of water is not 46 MARINE PROPELLERS. less than one-third the beam. The examples on page 47 illustrate the above. Limitation of Speed due to Form and Size. — That each form of ship has a limit to speed is well known ; also it is equally obvious that the length and size of a ship must have considerable influence on that limitation. A small ship with a prismatic coefficient of 0*700 could not be driven economically at a higher speed than 10 knots ; any increase in power developed is employed chiefly in wave making, whereas a ship 500 feet long of the same form could be driven economically at 19 knots per hour. Again, for 19 knots with a ship 300 feet long the prismatic coefficient should not exceed 0*610, otherwise the power would be excessive by comparison. In making investigations on this subject the author found that in this case also */L was one of the leading factors for determining these questions, and the formulae he arrived at involving it he has found to give results in agreement with good practice ; they are as follows : — (1) The highest prismatic coefficient for a speed S and a length L is /=0-39 Z/L+ l/S. (2) The fastest economic speed for a length L and a coefficient /is g / 0-39«/L \» (3) The shortest length of ship having a prismatic coefficient / for a speed S is V 0-39 / Froude's investigations on skin resistance led him to fix a set of values for j varying from '00963 for ships 50 feet long to '0088 for those 500 feet long when the index value for v is taken at 1*825. Some more recent investigations have shown that that index value requires modification, and that really — Index value of n for ships from 100 to 500 feet long is 1*829 for modern enamel paints when quite clean, and 1*827 for clean bright copper and zinc, and 1*843 for these metals when rough from corrosion. RESISTANCE OF SHIPS. 47 s o fa "I M w H >h W Q H < « O CO Q H H H CI IZi I— I o a 2 rH in p s? rH Ol t- in rH rH o CO © a. OS 00 00 Oi CO Ol CO lh CO os in lis to 4n in in i>. X t- t- CO r- a S o o o o o O o o o O o CO O o o © © © o © O CO © © CO -# ::■ o ■"■■ Ol m m m co '.- 1 r- Ol -* Ol © Cl m o o m -*i LlO mi 00 o_ ■■,- 1 m >n o 00 °i CO ■* m m_ cO rH Ol OS rH r- •* «# CO -* r~ Ol 3 GO co" ~ m "* ■3S co 1-1 rs Cl ~ cf OS CO CO CO Ol iH O CO m rH Cl o en 3 co -o m o m ■« Tf -K CO CO CO CO Ol 01 Ol > l-*5 J3 _; ! m rH co o ■TO o o o L-- X c CO o CO o © © m o •* CO © © © o -;-. Ml :o o o » ~ O o 1-1 o ■;:■ CO 1^- CD ■^ p 01 m CO 3 Ol c 3- CO 01 "* i~ Cl i-H CO CO in OS OS CD i-0 CO t- O rH Ol X -n in CO © CO as Oi m o CO t> o m m r-l Ol o I-l of r~ m m Cl r-t rH Ol m l-H rH o CO CO co CD m m m -♦< ■V -* co co CO Ol Ol Ol Cl Ol Ol Ol rH rH rH •" H 13 +3 -4^ CD O OS O O o o O Tjl Ol Cl X rH CO o © m m © © © O o m o o o Cl Ol CO ■i m CO ,,.. I- ;o CO co CD CO Ol "* CO CD X o rH CD Ol ■4-1 m /.! m Ol in (M fc» CO 1- o CO CD i-H O i-H_ Cl_ o rH rH_ 00 o CO 33 01 | 00 <*" o CO ci" o" cf in Ol cf Z> —■ t- CD •*" ■*" l-H ©" ©" Ol" CO" r-i Ol o" »— ; co 'J CO o m m -J* -* CO CO CO CO Cl Cl Ol Ol Ol Ol Ol Ol l-H l-H rH 1 m O CO CO r» t- 00 l-H © M ■„•■ -f *n o ■* CO "* CO l-H u CO OS CO oo OS OS -* If \ M lO T-H co OS X CD CO co of r- m CO Ol CM Ol OS Cl Cl "2| ig S2 CO o CO m O co m o r~ 1— 1 -# -1 o Cl t^ m ■* I— rH m Ol "* CO l-H -« l-H © Cl r-T HH u - 1 ««i o*= 1 ■';■-. "-• Cl "* m r-f T-H CO m m OS co rH ■;i <-< Cl Ol © 3 « as 1 m s* ai Cp CO I— i~ i-- 1> t- »>- i-. 1 ® o o O o © © o © s *5 tp cp CO ep r~ CO in o m o © b © 6 o b b b o b b b b b © © b © b © © b b © 3 o o o o "* co o CO o r~- ■* CO o Cl ■* Ol >H i^ as © © r- o rH Ml a o o — CO CO o I~ o © CO M => CO i-H i>. C^ CO r~ t~ m m CD CD m in m cm t~_ Cl o_ OS >* O 1- ■':':, m **, t- cO in m 0_ Ol CO CO i-H rH t-- H 00 CO CO r* l-H co" in m <* r-i CO CO Os" X r- co co" co" in" m CO Cl Cl" |H o o CO fM m CO m o O O o CM 1^ o Ol © © o m © m o m © CD oi GO Cl r~ -f • a m -o o m Ol o in CO oo CO Ol in ** CO [— -.t' 00 OS CO a cu CD c- CO •o -o CO in CO m t~ in -*" CO in m co m in -* ■* -f CO CO CO p J3 *= iD o CM o o o m 1^ o CO o o m m o m © © CO © © o m o Esfl CO CO co l- m ":■ i~ m r- CD CO in - > Cl Ol X oo CO CO © CO J^ rH CO c ID H> to CO CD m in CO •*< m •* CO ■* -n CO --0 CO Ol Ol Ol CO tJI ■',-J Cl CO Ol CD V) 3 O M S Hi CD 1—1 ; rf „ ri +J d go o CD - s « CD o cd CO '5 T3 CO CD CD Hi TO o CD -a S M "o UJ CD ft a e8 'a ft a CD o p = en W to p- 3 a « "S 1/1 3 3 CD o 3 S3 ■4-i o3 s- 3 ■-: CD 3 CO £ tH 3 '3 "3 *h _« CD p. a « CD CD Jh 3 u; HJ X P .5 tO o g e > .2 o CJ !> M 00 CD o p tH O O Hi CO CD CD .a Ph .D ;j" C3 Eh HH i>> 6h ■u a CD o UJ P <1 w O P X ^< Ph Ph Ph T3 cU i> - W » CO X CO CO CO CO _ " CO X; ■fl VI' CO X CO CO X X - l-H ^J l-H ^H i—i wj " CO CO - X CO co" J- CO OQ CO CO X CO CO X X 0*1 <( r*i o-. <-, ^. CO xJ CO CO CO x' H H H EH H H H H H P H CO m a W W M W CO H H H H H 43 MARINE PROPELLERS. As it is in a general way more convenient to deal with these questions on a reference value per square foot at 10 knots speed, the following table gives the resistances per square foot of surface of the different materials as probably actually found in practice as deduced from the foregoing experiments and observations on ships on trial. Table V. — Average Resistance per Square Foot of Different Material and Conditions in Actual Practice. Under Tinder Under Under 100 Feet 2U0 Feet 400 Feet 600 Feet Long. Long. Long. Long. (1) Copper bottom, new, bright and clean ■898 •874 •856 "850 (2) ,, ,, clean .... •966 ■943 •926 •926 (3) ,, ,, corroded .... 1-250 1-230 1-175 1-140 (4) Enamel paints, best kinds freshly done ■970 •944 •916 •900 (5) ,, ,, good and clean . 1*030 1-000 •970 •950 (6) Ordinary paint ; tar, etc. . 1140 1-100 1-070 1-050 (7) Tar and plumbago polished •922 •900 •884 •875 (8) Fine grass on paint 5-200 4-800 4*670 4 540 The resistance or losses due to wave-making and eddying are called generally residual losses, and may be calculated with a fair amount of exactness by the following rule : — Residual resistances = x MxBxv x/B + zL 2 ' M is the area of immersed midship section in square feet. B is the extreme breadth in feet. v is the velocity in feet per second. K is the ratio of length to breadth ; that is, L-l-B. H is the ratio of the length to velocity square; that is, L~v 2 . # is a factor got by 150 ~K 2 . z is a factor got by 0*55^ ^/100H = ^^. (a) The total resistance is then made up of the skin resistance plus residual resistance. It was the common practice formerly to determine the total resistance by the following formula, where D is the displacement in tons and S the speed of the ship in knots. RESISTANCE OF SHIPS. 49 (b) Total resistance = D% x S 2 in pounds. It was due to the following, viz. : — Resistance x S x ^° = E.H.P. x 33,000, 60 or RxS = E.H.P.x325 foot-lbs. E.H.P. = k x I.H.P. Then E x S = I.H.P. x 3257c, By the Admiralty speed formula, I.H.P. = - ^ S3 . By substituting the above value of I.H.P. the following holds: — RxS D%xS 3 p D%xS 2 QO , 7 335*= C ;orE=-_ ^_x325fc Now if the efficiency, etc., of the ship is high, h will be high, as will also the value of C. Taking h at 0*6, then Then if C = 195, which would be its value with the old ships of Rankine's day, R = D% x S 2 . To-day we have ships with an efficiency as high as - 70; the coefficient C, then, would be 227'5. If the efficiency is as high as 0-75, then C would be 244. Test of efficiency of the propulsive powers of a ship can be, and has been, gauged for more than fifty years by the magnitude of this coefficient ; as also by that of another coefficient which takes into account the size of the immersed midship section. The formula is: — C = area of immersed midship section x speed -=- I.H.P., usually written *g^. The older engineers set more value on this latter criterion than on the former. To-day, however, the latter is not often referred to. Example. — To find the wetted skin resistance and E.H.P. of a steamer 300 feet long, 38 feet beam, and 13'5 feet mean draught. Her displacement is 2400 tons, prismatic coefficient is 0*61, block* coefficient '546, and she is to steam at 18 knots. Allowance for keel 6 inches. 4 5 CO CD i> iO O m •-i i~ -O CD 1^ CO CO 4J •* -* -f lO "* CO CO Cl CO Cl CM i-H CO CO d 35 x X X X X X X X X X X X X X O ^M O O 1- in m m — O Cl l-H r> m Cl CM 01 CO CO t-- CO 1— 1 l-H rH rH l-H ^ m ci (O .-h a r-l O CO i-H CO m CO .0 rH 00 1- co m -H ci Ol ci CM r-> r-l i-H l-H r-l " CO |PL| OD 1— >0 f~ O 00 iH CM rH CO OJ iH CD Ph CO 1— a O Cl CO -* CD CD S -*k ;o Cl a I-~ in CO m i-H l- CO ■70 co CO -*1 CO ■£> ■:o 01 Cl i-H CO r- ( +j « ° (N r— Cl ■n m m 00 CO ■-> r- D Cl co ;i-| m OS •91 i-H ■ ~. m m 1-. CO ■_;■ o< t~ CM I-H r-l i-H Cl Cl l-H a> ft > O O iO 1- O CO O O CO O CM 'CO CO m r— tf CM co CO 01 CO ^f -* -H CO m CO Cl ^ O 01 m m CO m « co i~ O i— ;i Cl r- 00 i- I— 0» l~ CO 1- CO CO ■* -f -** m — < H CO _ -- — -- 0) ft« o .S3 £h Q tii 00 CO Cl CO ■ ' I- cc —1 CO Cl CM l-H q o £ 00 « CC' CD m r- m Cl Cl m co -n 00 CO CO l^ r-l CO co 00 1-- CD CO CO CO CO co Cl Cl 1—1 Cl CO P o CHAPTEE VI. DIMENSIONS OF PADDLE WHEELS. The diameter of a paddle wheel depends on the speed of the ship, the revolutions of the wheel, and the rate of slip. Let D be the effective diameter of the wheel, which, for purposes of calculation, shall be taken to the float " centres " of a feathering wheel and the middle of opposite floats of a radial wheel ; A the area of one float in square feet ; V thevelocityof thewheel at float centres, etc., in feet per second; S the speed in knots ; v the velocity of the ship in feet per second ; s the speed of the ship in knots ; e the efficiency of the machinery ; E the efficiency of the machinery and propellers ; R the revolutions per minute of the wheels ; T R the tow rope resistance of the ship in lbs. ; (1) The power delivered to the wheel N.H.P. = I.H.P x e ; (2) The power delivered by the wheel P R .H.P. = I.H.P. x E ; (3) The stream of water from wheel per second = A x V cubic feet ; (4) The weight of water projected = A x V x 64 pounds ; 54 (5) The mass of this water = Ax^Xt- = 2AxV; (6) The acceleration given to the water = V — v per second; then the pressure on the float or the thrust on shaft bearing from reaction at float = 2AxV(V- v). (a) The work done = thrust multiplied by speed of ship in feet per minute = 2AxV(V-7?)x6Oy = 120AxV(V-v>. v 7 33,000 275 Now since V = — — — l and the speed of the ship is a known DIMENSIONS OP PADDLE WHEELS. 8 1 quantity, it is easy to determine from the fundamental equation (a) the relation between D and A. D = 6 °— - ^ + ( v ~^ )60 7rE 7rE That is, the diameter of wheel = speed of ship plus slip (both in feet per second) -f ?r x the revolutions per minute. Examples. — To find the diameter of the wheel for a ship whose speed is to be 20 knots when running 50 revolutions per minute with a slip of 20 per cent. Here v= — = 2027 feet per minute, or 33'8 feet per second. SlipV-v = ^~; orX^ = 2027; then V = 2534 feet per minute, or 42*2 feet per second. D = 2534 -r 3-1416 x 50 - 16-14 feet. Reaction on the float of a paddle wheel whose diameter is D at E revolutions per minute and the slip 20 per cent. The float area being A, the speed of wheel per minute is 7rD x E or 7r b per second ; ,, if- 20 tDE ttDE the acceleration is -— x ~^-= OAh ; 100 60 300 weight of water moved per second = X 64 ; the mass of water moved per second = x kt ? = — — ^ — ; f , f . ! AxttDE ttDE A(ttDE) 2 a /DE\ 2 thrust m pounds = -- — x „„^ = — V^^tt^- = A( — ] ; F 30 300 9000 \30/ 80 the speed of the ship is — — x 7rDE = 2*5 DE feet per minute ; work done = A x ( W x 2-5 DR = **&*? ; \ oO / 3o0 H Ax(DE)^ Ax(DE)3_ /DEy. ' " ' 360x33,000 11,880,000 \228/ ' that is, the area of one float = E.H.P. at one wheel x ( r—= ) \JJli/ = I.H.P.xExQ 3 . 82 MARINE PROPELLERS. Example. — To find the area of a float of a wheel whose diameter is 22 feet, the revolutions 36 per minute, the I.H.P. per wheel 2100, and the efficiency 0*66. Area = 2100 x 0'66 x(~^-Y = 33*5 square feet. Area of Floats. — In practice the following rule holds good. Let I.H.P. be the total power developed by the engine, then each float should have the area A in square feet, D being the diameter in feet to float centres and R the revolutions. Rule.— A = LH.P.xEx (^~) xF. \1 ) x Pi/ For high-class machinery and feathering floats . E = 0'7 For ordinary „ „ „ . E = 65 For high-class machinery and radial floats . E = 0'60 For ordinary „ „ „ . E = 055 For stern-wheelers with one wheel only. Radial . F = 0*70 For „ „ „ Feathering F = 115 For side-wheelers with two wheels. Radial . F = 0"40 For „ „ „ Feathering . F = 0-60 The pitch of the floats — that is, their distance apart — must depend largely on the speed of the wheel. Each float sweeps the water before it as shown in fig. 14, leaving a hollow behind into which the oncoming water flows by gravity ; consequently there is a slope of water always behind each float, while the surface at which the next float will touch will be below that of still water ; that is, below the plane of the flotation of the ship when at rest. It is easy to calculate how far the back of a float may be bare from the common formula v = 2gh; taking g at 32 and v as the velocity imparted by the float in feet per second, we have h=— m Example, — To find the amount of the baring of a float moving at a velocity 50 feet per second when the slip is 20 per cent., the accelera- tion is 20 per cent, of 50 or 10 feet. Then the head 7 10 2 100 , , a . , hz= — — _^ = 1'56 feet. 2x32 64 That is, the float in question, if its dip is such as to bring its inner edge on the water level, will have its back bare for a depth of 1-56 feet when the ship is moving at a speed of 23"6 knots, DIMENSIONS OP PADDLE WHEELS. 83 Number of Floats. — An old ride for member of floats is Number of floats = 60 -f Jll So that with constant revolutions the larger wheel has the greatest pitch of Heats. Proportions of floats of a feathering wheel differ from those of a radial, thus : — In a radial wheel ratio of float length -f- breadth = 4 to 5. In a feathering wheel „ „ „ = 2'8to3. Originally, floats were made usually of English elm. To-day that timber, as well as Canadian elm, makes a satisfactory board ; in fact, what is wanted is a strong, tough wood which stands the action of water and which does not easily splinter or warp; anything having these qualities and not too heavy will do for the purpose. For in- shore working tug-boats even pitch pine has been used; but this splits too easily, and is, moreover, a fairly heavy wood. Thickness of wooden floats on a radial wheel was usually about one-eighth of their breadth, those of a feathering wheel one- twelfth their breadth. Eatio of breadth of float to thickness is for radial 8, feathering 12. The large floats are made in more than one piece, and in any case it is a good thing to through-bolt them, although it is urged that in doing so they were weakened against cross bending by having the holes. A better method of tying them together is therefore by bolt- ing through and through flat bars, or even angle bars, fitted across them, as the float is also stiffened thereby. The edges, especially the outer ones, of wooden floats are always carefully bevelled, as are also the ends, both being on the back sides. Steel floats are often used to-day instead of wood, because they offer less resistance at entry and exit as well, and are easily kept with a smooth surface by the application of black varnish or enamel paint. Such floats may also be made curved as shown in fig. 19, and are stiffened by angle bars as well as by the feathering gear. Steel plate floats are heavier than wood, and in the opinion of the old engineers, whose experience was largely with paddle steamers, the latter is preferable, inasmuch as a wooden float is incapable of getting bent out of shape as a steel one is, and therefore cannot do mischief to the boxes, to the wheel, the gear itself, and other parts. They are 84 MARINE PROPELLERS. more easily removed than a bent steel one, and the ship's carpenter can always make a new one as well as repair an old. Construction of the paddle wheel is seen and can be studied by referring to figs. 18 and 19. It consists of a large boss with conical flanges at each end to suit the splay out of the arms. The old radial wheels had a boss with parallel flanges and the larger ones with a flange in the middle to suit the three sets of arms, rims, etc.; and as there was no feathering gear to arrange for, the boss was as long as the wheel was broad at the rims. Paddle-wheel bosses were made of cast iron of the best and toughest quality, and great care was taken to avoid splitting at the flanges. To guard against the contingencies of hidden cracks, and to fortify the boss to withstand the heavy shocks it gets in a sea-way, the cast-iron boss had been shrunk on the edge of each flange, and on spigot ends formed around the shaft-hole rings or tyres of wrought iron, and latterly of steel. (See fig. 19.) Since steel castings have been available for the purpose, it has been customary to make these bosses of that metal, taking care that they were carefully annealed and " let down," so as to have no initial stresses due to cooling in the broad flanges. Formed on the flange are fillets, between which the arms are care- fully bedded so as to take all shear off the bolts. In the same way these fillets are fitted to the facings for the diagonal ties, so that there may be as little shear as possible on their bolts. The boss is carefully bored out so as to fit on the shaft end ; and although it was the old practice to bore the hole parallel, which was an improvement on the system preceding it, wherein it was rough and the boss was " staked " on with four stakes or long wedges, it is better now to make the shaft end slightly taper, say one in 48, and key it on tightly with two good-fitting keys at right angles. Let cl be the diameter of the inner journal of a paddle-wheel shaft as found from the formula d= I - — — x F. / I.H.P. , V E ' (a) For an engine with a single cylinder . . . F = 80 (b) For an engine with two cylinders and cranks coupled at right angles F = 58 (c) For an engine with two cylinders and intermediate shaft . . . . . . . p .F = 50 (J) For an engine with two cylinders, solid crank shafts at right angles F = 55 DIMENSIONS OF PADDLE WHEELS. 85 Thickness of cast-iron boss around shaft ,, „ cast-steel ,, ,, ,, Diameter of outer flanges Thickness of cast-iron flanges „ „ cast-steel „ Breadth of two hard steel keys, each Thickness „ ,, = 0-28 xd = 0-20xd 4xd = 015xd = 0-llxd -0*18 x d + 0'25 in. -0-09xt^ + 0-25 in. KB, — All shafts above 8 inches diameter should have two keys of about the size given above. When the diameter is less than 8 inches, one key is sufficient, and it should be not less than given by the above. Paddle-wheel arms in small radial wheels are of flat bar wrought iron or mild steel (see fig. 20). In larger wheels ordinary bar can only be used if the power is comparatively small. In case of large power and of all feathering wheels, the arms are forgings, formerly of iron, but now always of steel, as being cheaper and easier to obtain (see fig. 19) ; the whole arm can be smithed from one piece instead of by welding on the brackets for floats. Let N be the number of floats on a wheel and each float be sup- ported by a pair of arms whose breadth is b and thickness t. Then, if the framework of the wheel is properly designed and fitted so as to distribute the load, the resistance to bending of each arm =f(t x ft) 2 and — Total resistance = 2nf(t x 6 2 ). The shaft inner journal has to work against this same resistance, so that, if it be taken as P, Then P x — is the torsion on the shaft, and also equal to the bending moment on the arms at the centre. 7rd z The torque on the shaft = -^ x/; then p x ^ = ^A = 27^x& 2 ); that is, tx¥=^. -^r =-0982xd«x4 / 2 x 16 / Eatio of b to t at boss, 5, and near rim, 3*5 ; when there is only an inner rim the ratio of b to t outside the rim will be 6 to 7, and for mild steel ^ = 0-8. 86 MARINE PROPELLERS. It follows, then, that 6 2 x -t = O0982#x 0*8 or 6 3 = 0*393d 3 o b = d ^0-393. Therefore :— breadth of arms near boss = 0*732 x d ; thickness „ „ = 0*146 x d ; and breadth of arms near rims = 0'511 x d. It must always be borne in mind that the stresses on a wheel in a tug or other open sea-going ship are far greater in proportion to the driving power than is the case in ships running always in smooth water, for there is, in addition to the shocks from the waves, the liability of the whole available power to be transmitted to the one " lee " wheel, the other, or " weather " one, being quite out of water. Unless the floats have been designed for such an emergency the engine will race and then probably develop the full power ; for other- wise, in such rough weather the engines would be probably "eased down." If d is determined to suit such conditions the I.H.P. in the formula, p. 84, instead of being one-half, will be three-quarters the whole power developed. Consequently, by taking d as a basis for calculation, the wheel will resist to the same extent as the shaft. It may also be noted that for smooth-water steamers a small factor f of safety is sufficient. In these cases ^ may be taken at 0*65 to 0-60 instead of 0'8. Paddle-wheel rims are of flat bar iron or mild steel in section, depending on the pitch of the arms and the size of the parts of the arms near them. In the old wheels, where the floats were closer to- gether than obtains now and the number of floats was consequently large, the rims were generally of about the same section as that of the arms ; in fact, small radial and feathering wheels were often made of one section throughout. In the case of a modern feathering wheel with two rims the following holds good : — Breadth of outer rim . . . = 0'40 x d „ inner „ =0'40x^ Thickness of outer rim . . = 0*08 x d „ inner „ . . . = 0*10x^ Inner ring, when no outer, breadth = 0'50 x d ,, ,, thickness — 0*14 x d Diameter of bolts throughout . = 0*12 x d, double rims. n ■> • = 0*1 5 xd, single rims. DIMENSIONS OF PADDLE WHEELS. 87 Tie bars are placed diagonally from the inner rim between each pair of arms to the flange of the boss, and cross-bars to arms behind the floats. Diameter of diagonal ties = 018 x d + £ inch. „ cross-bars = 0*18 x d + f inch (made hollow) Fig. 22.— Details of Feathering Gear of a Paddle Wheel. It need hardly be said that after making these calculations for a proposed wheel the nearest standard section sizes must be chosen and dimensioned accordingly on the drawings. Feathering gear as now fitted to wheels differs very little from that found on the Morgan wheel as designed by Elijah Galloway seventy years ago. There have been large numbers of other methods of effecting the feathering, some of which were as ingenious, some as MARINE PROPELLERS. simple, and some even as effective ; but none of them have survived, because none possessed all these qualities in the same degree as Galloway's invention, which is shown in fig. 19, the feathering of the lioats being effected in that case by their levers being connected to and revolving about a pin fixed to the sponson beam of the paddle box set eccentrically to the paddle-shaft axis. (See fig. 22.) This method is a very convenient one, as it takes the gear to a place where it forms no obstruction. Unfortunately, however, the eccentric pin is affixed to something which at all times is liable to spring and never forms the rigid base desirable for it, for it sometimes s i/,./, mam W gg«g D ^s Fig. 23. — Details of Float Bearing, etc., of a Feathering Wheel. gets so badly bent or sprung by contact with a quay wall or dolphin as to damage the feathering gear and put the wheel out of action. It was therefore arranged many years ago, and by some engineers adopted as their general practice, to fit an eccentric to the outer end of the shaft main bearing of such a diameter as to permit of feathering sufficiently and clearing the bearing and its base, as shown in fig. 18. In this case the frictional resistance of the large hoop is necessarily greater than that of the pin carrier of the older method. It can, however, be lubricated with oil and attended to in a way that was not possible with the pin on the sponson beam. The working parts of the feathering gear of a paddle wheel should DIMENSIONS OF PADDLE WHEELS. 8 9 have the pins cased with hard bronze and the holes in which they fit bushed with lignum viUe. (See fig. 23.) The feathering pin should be treated in the same way, as it is always being drenched with water, so that no effective oiling is possible. White metal bushes (Fenton's) with hard steel pins can be employed satis- factorily, especially in sandy or muddy water. Dimensions of Float Fittings. — The diameter of float gudgeons = O'l x breadth of iioats + 1 inch. Length of bush for gud- geon = 1*40 x diameter of gudgeons. The diameter of radius rod pins = 0*50 x diameter of gudgeons. The king rod should have a double jaw support to its pin and the float gudgeons strengthened to take the load of driving the feathering gear. The locus or path of the paddle float "centre 15 is shown in fig. 24. The position of the float or the angle it makes with the vertical at each point is clearly seen throughout the whole course, but the most im- 90 MARINE PROPELLERS. portant and interesting period is that during which the float enters, passes through, and emerges from the water. It is evident, on examining the diagram, that the aim of some of the older engineers to have a float vertical throughout that stage was futile, and such an arrangement would have detracted very much from the efficiency of the propeller. To enter and leave the water without shock is necessary to efficiency, and this could only be accomplished by causing the float to be as nearly tangential to the locus as possible at those points. The path is easily constructed by remembering that while the wheel is turning round on its centre it is advancing with the ship ; and since the tangential velocity is at a higher rate than the horizon- tal advance of the ship, there will be the loop at the immersion period, due to the differences in direction and velocity, so that the horizontal difference in position of a point on the path is the measure of the acceleration imparted to the water in the time taken in effecting the movement. CHAPTER VII. HYDRAULIC PROPULSION: INTERNAL PRO- PELLERS AND JET PROPELLERS. As early as 1698 Savery patented his engine, whereby large quan- tities of water could be drawn into it and expelled at a comparatively low velocity with a moderate consumption of steam by means of the simplest of appliances. It is true that it was necessary to have a boiler capable of producing steam at a pressure in excess of the atmosphere, but this he must have had under any circumstances. His engine, in fact, was the original and rudimentary form of the modern pulsometer, but it was not automatic. Not till 1729, however, was it recognised as a fundamental principle that the propulsion of a ship was effected by projecting a stream of water in the opposite direction to her motion, for in that year John Allen took out a patent (see p. 4) for propelling a ship with a machine like Savery's, but, strange to say, instead of steam he proposed to use what we now term " internal combustion " gases to obtain the pressure for ejecting the water. Had this principle of propulsion been known to and recognised by Savery there might have been a steamship, something like that shown in fig. 1, in A.D. 1700 ; and the self-propelled ship might have come into general use a hundred years earlier than it actually did. The simplicity of this means of propulsion will always incite a strong attraction for it, and especially will it be so to those men who constitutionally hate intricacy and complications, and who con- sequently condemn any deviation from it in design and construction, however good such may be both in principle and practice. A steamship propelled by such means was actually built and tried so late as 1876 by Dr Fleisher in Germany. She was named the " Hydromotor," and was of 105 tons, the length being 110 feet, the beam 17 feet, and the draught of water 62 feet. It is said that 92 MARINE PROPELLERS. she attained a speed of nine knots with the equivalent of 100 I.H.P. The two receivers in this case were cylindrical, lined with wood, and in each was a wooden piston floating on the surface ; they worked alternately as a pulsometer does, and ejected about 12 cubic feet of water per second at a velocity of 66 feet. Such success as was attained with this ship was apparently insufficient to encourage the building of others. It should be noted, however, in passing that Joseph C. Napier patented a very similar arrangement in 1817. Hydraulic propulsion was a very proper designation for this method of propelling ships, inasmuch as there was absolutely no mechanical instrument which might be worked on any form of propeller, however much it might differ from the paddle or screw. When, later on, ships were built and their propulsion attempted by such means, the stream issuing was small in section, and, with high velocity, it had another descriptive name, viz., jet propulsion, which was so called because the stream issued in jets of high velocity from either side of the ship at or near the stern, instead of through one large orifice at a low rate of flow. Allen's system of 1729, already referred to, was succeeded by that of his new patent of 1830, wherein he claimed to use a JSTewcomen engine to pump the water through the nozzles, he having no doubt found that his internal combustion machine, like so many new, immature, and untried inventions, could not be relied on to success- fully compete with the older and improved steam engine, either in economy or continuous good working. But even with the help of the Newcomen engine there must have been still other impediments, for little or nothing more is to be found in the records of Allen or his inventions. Eight years later, however, we find that Bernoulli proposed a system of jet propulsion very like Allen's, so that it may be that he had learned from London what Allen had done, and perceived in his experiment the germ of greater things which were to be achieved when the defects of Allen's gear and apparatus were removed, or perhaps he hoped that when modified under the brighter guiding light of Continental science they would be rendered less harmful and approach nearer to efficiency. James Ramsay's system of 1792 differed from Allen's by his proposing to use a centrifugal pump instead of a common reciprocating pump. He either could not get such a pump made, or thought so little of his invention as to abandon it for an ordinary vertical pump, for in the following year, 1793, he had built and tried HYDRAULIC PROPULSION. 93 on the Thames a boat of considerable size driven by a common pump 24 inches in diameter. The jet or orifice was, however, only 6 inches in diameter, so it is not very surprising that the speed of the boat was only 4 knots. This, however, was about as good as Symington's paddle boat of 1788. Had he tried a centrifugal pump and a 10-inch orifice the early history of the steamship might have been different. Internal propellers seems to be a proper and distinguishing name for the pump pistons and the centrifugal pump impeller of Ramsay. It certainly is appropriate to that of William Hales, who in 1827 to 1836 employed a vertical helix or screw turning in an enclosing cylinder as the pump to draw in water from the bottom and expel it through the stern. John Ruthven's system, as patented in 1849, was, however, the best known and most carefully thought-out and extensively tested of the various systems for obtaining motion without the use of an exposed screw or of the still more exposed paddle. Just as K P. Smith was by no means the originator of the screw propeller, nor really the first who had taken out a patent for a propeller of the kind, yet our veneration is due to him for the perseverance with which he acted and proved the merit of his inven- tion, and our thanks as well as his to Mr Wright for providing the means for doing it ; so Euthven, while entering the field so late with his plan for hydraulic or jet propulsion, was the first to succeed in having the system thoroughly tried on a large scale ; it was, however, seventeen years after the patent was taken out that the Admiralty was persuaded to build a special vessel of 1160 tons displacement, to satisfy the public whether Ruthven's claims were or were not so well founded as his backers maintained. Referring to the extract from Ruthven's patent No. 12,739, set out on p. 24, it will be seen that he claimed to use a centrifugal wheel with curved blades — in other words, a particular form of the centrifugal pump which Ramsay had claimed in 1792. Fig. 25, p. 94, shows a rough plan of Ruthven's propeller and pipes as supplied to H.M.S. " Waterwitch" in 1866. The terminal arms or elbows shown to be fitted on each side could be swivelled round so as to deliver the stream of water right aft, right forward, or at any intermediate position ; and as the gear for moving them round was operated from the bridge, she could be steered as well as have her speed regulated by the officer on the bridge without reference to the officer in the engine-room, much to the delight of the former, as all 94 MARINE PROPELLERS. the latter had to do then was to keep the engines running at full speed. The " Waterwitch " was put into competition with her sister ship H.M.S. " Viper," as may be seen by referring to the figures in Table VII., wherein are to be found recorded results of the best trial of each ship. An examination of all the trials, however, shows that there was really not much to choose between the two ships, and that the efficiency of both was wretchedly poor, as may be concluded by- comparing them with the performances of two single-screw ships of Fig. 25.— Ruthven's Hydraulic Propeller, as in H.M.S. "Waterwitch. similar size and shape made since, and with one single-screw naval ship somewhat older ; all three were of about the same speed. Two other ships, one a twin-screw and one a paddle steamer of higher speeds, are added, to show that even then their efficiency is much greater than that of the " Waterwitch/' in spite of the increased and somewhat high speed for their length. The " Waterwitch " speed was low even for 1866 ; to-day no kind of naval ship would be of any use whatever at so slow a speed. To quicken her machinery up sufficiently for the speeds of to-day would HYDRAULIC PROPULSION. 95 be an impossibility ; and if even 12 knots only had been required from the " Waterwitch " it could not have been obtained, except by increasing her draught of water to permit of the heavier boilers, etc., and the enlarged water-passages and chamber for the wheel. Table VII. — Trial of the Hydraulic Propeller H.M.S. "Water- witch " COMPARED WITH THOSE OF OTHER SHIPS OF ABOUT THE SAME SIZE. Dis- place- ment. Dimensions, Feet. Trial Results. Name of Ship. Draught D= x $' Tons. Length. Beam. of Water. Speed. I.H.P. I.H.P. Knots. H.M.S. "Water- 1101 162'0 32-0 11-17 9-3U 7G0 117 Hydraulic propeller, witch " Ruthven's. H.M.S. "Viper" 1180 162'0 32-0 11-83 9 '58 690 141 Twin-screw propeller ; expansive engines. H.M.S. "Chame- 113G lc5'0 33-2 13-50 10-21 584 190 Single-screw propel- leon " ler ; horizontal engines. S.S. "Bulania" 12S6 200-0 30-0 12-1 1U-80 661 228 Single-screw propel- ler; vertical triples. S.S. "Polo" 1UM5 1600 25-0 13-0 9-50 430 203 Single - screw pro- peller ; quadruple triples. T.S. "Ivy" 1080 204-0 34-0 10*3 13 20 1018 240 Twin - screw propel- ler ; vertical triples. P.S. "Jr. . . N" 1378 245-0 32-0 12-0 14-50 1684 227 Paddle wheel ; oscil- lating expansive. Thornycroft's system, based on that of Ruthven's, was a great improvement on it, consequent on the experience gained in the " Waterwitch " trials ; and the application of the science of hydro- dynamics by Sir John Thornycroft, of which he is a past master, warranted the Admiralty to engage in some further trials in 1881, when a torpedo boat was built, 66 feet 4 inches long by 7 feet 6 inches beam, and a draught of water of 2 feet 6 inches only ; the displacement was 14*5 tons and she was fitted with the Thornycroft impeller, of which a plan is shown in fig. 26. It was driven by a compound engine having cylinders 8 inches and 14 '5 inches in diameter and 12 inches stroke, which developed on trial 167 I.H.P. The speed on a carefully conducted trial was, however, only 12*6 knots, which was low and a keen disappointment, seeing that a similar boat but 3 feet 4 inches shorter with 13 tons displacement ran at a rate of 17"3 knots with her engines developing 170 I.H.P. The efficiency of the latter was not particularly high, being only 0*5, but that with the centrifugal pump was only 0'254, as estimated by Mr Barnaby. 9 6 MARINE PROPELLERS. Notwithstanding this very low efficiency scarcely the last word has been said for jet propulsion, seeing that it possesses the advantage of having the propelling instrument so well protected, and the ship itself is in quite as good a condition for use as a sailing ship is for moving Fig. 26. — Thornycroft's Hydraulic Motor in a Torpedo Boat. under sails. It is, therefore, eminently suited for such purposes as propelling lifeboats and similar craft that knock about in rough weather near ships and landing-places. It should be also an admirable addition to the long-voyage sailing ship to enable her to pass through calm belts, and to get into and out of harbours, docks and estuaries, HYDRAULIC PROPULSION. 97 inasmuch as it does not in any way detract from her ability to sail well, and high efficiency of machinery is not of first importance. In 1888 Messrs E. & H. Green built for the National Lifeboat Institution a large lifeboat into which the hydraulic machinery of Thornycroft's make was fitted ; the boat was tried, when again it was found that the efficiency was lower than that of the screw-driven boat ; but, on the other hand, her efficiency as a lifeboat was sufficiently established to warrant them building others from time to time. The efficiency of the machinery, as measured solely by the ratio of the useful work done by it to that generated, is of small moment in a lifeboat, and is not always of prime importance in other ships. Perhaps in these days of scientific investigation accompanied by scientific phrases, a little too much importance is sometimes given to Fig. 27. — Bessemer's Hydraulic Propeller. them, especially when they dazzle the eyes of the public so as to blind them to the real faults of a system. For example, high efficiency without reliability in continuous good running is futile. High mechanical and steam efficiency is sometimes accompanied by great liability to breakdown, and often by costly upkeep and great wear and tear. Further, high efficiency may sometimes even be purchased at such a high price that capital charges form a very serious set-off to working charges. In the same year, 1849, that Euthven took out his patent the late Sir Henry Bessemer proposed an arrangement of internal propeller as shown in fig. 27, which literally consists of a screw like Ericsson's, fitted in a case and with channels leading the water direct to the blade zone and from it to the stern, on either side of the wheel or screw, the case being contracted gradually to suit the channels or tubes. Inside the wheel case is an egg-shaped centre chamber taper- 7 98 MARINE PROPELLERS. ing at each end so as to guide the water to and from the impeller ; it acts as and is virtually a boss to it. The long after-end tapers with and as far as the taper of the outer case. There appears to be no record of Bessemer having tried this ingenious arrangement or done anything with it beyond the model stage, for which, however, there need be no surprise, as at that time he had not acquired the very considerable riches (whereby he could have proved this invention) that he had later on. This propeller, especially if assisted with guide blades on Thornycroft's principle, having only short channels, might have proved more successful than Euthven's. But after all is said, however, it remains that Bessemer's design had the same damning features as are common to all those with an internal propeller ; and if it was subject to the same constructive criticism, it would cease to be an internal propeller ; for it may be urged with all propriety and truth that for efficiency with such a system it is essential to keep down the velocity of flow, inasmuch as the frictional resistance of the surface of the wheel chamber and the passages to and from it is one of the chief sources of loss of energy and consequent low efficiency. Increased efficiency would be obtained by making the passages of larger sectional area, which leads to the dis- advantages of taking up more room in the ship, as also of an increase in the weight of the machinery, metal passages and water required, to which may be added the loss arising from there being less space for cargo. But to counteract all these objectionable conse- quences, it may be urged that the inlet shall be made as close to the impeller as possible and the delivery channels be shortened to their least possible length. It is obvious, in fact, that the shorter these are. the higher must be the efficiency, and thus the maximum efficiency will be attained when their length is zero, that is, so that the impeller delivers directly to the sea. In plain language, the propeller acts best when there are no inlet or outlet passages at all. But as some propellers, if devoid of chambers and guides, may lose in efficiency tremendously, they must be replaced by the screw or other implement which requires no external assistance to render it efficient. The fact is that few people who have strenuously advocated the adoption of the jet propeller system for general purposes can have taken the trouble to ascertain some of the physical facts and figures involved in its use on board even the ordinary ship of commerce. To take a simple case. Suppose a passenger steamer, such as was employed a few years ago in cross-channel service, having a dis- HYDRAULIC PROPULSION. 99 placement of 1750 tons and capable of a speed of 17| knots on trial when the engines developed 4000 I.H. P. with an efficiency of 0*66, to be fitted up with a Ruthven installation instead of twin screws about 12 feet diameter running at 130 revolutions per minute. With twin screws E.H.P. = 4000 x 0'66 = 2640. Now, assuming the efficiency of the hydraulic machinery, etc., to be 0-26, then gross LH.P. = 2640 4- 0*26 = 10,154, an enormous increase in power, being 250 per cent., and such that the ship could not contain boilers enough for it. Even if, for the sake of argument, it be supposed that by improve- ments, etc., an efficiency of 0*45 could be obtained, and at present that seems impossible, then Gross LH.P. = 2640 -f 0*45 = 5867, a sufficiently alarming increase, and only just possible to be in- stalled with great care. The indicated thrust of such a ship will be 4000x33,000 2000 ' = 66,000 lbs. The actual mean thrust will be about 56,000 lbs. Then taking the usual symbols, and allowing real slip at 25 per cent., making A represent the combined area of section of passages ; and assuming the speed of the ship v to be 1773 feet per minute and the mean 591 velocity of the stream 2364, the acceleration is 591, or — =9*85 feet per second. ^ f A X 2364 591 64 \ Kc nr^n Then thrust = { ^ — x 6Q - x ^ = 56,000, or, A = 56,000 -^ (39-4 x 9-9x2) = 72 square feet. If there was a channel or tube at each side, the area of section would be 36 feet, or 6 feet square ; rather an appalling orifice as well as an undue appropriation of space in the hold of the ship. The velocity of flow through the passages will be at the mean rate of 2364 feet per minute, or about 23 knots. The resistance per square foot of surface will therefore be (^xl-25 lbs. = 6-6 lbs. IOO MARINE PROPELLERS. For each foot of channel the friction will be 6 x 4 x 6*6 = 158-4 lbs. The equivalent in I.H.P. of work done in overcoming this is 11*35 I.H.P. per foot of each channel, and as the resistance of these and their bends may be equal to 300 feet of straight length, then loss from this cause alone will be 3405 I.H.P, If so small a slip be objected to, let it be taken at 33 per cent. ; then speed of flow will be 1773 x - = 2660 feet per minute. Li The velocity of flow is now 26 knots, and hence /26 s2 friction = f — \ x T25, or 8*45 lbs. per foot. Taking, however, the effect of the higher acceleration of water in reducing its quantity, A = 56,000 -:- i 2 ^- x ^ x 2) = 42-7 square feet. \ 60 60 / Each channel then will have an area of 21*4 square feet and a side 46 feet or a periphery of 18 '4 feet. Then the frictional resistance of channel per foot = 18"4x 8*45 = 155-5 lbs. The I.H.P. =155-5 x 2660^-33,000 = 12-53. As before, taking the resistance as the equivalent to 300 feet of such a channel, then — Loss = 12 '53x300 = 3759 I.H.P. To estimate the power of a jet or hydraulic installation it is necessary to know the velocity of the " feed " stream, which is the water entering the ship to supply the impeller, to take into account the velocity of the delivery stream as it issues through the nozzles or apertures, as well as the quantity of water passed through them in a second of time. If the area of transverse section of the feed channel is to that of the delivery as the velocity of flow through the latter is to that through the former, then the feed water will enter the whirl chamber of the impeller at constant velocity, due to the motion of the ship. To it is then imparted an accelerating force so that it leaves the ship at a higher velocity than that of the passing water. If, however, the inlet is made large enough for the water to enter at the velocity v, being that of the ship, and the channel is gradually HYDRAULIC PROPULSION. IOI contracted, the acceleration of the water will be then gradual up to the impeller, where it receives its final impetus and may leave the whirl chamber at the high velocity V. The delivery channel may then be so tapered that its section at discharge through the side is such that the water is delivered at the velocity v. As a matter of fact, however, such physical conditions are imposed on the engineer that neither of these refinements can be carried out in practice. The inefficiency of the hydraulic propeller is, however, due to other causes besides that of friction in passages and case ; these seem to be inevitable with the system. In the case of the " Hydro- motor "' the active machine was practically a bad form of Newcomen engine, inasmuch as the expelling force was that of steam acting on the wooden piston and the atmospheric pressure forcing it back again with the water behind it — a single-acting condensing engine. And although the steam was supposed to exhaust to a surface condenser, no doubt a large proportion of it was condensed in the chamber. The efficiency of this arrangement would be therefore far below that of any ordinary modern steam engine. With the centrifugal pump or other internal impeller, the losses in the chamber from various causes are huge. The wheel or impeller has an enormous surface exposed to the action of the water, so that with its high velocity the frictional resistance is most serious. The same may apply to the whirl chamber, whose sides and ends must, resist the water largely and set up eddies, which are always so per- nicious in their action on all such instruments. Then, too, the flow of water into the chamber, or, as it is called, the feed, which is induced by, and takes from, the power of the pump, is checked on entering it, and the whole or greater part of its energy is virtually lost. Besides these drawbacks, the whole of the water has to be lifted above the level of the sea at the expense of energy which is thus completely lost; if, however, delivery took place at or below sea flotation line, there would be then a great loss, due to dragging nozzles and fittings through the water, in fact, greater than that expended in raising it. The whirl chamber itself seems in all the experiments to have been too small for free working, and the impression gathered from examining the plans is that the water would always undergo a kind of grinding process before passing on. The Duty and Efficiency of a Hydraulic Propeller.— If A is the combined area of section of the delivery nozzles, V the velocity 102 MARINE PROPELLERS. in feet per second of the flow through them, and v the velocity of the ship, also in feet per second, the weight of a cubic foot of sea water 69 lbs., and gravity 32. The quantity of water passed per second = A x V x 64 lbs. T , _ . A x V x 64 A w Tr Its mass is — = 2A x V. 32 The acceleration, or, as it it called, slip = V — v. The reaction, or equivalent of thrust =2A xV(V-d). The work done is therefore = 2Ax V(V — v)v ft. lbs. per second. The energy of the " race " will be A x V(V-v 2 ). These things being so, then — The work done in propulsion = the work done + energy of race - 2 A x V(V -v)v + Ax V( V - v 2 ). = AxV(V 2 -^ 2 ). The efficiency of the jet itself will be, then, — — L~ — ~~ = — — -. J J AV(Y z -v 2 ) Y + v Without the modifications of channel alluded to, the energy of the feed water is dissipated and virtually lost, and is represented by A x V x v 2 . This must therefore be added to the total work so as to make up the full amount expended in propulsion. The total work is then = 2AV(V-v)v + AV(V-^) 2 + AV x v. Simplifying this it is = A x V 3 . Under these conditions the efficiency of iet = -~ — — L J J A x V 3 _2(V-^ - v 2 ' Taking the particulars of H.M.S. " Waterwitch " as given by Sir William White, viz. Ax V = 154*7 cubic feet, (V-v) = 13'3 feet; V being 29 feet and v 157 feet, and the I.H.P. as 760, 2 x 13-3 x 157 29 2 Efficiency of jet = ' X1 '* X10/ = 0'5. Total efficiency of the system = useful work-rl.ELV. x550 2 x 154-7 x 13-3x15-7 760x550 -0-155. The resistance of the "Waterwitch" at 9*3 knots was about 4900 lbs. and the resistance H.P. = 140. Efficiency in this was 140-760 = 0184. HYDRAULIC PROPULSION. IO3 The efficiency of the (< Viper," calculated on the same lines, was only 0-23. The efficiency of the system as installed by Thornycroft in a torpedo boat was estimated by Mr Barnaby as follows : — Efficiency of engine was 077 „ pump „ 0*46 » jet „ 0*71 Total efficiency of system = 077 X 0-46 x 071 = 0*2514. It is unfortunate for us that no progessive trials were made with these ships. Even the old "half-boiler'* trial, if made with the " Water witch," would have given some good information on which to make further investigations, and perhaps have given some indications of the road to follow to avoid the heavy losses. CHAPTER VIII. THE SCREW PROPELLER : LEADING FEATURES AND CHARACTERISTICS ; THRUST AND EFFICIENCY. A screw propeller is an instrument consisting of a central hub or boss secured on the end of a shaft projecting through the end of the ship, to which boss are affixed two or more wings or blades, each of which is shaped and formed like the others and set angularly equidistant from one another with their centre lines at right angles, or nearly so with the axis of the shaft. The face or surface of the blades acting on the water is helical ; that is, it is formed of a part of one or parts of more than one helix. The helix is a surface developed by the revolution of a line about one end, which end travels or ad- vances lineally on an axis at right angles to it as it revolves. (See fig. 28.) A true helix is thus formed when the angular movement and the lineal advance are synchronous; that is, if L is the length of the one complete convolution of the helix and nxl = L, then an axial advance of I and an angular movement of 360° ~n are made in the same time. The pitch of a helix is the length of the axis or the distance travelled when a complete revolution of the describing line has been made ; or it may be defined as the distance between two parallel Fig. 28.— Screw Blade on a True Helix. THE SCREW PROPELLER. IO5 planes at right angles to the axis, which thus cut off a complete con- volution of a helix. The angle of the helix at a distance r from the axis is one whose tangent = pitch -^-2tt?\ The surface of the helix of radius R, pitch P, is Ex VP^ 2 R 2 = RVP 2 + 10R 2 . Common screw was the name given to the screws first used by F. P. Smith and for a long time after in H.M. service. The boss was cylindrical and comparatively small in diameter, the acting surface was a part of a helix cut by parallel planes a fraction of the pitch apart, latterly about one-eighth. Taking the length as I, then — • Surface of each blade of a common screw = ^ x R \/P 2 +10R 2 . As the common screw had usually two blades and sometimes more, each was a portion of a separate helical convolution cut off by the parallel planes. Developed Surface. — Such screws naturally had blades with very wide tips, and the actual acting surface which was fan-shaped was said to be the developed surface when laid out on a plane surface. Projected surface of a screw is that made by projecting or throwing the shadow of the blades on one of the transverse planes ; its area is often taken as the measure of the ability of the screw to make thrust. At one time there was a disposition to consider this surface only, but it is not convenient, as it varies with the alteration in pitch, whereas that of the acting surface of the blade itself remains the same. Diameter of a screw is that of the circle described by the outermost point of a blade in making a revolution. True screw is the name now applied to the screw whose whole acting surface of blade is part of one helix only. That is, the pitch is uniform throughout. Variable pitch screws are those whose blade surfaces are helical, but made up of portions of various helices, as the pitch varies from one part to another. Bennet Woodcroft claimed special virtues for the screw, the blades of which had a coarser pitch at the following part than at the leading edge (see fig. 29). He thought that the water would be thus gradually io6 MARINE PROPELLERS. set in motion sternwards and leave the screw at the highest velocity instead of starting it suddenly and traversing the blade at the same speed. With a slow revolution screw there would be in reality this theoretical gain, but with the screws of to-day, whose speed at the tip is 60 to 70 knots, it is scarcely to be expected that such a refinement as Wood- croft patented would make any difference. Still, there are plenty of engineers of the mercantile marine who seriously claim that such propellers do give better results in the steamers under their observation, while others seem equally Fig. 29.— Increasing Pitch Helix ( Woodcroft). convinced that there is really more virtue in a propeller whose pitch does not vary transversely, but which does increase from the boss to the tips ; others, again, even fancy a propeller with a coarser pitch at the boss than at the tip. In the naval service a true screw is the rule now, and it is that generally used in the mer- cantile service. Woodcroft's original screw was of one convolu- tion, so that the pitch gradu- ally increased from 10 feet at the fore to, say, 12 feet at the after or delivery end, AYW. Fig. 29 shows the line AYW, due to a pitch vary- ing from BL to BC, and is constructed as follows : The base-line AB repre- sents the circumference of the circle whose diameter d is that of the propeller at any point required ; therefore AB = wd. BC is perpendicular to AB and is the pitch at delivery, while BL is the pitch at entry edge of blade. Fig. 30.— Screw Blade Curved Forward, Griffiths' patent. THE SCREW PROPELLER. 107 Fig. 31.' — Screw Blade thrown back by making centre line Spiral on the Bed. Divide LC into the same number of parts (say 8) as AB. Then BK is the pitch at £ the distance from the leading edge, and AK is the angle made by the blade at that part, and so on for BJ, BH, etc. to BC. Draw QP, ES, TV, etc. perpendicular to AB; AP is then the eighth part of AC. Draw PS parallel with AD cutting RS at S v S t parallel with AE, Vj cutting TV at V 1? etc. etc., and XW parallel with AL. The polygonal line APVXW can be resolved into a curve, and if the figure, is then wrapped around on a cylinder of circumference equal AB, APVXW is the trace of the helix of increasing pitch required. But Woodcroft and others, as screws got shorter, made the leading half of the blade of one pitch, the following of another, and faired one into the other with the curvature shown in fig. 56. Griffiths and others since his time have shown a preference for a blade of which the acting surface is developed by the revolution of a curved line instead of a straight one; in Griffiths' case the generating line would be straight for a half to two-thirds of its length and then bent towards the bow. (See fig. 30.) Very many engineers prefer the surface to be developed by a straight line at less than a right angle with the axis, so that the surface is a portion of what is called by mechanics a " V thread. 1 ' (See fig. 32.) In all these cases every portion of the blade is that of a true helix, Fig. 32. — Screw Blade thrown back by Conine; the Bed. io8 MARINE PKOPELLERS. as there is no variation in pitch. Some makers of screw propellers who like to have the blade ends thrown away from the ship, do so on a true and common helix by curving the centre of the blade on it so that the tip is lower than even the end of the boss (fig. 31). Others, again, make a bent blade like the Griffiths, but make the generating line at less than 90 degrees with the axis. (See figs. 32, 33.) Pitch ratio is the ratio of the pitch to the diameter of the screw, P-i-D; the exigencies of to-day to suit steam turbine motors re- quire the pitch ratio to be often less than 1 '0. Experi- ence with screws driven by reciprocators has shown in the past that a low-pitch ratio was not satisfactory, and generally resulted in a low efficiency of screw ; it was also often accompanied by, if not the cause of, negative slip. On the other hand, screws with pitch ratios of 1*2 to 1'5 invari- ably give satisfactory results accompanied by moderate positive slip. Surface ratio is the ratio of the area of the acting surface of a screw to that of a circle of the diameter of the screw. Fig. 33. —Screw Blade Curved Back. With two-bladed screws Common Surface ratio 0*275 to 0*320 two Griffiths u 0-260 to 0*290 three ., Common )j 0*350 to 0*400 three „ Griffiths >) 0*335 to 0*380 three Admy. leaf j) 0270 to 0310 three „ Circular j> 0*370 to 0-400 three Broad tips n 0-420 to 0*440 four Common i> 0*388 to 0*450 four Admy. leaf j> 0*330 to 0-380 four ,, Circular j' 0*500 to 0*520 four „ Broad tips >> 0*530 to 0*550 four j Mercantile ordy 1 Square tip } 0*290 to 0*400 THE SCREW PROPELLER. IO9 In these pages it is proposed to use always the following : — P the pitch and D the diameter of the propeller in feet. V as the velocity of the propeller in feet per second. v „ ,> „ ship S „ „ „ propeller in knots. 5 „ „ „ ship E as the revolutions of the propeller per minute. Velocity of propeller is the pitch multiplied by the number of revolutions. Px R V = feet per second, or S= T0F3 knots ' Resistance of the ship, tow rope, is that due only to the resistance of the ship to her passage through water, as is the case when being towed without a propeller by another ship at a consider- able distance from it. Such resistance is measured by the tension of the tow rope, as was done by Dr William Froude in his experiments with H.M.S. " Greyhound." He found that the resistance of the "Greyhound " was 0*6 ton at 4 knots, 1*4 tons at 6 knots, 2 '5 tons at 8 knots, 47 tons at 10 knots; at 12 knots 9*0 tons, which was excessive, as this ship was designed for 10 knots only. A much larger ship, the " Merkara," which had a resistance of one ton at 4 knots, had only the 9 tons at 12 knots. The resistance per square foot of wetted skin was 1*396 lbs. at 10 knots with the " Greyhound." With modern ships at 10 knots the resistance is 1 lb. per square foot of immersed skin when fresh painted to 1*4 lb. in ordinary clean condition. (Vide Table V„ Chapter III.) Augment of resistance is the increased resistance caused, first of all, by the increase in velocity of the water past the hull by the suction of the propeller ; and, secondly, by the diminution in hydraulic head or pressure at the stern due to the action of the screw. Screws of large diameter and fine pitch have a special tendency to aggravate this loss of " head." Total resistance of the ship is the resultant of the resisting forces thus created and made active ; it is these that the propeller thrust has to overcome, and thrust is equal to them when the ship is moving at uniform speed. Resistance of propeller, if of no pitch, is that due to the skin HO MARINE PROPELLERS. friction on moving through the water and to the resistance due to the bluntness of the edge and the form of the body forced through the water. The energy absorbed in overcoming this is all lost, and amounts to a large fraction of the total energy imparted to the screw by the engines ; for at a velocity through the water of 10 knots each square foot of surface like that of a screw propeller exerts a resistance of 1\ lb. (in ordinary bronze blades or those merely painted). Highly polished blades and those having fresh enamel paint will offer less — probably only 1 lb. per square foot. The back of the blade may not set up so great an amount of friction, but it will not be materially less, taking into account its shape, etc., so that it is usual, in calculating frictional resistance of screws, to take twice the acting surface and assume that the mean resistance is based on an allowance of 1\ lb. at 10 knots at any part of it in its passage at that. The resistance will vary, of course, as the square of the velocity, so that those propellers whose rate of motion near the tips is 60 knots will resist at the rate of 45 lbs. per foot at the tips, or, taking back and front, 90 lbs. per foot of acting surface. The other resistances named will vary roughly with the number of blades, but inversely as the breadth of blade, and in everyday practice with good screws may be taken at 5 per cent, of the total skin resistance for each blade. That is, if there are four blades, the total resistance is 1*2 multiplied by frictional resistance. Frictional resistance of a screw blade may be found by the following simple methods : — Fig. 34 shows the outline of the developed surface of half a blade whose figure is symmetrical about GB. The propeller is moving at a uniform rate of revolution so that BC represents the velocity through the water at the tips to a convenient scale. That is, the velocity per revolution at B and at any intermediate point is v 1 = VpitchHOirS) 2 , cl being the diameter at any point taken. If BC, etc., GK, represents on a convenient scale the velocities at B, etc., G-. A curve drawn through C, etc., K will permit of the velocity being ascertained at any intermediate points by taking the intercept between BG and CK at these points. The resistance per THE SCREW PROPELLER. I I I square foot may be calculated at three or four points by the rule y = l*25 ( — J lbs. and a curve GD set up in the same way so that intercepts will give the resistance at any intermediate points. Now, taking narrow strips of the blades at three or four stations and multiplying by the resistance at these stations and doubling the result to allow for the blade backs, a curve HE is obtained so that intercepts again give the resistance at various stations, and the area is the measure of the total resistance of one blade. Proceed, then, to multiply the resistance of the strips as obtained Fig. 34. — Showing Curves of Friction, Resistance, etc., of a common Screw Blade. above by the space moved through by them in a minute, and the work absorbed in turning that blade is measured by making a curve HF by means of a few of the ordinates so found as before. Intercepts between HF and GB give the work absorbed in moving those strips through the water, and the area GBFH represents the total power in ft. lbs. absorbed in turning that blade through the water. Dividing it by 33,000, the horse-power required to overcome h is obtained. Figure 34 represents the equivalent resistance of two of the four blades of H.M.S. " Amazon," and fig. 35 is that of one of the two I 12 MARINE PROPELLERS. of the Griffiths screw which replaced it and gave so much better results. The ill effect of the broad tip is seen at a glance, as are also the losses arising from excessive diameter, for, by taking six inches off each tip, the resistance is in both cases very much reduced, especially so in the case of the four-bladed screw. Froude found the efficiency of the " Greyhound's " machinery to be exceedingly low, and attri- buted it chiefly to engine resistance, whereas it was largely due to the absurdly large diameter of the screw, it being 12-33 feet diameter with 52 square feet of surface ; whereas the " Eattler," of similar size and power, had a screw 10 feet diameter with only 22 - 8 square feet of surface, which elaborate ex- periments years before had shown to be sufficient. Moreover, the " Eattler " had a speed coefficient of 224 against that of 142 of the " Greyhound," which ought to have opened the eyes of the authorities in 1865. With the high speed of revolu- tion necessary for the efficient working of turbine motors, as also for the speed of revolution possible with modern reciprocators, especially the enclosed variety with automati- cally forced lubrication, propellers of small diameter are absolutely necessary for safe running, while to prevent cavitation the blade area must be relatively large. Hence we find the modern propeller is gradually getting nearer and nearer in width of blade to our old friend the common screw of sixty years ago, and differs from it now chiefly in its having nicely rounded corners instead of the rigidly square ones of our grandfathers' time. Fig. 36 shows one blade of H.M.S. " Eattler " of 1845 ; the dotted line is that of a blade of a modern turbine motor steamer. Now, although the difference in blade is small to look at, the action when at work is very different. Those corners of the old screws caused violent vibration at high speeds ; but when they were cut away there was a very marked improvement. Fig. 35.— Showing Curves of Friction, etc. . of Griffiths Screw. THE SCREW PROPELLER. 113 Frictional resistance of a screw propeller may be cal- culated with a close approximation to the truth by taking the velocity at the tip and the total area of acting surface, using multipliers in both cases deduced from the close calculation of it with screws of different types. Let S be the velocity of the blade tips in knots per hour. Let R be the revolu- tions per minute. Let D be the diameter in feet. Let P be the pitch of screw in feet. Let A be the area of acting surface in square feet. The resistance of a square foot is assumed to be 1| lb. at 10 knots. S= VP 2 + (7tD) 2 xRx60^-6080 = Rx JW+JwD^^-101'3. _S\ 2 vlO/ Fig. 36. — Modern High Revolution Screw compared with that of H. M.S. "Rattler," 1845. lbs. Resistance per square foot=l"25 Resistance of screw = 2 Ax 1*25 ( — J x/lbs. For a common screw . . . /= 0634 For a fantail-shape screw . . /= 0*581 For a parallel blade . . . /= 0*550 For an oval 7=0*520 For a leaf shape . .... /= 0*450 For a Griffiths 7=0*350 The horse-power absorbed in overcoming the frictional resistance may be found now by multiplying the resistance by the space in feet moved through in a minute and dividing by 33,000. 114 MARINE PROPELLERS. The mean space moved through by the blade surface from tip to boss of an ordinary propeller = 0*7 X distance moved through by the tip. u Sx 07x6080 K nQQ Hence mean space = — = 70*9b. Then I.H.P. expended = 2Ax 1'25 (^Yx/x70-9S-=- 33,000 _ AxS 3 x/ 18,612 ' Example. — A screw 12 feet diameter, 15 feet pitch, has 42 square feet of surface and moves at 130 revolutions per minute (three leaf blades). Here S= V225 + 1440x 130-M01-3= 1Q * 3 ' =52*4. t* ■ *• i • * tip 42 x52-4 3 x 0-515 _ _ . Fnctional resistance screw H.r. = 0C10 = 166-8. lOjblz Edge resistance here will be 3x5 per cent, or 15 per cent, of 166*8 = 25 H.P. Then total resistance of screw = 166*8 + 25 = 191*8 H.P. The thrust of the screw is the resultant of all the pressures on the screw acting in a direction parallel to its axis and applied through the screw shafts to the thrust block. There is no method of calculating this with any degree of certainty from purely theoretical considerations ; therefore for H.M. ships and other ships of high speed and great importance, the thrust and other characteristics of any proposed screw are ascertained by experiments with models in a tank fitted with special apparatus for towing and accurate recording instruments. This is, of course, both an expensive and tedious method, and a luxury at present only enjoyed by a few; moreover, such experiments, while being very interesting, are not necessarily con- clusive as to what may be expected from the full size screws. Indicated thrust was a term introduced by the late Dr William Froude as a means of expressing a value for the thrust of a screw at various speeds by which its performance could be analysed and com- pared with that of other screws ; and, although he did not claim that the thrust so estimated was the actual thrust, it was not difficult to deduce from it something approximate to it. It is clear, however, that even this claim must be taken with reserve, seeing how misleading may be the results of such calculations with screws of abnormal THE SCREW PROPELLER. I I 5 proportions, as will be shown later on, unless they are treated with caution. Froude's rule was as follows : — T ,. . ,,, LH.P.x 33,000 . , Indicated thrust = — v — - in pounds. For example, suppose a ship to travel at 15 knots with the I.H.P. 3000, and the apparent slip of screw 12J per cent. ; Here v = 15 x 6080-^60 = 1520 feet per minute. Slip=V-*> = |, 8 8 then V = - v = = x 1520 = 1737 feet per minute. t r * a ^ 4. 3000x33,000 _ ... ,, Indicated thrust = ^ ' — =57,000 lbs. I * o ( Practically this means that the power delivered by the screw is £ of the I.H.P. With the 20 per cent, slip of the older engines Froude's suggestion did give a good approximation to the real one ; but with the machinery of to-day with 20 per cent, slip it would not be nearly so correct. Froude also very properly pointed out the directions in which the power generated in the engine is expended, and drew up a sketch balance-sheet as a model for engineers to follow in making up their accounts. First of all, the engine itself requires a certain amount of the gross power it develops to be absorbed, as it were, in overcoming its own resistance in the way of friction, and for losses due to the inertia of the moving parts. Secondly, the working of the air, feed, bilge and other pumps make further inroads on the I.H.P. Thirdly, the resistance of the shafting and thrust requires another portion. The balance is the power transmitted to and put into the screw, and may amount to 90 per cent, of the I.H.P. or even more, with good modern engines having only air pumps to drive and all the working parts carefully fitted and thoroughly lubricated. In the older ships with circulating, feed and bilge pumps, in addition to the air pumps, worked by the main engines and having surface condensers and work- ing parts well made throughout, the balance was about 85 per cent. ; I I 6 MARINE PROPELLERS. while with horizontal jet condensing engines and the bearings and guides such that there was always a tendency to run hot, the balance was often as low as 70 to 80 per cent. If N.H.P. be the gross power imparted to the propeller, then — Efficiency of the engines and shafting = N.H.P. -j-I.H. P. Of this power a considerable portion is required to overcome the resistance of the propeller itself, as already shown, and T.H.P., the balance, should be employed in making " thrust"; but it may be that some more power is being wasted in disturbing and making eddies in the water and dissipating its energy in other ways than that of " projecting a stream of water in the direction opposite to that of the ship's motion/' The net balance, in any case, is employed in making thrust, and the power may be called Thrust Horse-power or T.H.P. Then: — Efficiency of the propeller = T.H.P. ~ N.H.P. Total efficiency of machinery and screw — T.H.P. -^I.H.P. But, as already shown, the screw when working is always the cause of an increase to the ship's resistance ; in some cases that increment is a most serious one ; hence if the tow-rope resistance of the ship is R and the H.P. corresponding to it is Tr.H.P., then : — True efficiency of machinery \ propeller ; and the ship = Tr. H.P. -f LH.P. The resistance of the ship can generally be estimated from the information obtained from trials at low powers when the wave-making is not serious enough to be taken into account, and when there could be no cavitation or other disturbing causes at the propellers, or by assuming that each foot of wetted skin has a resistance of 1 lb. to li lb. at 10 knots. General efficiency of propeller is a thing hardly attainable in theory, inasmuch as it is evident that the screw propeller whose pitch and surface are the best for a speed of 20 knots cannot be the propeller of maximum efficiency at 15 knots. It is therefore a very debatable point whether it is better to design the screw to suit the maximum contract trial speed, or for the service speed in the case of mail and passenger steamers, or for the maximum speed consistent with naval service conditions in H.M. Navy for warships. The Actual Thrust of a Screw Propeller. — In the early days of screw propulsion attempts were made to measure it by means of a THE SCREW PROPELLER. I I 7 dynamometer applied to the screw-shaft end so as to take the whole thrust. With the shaft geared to the engine shaft, as was then customary, this method was easily adaptable, but the "readings" were such as to render any results deduced from them to be always taken with reserve, and they were generally open to extreme doubt ; it is therefore not surprising to find the records published by the Admiralty of trials of H.M. ships with the dynamometer quite inconsistent one with another ; and only those taken on board H.M.S. " Battler," when it may be presumed the instrument was new and carefully calibrated, are worthy of careful analysis. Since those days other attempts have been made to measure the thrust of a screw propeller of a size beyond the model stage — notably those of Mr Yarrow with a torpedo boat, as published by him in a paper read before the Institute of Naval Architects. His experiments were ingeniously devised and most carefully carried out, so that all disturbing elements were minimised and most of them eliminated. Mr Yarrow's experiments made in 1883 with a 60-ton torpedo boat are very interesting and instructive. He carried out a series of trials at speeds varying from 9 to 15 knots. (See p. 237.) (1) Propelled by her own engines, from which the indicated horse-power was noted as I.H-P. (2) The thrusts in pounds and the horse-power corresponding were also carefully measured and noted as T. H.P. (3) After removing the propeller the boat was towed by another and larger boat, so as to be as free as possible from any disturbance due to the latter ; the tension on the tow-rope was carefully measured and noted and the horse-power deduced, denominated Tr.H.P. It was found that the maximum value of Tr.H. P. -=- LH.P. — that is, the maximum efficiency of the whole — was 0672 at about 11 knots while at 9 knots and 15 knots it was 063. The maximum efficiency of engines and propeller as shown by T.H.P. -f-LH. P. was also at about 11 knots and the value 0*852; at 15 knots it was only 0'733, while at 9 knots it was 0'800. The maximum efficiency of the propeller as taken by the ratio TR.H.P. to T.H.P. was 0*860 at 15 knots, while at 9 knots it was 0786 and at 11 knots 0*790. Mr Yarrow expressed some doubt as to the perfect accuracy of the I.H.P., and judging from some observations made with quick- running engines that have come under the author's notice, there is I I 8 MARINE PROPELLERS. reason to suspect that Mr Yarrow was right, and that the power is quite 10 per cent, below what it should have been had the instru- ments been fitted direct to the cylinders. Making this addition, and deducting a fair allowance for engine friction, etc., it would appear that the losses at the propeller itself were 1*5 Ii.P. at 10 knots, 4*1 at 11 knots, and as much as 70 at 15 knots. Taking these figures, and calling the net horse-power delivered to the screw N.H.P., then T.H.P. -fN.H.P. is a maximum at about 10 knots, and is 0*759, while at 15 knots it was only 0*61. Under these conditions the screw losses at 10 knots were only 5 per cent, of the N.H.P., while at 15 knots they amounted to 29 per cent. The augment of resistance at 10 knots was 20 per cent of the K.H.P., and at 15 knots only 10 per cent, of the N.H.P. Mr R. E. Froude, whose researches and extensive experience in screw propeller experiments not only entitle him to a most respectful hearing, but to the gratitude of every marine engineer and naval architect for the most useful and invaluable information he has given them without stint ever since he succeeded to the place so long occupied by his honoured father Dr William Froude, has propounded a formula for calculating the actual thrust of a propeller and given the multipliers deduced from carefully made model experiments, so that it can be applied to any screw. He states : — The analysis, then, of the series of thrust and efficiency curves yielded by the experiments on the individual model screws was in the first instance based on the following simple formula for thrust in terms of revolutions per minute : — T = aR 2 -6R . . . (1) Where T — thrust ; R = revolutions per minute ; a = a coefficient depending on dimensions, etc., of screw; b> one depending on the speed and pitch. This formula embodies the following idea, which under certain ideal conditions would be theoretically correct. In any screw revolving in still water at various rotary speeds without axial advance the thrust will, of course, be proportional to the square of the rotary speed. This fact is expressed by the term «R 2 of the formula, represented by the ordinates of a parabola ABCD. If now we suppose the screw, while still revolving as before at various rotary speeds, to have a definite forward axial linear speed of advance V as well, there will, of course, then be a certain definite rotary speed THE SCREW PROPELLER. I 1 9 (revolutions per min. = E , say), at which the thrust will bo zero, below which it will be negative, and above which it will be positive, but of decreased amount. And the formula expresses this decrease of thrust in terms of revolutions per minute by the negative term &E, represented by the ordinates to the straight line ACE, cutting the parabola at C, namely, at E , the revolutions of zero thrust. Thus the straight line ACE becomes in effect the zero line for the curve CD, regarded as the thrust curve for the speed of advance V ; and, similarly, the same parabola ABCD may be made to furnish the thrust curve for the same screw at any speed, by drawing accord- ingly the sloping line ACE which represents the term ( — Z>E) of the formula. This formula expresses the thrust curves of experiment, in general, remarkably well ; since, therefore, a new reduction was in any case necessary, this formula was unquestionably the right basis for at any rate the primary analysis of the results. Thrust. — Continuing the study of the formula, it will be seen that if, as is most convenient for such analysis, and as has been done in this case, we take as a conventional measure of the pitch the travel per revolution at the revolutions per minute E (of zero thrust) so that and, again, observing that for zero thrust at revolutions = R we must have aR 2 = 6E , we can eliminate the coefficient h in terms of a, V, and P. And, assuming the coefficient a to have been correctly obtained for a screw of specific design and unit diameter (diameter = D = 1), we obtain from equation (1) above the following two alter- native equations for thrust : — T = ? D2V2 (T^s? • ' • • (2) T = «D 4 E 2 S, (3) where p = -jy or pitch ratio, and S — slip-ratio as ordinarily reckoned, viz. (E — E ) -r E. The former of these, expressing thrust in terms of speed and slip-ratio, is perhaps the most intelligible ; while the latter is often more convenient for computation. To enable the thrust for given speed and revolutions to be 120 MARINE PROPELLERS. calculated by either of these formulae for a propeller of any given design and dimensions, it only needed to determine the coefficient a as affected by difference of design, the principal elements in which may be taken to consist in (i.) pitch ratio ; (ii.) type, and blade width proportion. It was found that the effects of these two principal elements might be taken as independent of one another, and that, as regards (i.) a might be most correctly taken as proportional to _£?(p + 21). As regards (ii.), the value T~9il > wn ^ cn ' as J us ^ seen > * s constant for varying pitch-ratio, was taken as the expression for the " blade- factor " B ; the purpose of which is to denote what may be called the thrust capacity of the propeller, as dependent on type, i.e. whether three-blade elliptical, three-b]ade wide-tip, or four-blade elliptical ; and within each of these types, on width proportion of blade. The value of this blade factor B as obtained from the experiments, and as dependent on these variants, has been indicated by the ordinates of three curves respectively proper to the three types just mentioned, on an abscissae scale representing blade width proportion, as indicated by " disc area ratio," namely, ratio of total blade area to disc area. At the same time, for a final test of the formula of equation (1) as accurately expressing the variation of thrust with revolutions, the thrust values of all the individual propellers, at a series of slip-ratio values, were carefully compared with thrust values calculated by formula ; and on this information the thrust formula was corrected by multiplying the right-hand side by 1*02 (1 — '08 S). Making this correction, and also substituting for a its value in terms of the blade factor B just referred to, equation (2) becomes the final thrust formula, as follows : — T _ D2Y2xB ^ + 21 ^ 1-Q2S(l-Q8S ) V X p X (1-S) 2 W To facilitate calculations, a curve was computed expressing the last factor (involving S only) as an ordinate { = y) to a base ( = #) expressing revolutions and pitch relatively to speed, as indicaiive of the slip ratio S. This curve is commonly called the "%y" curve. Conveniently for ship screw calculations, the numerical coefficients used in the computation of this curve were chosen for expressing, not thrust, but "thrust horse -power" (or T,JLP.) = H; speed = V, in THE SCREW PROPELLER. 121 knots; revolutions = E, in hundreds; diameter = D, in feet. We thus get as the expressions for x and y as follows : — x = R/?D V 1-0133 ' : 1-S. H B<> + 21) D 2 V : = -0032162 s ? ""^ 9 S) " (5) (6) Table VIII.— B Values. Disc area ratio . ■3U ■35 •40 •45 ■50 •55 ■60 •65 ■70 •75 •80 Four blades, elliptical ■0978 ■1020 •1050 •1070 ■1085 •1100 '1112 •1124 •1135 '1147 ■1157 Three blades, wide tip ■1045 ■1097 •1126 •1148 •1166 ■1182 ■1195 •1207 •1218 •1230 1242 Four blades, elliptical ■1040 ■1106 ■1159 •1197 ■1227 ■1249 •1268 ■1282 ■1294 •1306 ■1318 Dr W. Froude's analysis was proposed by him as a means whereby the efficiency of a screw could be determined, and its per- formance compared with other screws was arrived at as follows :— He assumed that at each revolution the mean pressure on the pistons multiplied by twice the stroke was equal to the thrust multiplied by the pitch of the screw. Let A be the area of L.P. cylinder or cylinders in square inches, L the length of stroke, P the pitch in feet, and p the referred mean pressure. Thrust x P =p x A x 2L, or thrust = Multiply both numerator and denominator by K, the number of revolutions per minute, then ^ , »xAx2LxE thrust — - ^ — r> ■ PxK Now pxAx 2Lx R = I.H.P. x 33,000, and, substituting, then I.H.P.x 33,000 thrust = : PxE This he called indicated thrust, and by calculating it for the various speeds taken on progressive trial, and using the results as ordinates to the speeds as abscissae, a curve drawn through their ends shows the indicated thrust at all speeds (fig. 37, p. 122), and the trend of the curve indicates whether there is a falling off or abnormal increase in thrust as the speed increases. If the latter, it is evident the screw is doing its duty, but the ship is not responding ; if, on the 122 MARINE PROPELLERS. other hand, the curve falls away the screw is too small for the power put into it, or is using the power in some other way than that of producing thrust. Fronde's curve of indicated thrust (fig. 37) is constructed as follows : — On the base line OX take points B, C, and D corresponding to the three different speeds of progressive speed trial. From the data given, calculate the indicated thrust by the following rule : — Indicated thrust = .— ^ — '—^--—^ _ pitch x revs, per mm. Draw BE, CF, and DG perpendicular to OX and proportionate to M Speed in Knots X Fig. 37. the indicated thrust thus calculated, on a convenient scale. Now take A so that OA represents the speed at the slowest rate of steam- ing observable, calculate the indicated thrust as before, and erect AH as representing the amount to the same scale as the others, and through HEGF draw a curve and continue the same if possible, so as to cut OY at T. As it is generally, in practice, impossible to get so slow a speed as to enable the prolongation of the curve HEGF to be made with accuracy, Froude suggested that the point T might be obtained by taking a point M between OA so that — OM_0-S7 OA 1-0 " THE SCREW PROPELLER. I 23 Draw MK perpendicular to OX, and touching the tangent drawn through H at the point K. Through K draw KT parallel to OA, cutting OY at T, which is then the point through which the curve should pass. If such a curve is produced it will not pass through the origin but cut the ordinate some way above it; this is the measure of the thrust at no speed, and it was thought by Froude to indicate " the equivalent friction of the engine due to the working load/' Froude differentiated and laid down that " when decomposed into its constituent parts, indicated thrust is resolved into several elements which must be enumerated " as — (1) The useful thrust or ship's true resistance. (2) The augment of resistance, due to the diminution which the action of the propeller creates in the pressure of the water against the stern end of the ship. (3) The equivalent of the friction of the screw blades in their edgeway motion through the water. (4) The equivalent of the friction due to the deadweight of the working parts, piston packings, and the like, which constitute the initial or slow-speed friction of the engine. (5) The equivalent of friction of the engines, due to the working load. (6) The equivalent of air pump and feed pump duty. It is probable that (1), (3), and (4) are all very nearly proportional to the useful thrust; (6) is probably nearly proportional to the square of the revolutions ; (5) is constant at all speeds — that is, the power absorbed in engine friction varies with the revolutions. A practical rule for estimating thrust of a screw has been devised by the author as the result of analysing a series of trials made at various times of actual ships by various engineers with such accuracy, both of observation and calculation, as to command respect and permit of their use in an estimation of the actual thrust. Mr R E. Froude, like others, is of opinion that thrust varies with the acting surface ; and, generally speaking, if it is understood that there is no other variable function, such as pitch or diameter changed, this is fairly true ; but it is still not absolutely true, for the effect of adding a square foot of surface near the tips will be very different from what would follow from making the same addition near to the roots of the blades. The broad statement is therefore by itself absolutely wrong and misleading, for it is manifest that a screw 124 MARINE PROPELLERS. 10 feet diameter having a surface of 40 square feet would produce a very different thrust from that given by a screw having the same surface but 13 feet diameter. Moreover, experience has shown that with two blades the total surface may be smaller than with a pro- peller of the same diameter and pitch, but with three or four blades Professor Cote rill has shown in a very interesting paper read before the members of the I.N.A. that it is only a common screw, that is, one with very broad tips, that can make a complete column of water, and then only at its outer part. The author, for these and other reasons, therefore, prefers to take diameter and square root of blade area multiplied together as one of the governing functions instead of area simply. He also finds that within the limits of practice thrust varies inversely with the pitch ratio. Then, if A is the aggregate area of active blade surface in square feet, D the diameter in feet, V the speed of the screw in feet per second, P r the pitch ratio, and G the ratio of the distance of the centre of gravity of the blade face from the boss to half diameter of screw, the rule is : — ™ . . , Dxn/AxV 2 , lhrust in pounds = p x G. This will give ; approximately, the mean thrust of any ordinary screw, that is, of any screw whose shape or proportions are not abnormal. In practice the following values for G may be taken when it is not convenient to calculate them : — Griffiths blade, broad . . G = 0'36 „ narrow . . . G = 0'33 Oval and leaf-shaped round tip . G = 0*40 Circular blade . . . G = 0*42 ,, broad tipped blade . . G = 0*45-0'50 Mercantile square-tipped blade . G = 0'42 Examples. — (1.) To find the thrust from the screw of a torpedo boat whose diameter is 6*5 feet, pitch 8 feet, revolutions 350, area of blades 12 square feet, leaf-shape. Thrust = 6 " 5 ii-^ 12x482 x 0-4 = 16,830 lbs. (2.) To find the thrust on each of the twin screw of an Atlantic THE SCREW PROPELLER. I 25 liner whose diameter is 18 feet, pitch 25 feet, blade area 81 square feet and Griffiths broad type. Eevolutions 90 per minute. Thrust = 18x >/81x37 j>2 = j 1-39 (3.) To find the thrust on the propeller of a cargo steamer whose diameter is 20 feet, pitch 21 feet, surface 110 square feet, leaf-shape. Eevolutions 60 per minute. Thrust = 2Q> i N/llQ J <242 x 0-40 = 40,270 lbs. 1-20 Mean pitch of a screw is usually meant to be the mean of all the pitches as measured at a series of positions on the acting face of the blades, adding them together and dividing by the number. This, of course, is an arithmetical mean only, and is convenient for using when calculating the slip. But it is by no means a true measure of the active or effective pitch of any screw, and far from being so when the difference of pitch is large and varies on a rule or method such as that devised by Woodcroft, Atherton, and others. Woodcroft and others down to Sir J. Thornycroft, who made screws with the pitch of the following half of the blades considerably greater than that of the leading edge, claim that the water is set in motion gently and the total acceleration given to it only completed at the following edge. If this is so, the pitch of the following portion of the blades is the effective pitch and the mean of it taken as the mean effective pitch. In making calculations for such screws as Woodcroft's some such method must be adopted for arriving at a measure of the really effective pitch. Again, in the case of the screw, which varies in pitch from the boss to the tip, so that the greatest pitch is farthest from the axis, no arithmetical mean will give the true indication of the effective pitch of such a screw. From observation, the pitch at the broadest part or that just beyond the middle of the blade seemed to be a fair measure of the capacity of the screw; and so long as it is a true screw circumferentially the pitch may be measured on positions on that circle and the arithmetical mean taken as the acting pitch. A further complication arises when a screw is made to vary in pitch from edge to edge and at the same time to vary from root to tip. Who can possibly say what is the effective pitch of such a screw, and how can any arithmetic mean possibly be taken as an expression of it ? 126 MARINE PROPELLERS. So many of the ships showing negative slip had screws with varying pitch in the old days, while the true pitch common screw seldom gave it, that there was always a strong suspicion that wrong estimation of effective or acting pitch was why the slip came out negative; and even now there is reason to look on it with great suspicion — especially on the curved sections, which may set up an acting or active water surface apart from the metallic surface as shown in fig. 15. Loss by water "slip." — There is another source of loss with the screw in common with all propellers, and that is from the frictional resistance of the water as accelerated in its passage through the sur- rounding water. It will be seen by reference to fig. 16 that the water in flowing to the propeller from the orifice would be subject to resistance in the channel through which it flows, and a certain amount of the power expended on suction would be in this way used up and wasted. Further, it will be seen in the same figure that in passing the screw and beyond it, there is more resistance with its consequent waste. Now in the ordinary screw steamer similar passages, etc., are formed in the still water as shown in fig. 17, through which the stream to and from the propeller flows. But a portion of these passages is formed by the skin of the ship itself, so that the friction on it is used up in retarding the motion of the ship. This amounts to a consider- able quantity in paddle steamers, and will not be small in those screw steamers having three and four propellers. In the case of the ice- breaker " Ermack " with four screws, it is said her speed is better with the bow screw going " astern " than " ahead," because of the influence of its stream on the ship's bow and skin. The efficiency of a propeller is measured on the same principle as all other instruments, and tested in the same way as everything else must be sooner or later in commercial life, viz. by comparing the useful work done by the propeller with the cost, that is, the power imparted to it while occupied in doing it. The screw, however, differs from most other instruments, inasmuch as it may have two quite different values assigned to it for its output, depending on whether it is looked on as a " pusher, " when the gross thrust is the force exerted by it, or simply as a propeller, in which latter case the resistance of the ship overcome by it is the measure of its useful work. The power imparted to the screw or paddle wheel is that THE SCREW PROPELLER. 1 27 developed by the engine, less the amount required to move itself and its appurtenances. The older vertical engines driving paddle wheels or geared to a screw shaft had two large so-called "air pumps," the capacity of each of which was generally one-eighth the capacity of the cylinder. The horizontal engine for driving screw propellers had two double acting pumps each about one-twelfth the capacity of the cylinder. These pumps had to deal with large quantities of water as well as produce and maintain a vacuum of 26 inches; con- sequently a large portion of the I.H.P. developed was absorbed in this duty alone, and it probably varied as the square of the revolu- tions, as would also that for the other pumps, of which each engine had a pair of bilge pumps and a pair of feed pumps of considerable size to allow for " blowing down " the boilers. It may be taken as fairly correct that the N.H.P. of the old jet condensing engines when well made was about 80 per cent, at full speed. In the newer engines with surface condensers, and the circulating of water done by centrifugal pumps, the efficiency at full speed may have been 85 to 87£ per cent.; in newer engines still, having three cranks and only the air pump worked by the main engine, 90 to 93 per cent, is not an extravagant estimate for the best made vertical ones, judging by some experiments made on engines running with and without the propellers. The ordinary friction of a marine engine practically varies with the revolutions, for although at very slow speeds the friction per revolution increases, the rate of revolution is then much less than generally obtains when dealing with propeller problems. In a general way the friction of an engine is proportioned to the size of its cylinders, that is, to the nominal horse-power ; for this piirpose Norn. H.P. may be estimated by multiplying the diameter by the stroke of the low-pressure cylinder (both in inches) and dividing by 15 for a compound, 13 for a triple, 10"5 for a quadruple engine. A fair allowance for internal resistance of a modern engine is "006 per Nom. H.P. per revolution. Example. — What is the frictional or internal resistance of a triple engine having cylinders 20 inches, 32 inches, and 52 inches diameter and 36 inches stroke, when running at 150 revolutions ? Norn. H.P. = 52 x 36 -f 13 = 144. Friction=*006 x 150 x 144 = 129*6 I.H.P. At this speed the I.H.P. would be about 1800. The efficiency is therefore 92*7 per cent. 128 MARINE PROPELLERS. The net horse-power thus imparted to the screw is absorbed in turning it round at so many revolutions per minute, besides causing it to overcome its own resistance, frictional and otherwise, in passing through the water ; it imparts acceleration to a mass of water ; it may also in so doing set up side currents, that is, a whirling action due to the obliquity of blade and friction of surface. The imparting of motion to the water axially sets up a thrust along the axis as already explained, which thrust may, or may not, be wholly employed usefully. Augmented resistance is produced more or less by every screw propeller used in the stern of a ship, inasmuch as it must take away some of the pressure on the stern due to the " head " of water. The larger the screw the greater must be the augment, and with a very large screw it is enormous, as may be seen by referring to the " Archer's" trials, wherein a screw 12*5 feet diameter was at first used when one 8 feet in diameter would have been sufficient ; and also severe when placed close to the ship, as in the case of the " Dauntless." The useful effect of a screw is, therefore, the gross thrust due to the acceleration of the column of water less the augmented resistance set up by it. In comparing the performances of ships and propellers, much may be done with simple means, and that without doing violence to scientific truth, as by assuming that the tow-rope resistance of a ship varies as the square of the speed and as the wetted skin, as indeed it does in all well-formed ships for speed, rather below that of full speed ; that for copper-sheathed ships, such as the old naval frigates and corvettes, an allowance of 1\ lbs. per square foot at 10 knots may be taken, in a general way, as fair, although with clean new copper carefully nailed, 1\ would be sufficient; that the early iron ships with such paints as were then used probably set up a resistance of l'l to 1*125 lbs. per square foot ; and that with the modern spirit enamel paints, freshly put on, an allowance of 1 lb. is enough. Thab is the rule for — Tow-rope resistance = WS x \jT\) x f' S is the speed in knots. f=l'Q for the best enamel paints. _/=l'l for the older anti-fouling compositions. /=1'25 for coppered ships. THE SCREW PROPELLER. I 29 Tow rope horse-power - WS x \jz ) x/xfeet moved through by the ship per minute = WSxS 3 x/t 32,560. This is the power required to move the ship through the water without producing heavy waves, and is, of course, the ideal con- dition. It may therefore be taken as the standard for comparison when dealing with ships and their machinery and propellers. Hence — General efficiency . . =Tr. H.P.-M.H.P, Propeller efficiency . . = Pr. H.P. -i- Net H.P. Engine efficiency . . ^Net H.P.-fl.H.P. With such means as these it is comparatively easy to analyse the results of any trial in the better way than merely noting the speed coefficients ; and when, in addition to a full-power trial, there is one at a lower power as in the old Navy days of ll half-boilers in use/ 5 there is a capital means of checking the results observed at the higher speeds. The more recent trials in the Navy always include one at 10 knots, which is of course now by comparison a very low-powered one. Ships as now designed can move at this speed without producing waves of sensible magnitude ; hence the resistance, then, is practically skin friction only. CHAPTER IX. VARIOUS FORMS OF SCREW PROPELLER. Smith's and Woodcroft's original screws consisted of a complete convolution of one helix around a boss having a bearing at each end. Both inventors soon dropped this form and resorted to one consisting of two half-convolutions, getting thereby with the same surface better results generally, and with a marked reduction of vibration in particular. Advancing further, both shortened their screws until the length was only about one-eighth of the pitch, with results still more gratifying and fully justifying the changes. For many years all screws were made of about that length, and to-day there is no departure from that practice worth noting. Robert Griffiths, who commenced to devote himself to a study of the screw propeller at that time, was apparently the first to perceive that the chief defect of the common screw was in having such a broad tip ; that is, so much of its surface was relatively remote from the axis of revolution ; he first of all proposed to make the blades as shown in Plate V., fig. 37, but eventually he came in 1860 to the design now so well known, and an example of which is seen in fig. 38, p. 131, which was the propeller of H.M.S. " Galatea/' a very large frigate and famous in her day. Other engineers, however, had suggested blade forms at about the same time that differed largely from the common screw in respect to the distribution of acting surface and worked fairly well, but whereas their proposals have all been dropped and probably most of them forgotten, that of Griffiths practically remains to-day; inasmuch as the fundamental form of nearly every screw now in use has all the leading features of his. Another development which had a fascination for and was suggested by more than one engineer in those early days of screw propulsion was that of cutting the corners away and making the 130" VARIOUS FORMS OF SCREW PROPELLER. 131 blades of practically the same width from root to tip. The desire to have a propeller that was masked by the stern-post when the blades were in the vertical position so as to cause practically no obstruction when the ship was under sail no doubt led to, and prompted the Fig. 38.— Propeller on Griffiths' Patent (I860) Adjustable Blades. adoption of, this form of blade ; and with a view further to minimise this, the temptation to make the blades narrow was great; the result was that such propellers generally had insufficient surface and so suffered from what is now called " cavitation," and was evidenced by excessive slip. Margin Screw. — To overcome this defect, Mangin, a French engineer, constructed a propeller which was virtually a pair of narrow two-bladed propellers placed in line one behind the other, as shown 132 MARINE PROPELLERS. in fig. 39. He likewise followed Woodcroft's idea by making the blades with an increasing pitch, the leading quarter of each blade being finer than the following three quarters ; sometimes, however, the following pair of blades were of greater pitch than the leading pair. These screws were fitted first to H.M.S. "Flying Fish" in 1854, and throughly tested ; later on quite a considerable number of ships in H.M. Navy, as the result of these experiments, were fitted with Fig. 39.— ilangin's Double Screw. these screws, say from 1860 to 1870, most of them having twin screws. A few merchant ships were likewise supplied with them ; they were apparently considered to be successful at that time, but judging from the results of trials (see p. 133) it will be concluded that it must have been rather on account of the smallness of their obstruction than to any superiority in propelling power that such a verdict was arrived at. With the reduction in the sails on steamships the Mangin propeller disappeared from service, and its use has never been revived. VARIOUS FORMS OF SCREW PROPELLER. 133 It may, however, be that some of the virtue it possessed was due to the same causes that made for the success of the two and three propellers of small diameter that circumstances compelled Mr Parsons and others to fit to the screw shafts to utilise the power transmitted through them. Some further interesting experiments were made with the Mangin screw in H.M.S. "Flying Fish" in 1857, when that ship was tried with each half screw separately and then with both halves fixed with the blades at an angle of 21*5° and then with them parallel, one ahead of the other ; the results may be seen below. As might have been anticipated, the slip with the after half only was enormous, although the efficiency, as shown by the speed coefficients, compared favourably with that of the complete screw. Table IX. — Trials of Mangin's Propellers on H.M.S. "Flying Fish," 1857. Conditions of Screw. Diameter of screw . ft. Pitch of screw . . , , Number of blades Area of acting surface, Pitch ratio Surface ratio Revolutions per minute Slip . . . per cent, y Speed of ship . . knots Indicated horse-power . ,, thrust . . lbs. D 2/3 xS 3 -rI.H.P. I " J sq. ft. 35 36 Two Two Halves Halves 21-6° Apart. Set in Line. 13'2 13-2 20*1 F 20-1 F 25*8 A 25-8 A 4 4 48'8 48-8 1'52 F 1*52 F 1*96 A 1'96 A 0*360 0*360 7675 70-75 28*31 F 21-34 F 44*17 A 38-75 A 10*794 11-025 1093 1093 20,640 22,170 117*8 125*5 37 Forward Half Only. 13-2 201 2 23-6 1-52 0-173 101*0 45'49 10*908 1270 20,650 1047 38 39 After Half Only. Common Screw. 13-2 13*17 25-8 19-95 2 25-2 2 56-4 1-96 1-52 0-185 89*0 0*415 79-0 51-25 25-83 11-038 1177 16,750 116-9 11-536 1052 22,00 148-9 It will be noted also that in the case of the Mangin screw the slip was excessive, but it seems singular that it was least with the blades in line — that is, one behind the other. It is also a little curious that the speed with the forward screw only should have been practically the same as when that screw followed the other by 21*5°, and that with the after half the same is obtained with both in line. Improved Common Screw. — Emulated by the success of the Griffiths screw, and with the intention of keeping from the scrap-heap so many of the common screws as were then in existence, the naval 134 MARINE PROPELLERS. authorities commenced to reduce their broad tips by cutting away the leading corner so as to approximate in surface and breadth of tip to the Griffiths. A marked improvement in vibration was manifest in all cases, and where the original screw had had an excessive surface the speed for I.H.P. was better. Further attempts were made in the same direction by cutting away the following corners, so that the blade was now an elongated hexagon. The results in this case were disappointing so far as the speed was concerned, even although there was a further diminution in vibration. A good instance of this treatment may be seen in the case of H.M.S. " Doris," p. 213. The area, after both corners had been cut off, was insufficient. Hirsch's Screw. — Although patents continued to be taken out for improvements in screw propellers, none of them are worth noticing until 1860, when Herman Hirsch prescribed and patented a form of screw whose leading features are set out in patent 2930 (p. 29). Later on, in 1866, he took out another patent for improvements on his former ; the new screw was as shown by fig. 40. From 1870 to 1875 the shipping world was from time to time excited by the reports circulated of the wonderful improvements made in speed and coal consumption, especially the former, of ships that had been fitted with the Hirsch screw in place of one of the ordinary type. There is little or no reason to doubt the accuracy of the statements made at that time, but an analysis of them has since made manifest that in most cases of success the Hirsch screw had replaced one of very bad design and proportions; and further, that where the Hirsch screw had seemed to fail, it was only so because it had been in competition with a highly efficient screw that was already doing all that could be expected under the conditions ; that is, when the old screw was driving the ship at the highest economic speed possible with the form of the ship, the Hirsch could do little or no more. And if an improvement in speed was obtained by it in some other ships, it was generally on account of the higher power developed by the engines with the new screw ; the coal consumption was then not always so satisfactory. The fact is, that Hirsch knew so much of the right principles on which to design a propeller that he always supplied a fairly good screw and one suited to each special ship submitted to him ; whereas, at that time, the majority of screw propellers were designed by rule of thumb in a happy-go-lucky way that sometimes produced good results and more often bad ones ; and as there were no progressive trials at that time, or other methods of VARIOUS FORMS OF SCREW PROPELLER. 135 analysing and differentiating the results of trials, it was never known with certainty to what to attribute apparent failures, or rather comparative failure, for half a knot more or less short on trial speed was not generally reckoned to be serious enough to justify further trials, or even troublesome investigations. How very far unfit screws Fig. 40.— H. Hirsch's Screw of 60°. Patented 1866. can contribute to speed failure was never appreciated until the trials of the " Iris " were made and analysed. The White Star s.s. "Adriatic" was fitted with a Hirsch screw in the place of the ordinary four-bladed one. Ten voyages across the Atlantic were made with each screw. The average time with the old screw was 18 days 9 hours 18 minutes; with the Hirsch screw it was 136 MARINE PROPELLERS. VARIOUS FORMS OF SCREW PROPELLER. 137 only 17 days 5 hours 1 minute. The Hirsch in that ship was stated to be a very efficient screw when going " astern," and no doubt it would be. Hirsch's screws were said to be more free from vibration than other screws, and instances were given which seemed to prove this. No doubt with some of the older ones the Hirsch screw would compare favourably under any circumstances, but we know now that vibration is not always due to the screw, and that a comparatively small change in number of revolutions per minute will effect a radical change in it. Consequently, if the old screw was revolving at a rate which synchro- nised with the ship's period of vibration while the Hirsch was faster or slower, the change for the better would be observable. It also follows that if the power wasted in vibrating the ship was, Fig. 42. — Oval Blades of Equal Surface for Different Diameters. as it could be, applied to propulsion, the coal consumption or speed would thereby be improved. The following facts were at the time vouched for by reliable witnesses : — knots. knots. The " Louise " (German Navy), Griffiths screw, 1337 ; Hirsch, 14*07 S.S. " Herder " (Mercantile), S.S. "L'Isere," „ „ S.S. " Louisane," „ S.S. "Pereire," S.S. "Conrad," Consumption of coal per 100 miles. S.S. "L'lsere," Griffiths, 4*55 tons ; Hirsch, 4*03 tons. S.S. "Conrad," „ 1118 „ „ 10*28 „ S.S. " Louisane," „ 23*32 „ „ 2077 „ The ordinary mercantile cast iron screw for cargo steamers is made as shown in fig. 41, with the boss and blades cast in one piece. The centre line of the blade is sometimes as shown, and sometimes the 11-48 '1 12-38 7*985 n 8-99 10*05 >i 11-10 14-459 )) 15-459 9*595 n 10-65 138 MARINE PROPELLERS. blades are " thrown aft " by the other means described in Chapter VIII. Sometimes the screw is made with the face square with the shaft, with only a slight curvature near the tip. Fig. 42 shows the contour of the blades of four screws differing Fig. 43. — Modern Bronze Naval Screw. largely in diameter but having the same area of acting blade surface, the blade shapes in each case being elliptical. The surface taken for purposes of comparison is 36 square feet, and the diameters range from 9 feet to 12 feet. The pitch of each is 15 feet, and the calcula- tion is based on the revolutions being also the same, viz. 100 per VARIOUS FORMS OF SCREW PROPELLER. 139 minute. The speed of the ship would be in each case about thirteen and a half knots. The following is calculated by means of the formula given for Thrust on page 124 and for blade friction on page 113. Table X. Diameter of screw . Pitch of screw . Pitch ratio Surface ratio . Speed at tip . Thrust as calculated Friction , , Thrust horse- power . Friction ,, ,, Edge resistance, do . Total horse- power . Friction and edge H.P-r Thrust H.P -r Total H. P. ft. knots lbs. 9'0 10*0 11-0 12'0 15*0 15-0 15*0 15-0 1-667 1*500 1*364 1-250 0*566 0-458 0-379 0*319 31-8 34*5 37-4 40-3 10125 12500 15125 18000 545 643 756 875 421-6 520*9 620*2 725-0 46-5 59-5 75-9 94-7 9-3 11*9 17-2 18-9 477-4 592*3 711*3 838-6 0*132 0*137 0-147 0-157 0-117 0*121 0*128 0-135 These oval forms are those in general use to-day in the Navy and express steamer service. In the case of turbine-driven screws the blades are fuller still, and in some cases so much so that they resemble a square with the corners well rounded off.. Fig. 43 shows in detail the screw propellers as now used in the Navy with reciprocating fast running engines ; the shape of the blade is a common one, but in some cases it is even fuller at the tips. The blades usually have oval holes in the flanges, so that their position may be shifted and the pitch increased or decreased as may at any time be desired. By fitting " cod " pieces in the ends of the holes touching the bolts, the blades cannot shift. CHAPTEE X. THE NUMBER AND POSITION OF SCREWS. The single screw continued in vogue for many reasons till the 'sixties of the nineteenth century, although several inventors and pioneers had claimed to fit more than one screw, and others insisted on two screws. Trevithick in 1815 claimed for his patent screw propeller that " it may revolve at the head or the stern of the vessel ; or one or more such worms may work on each side of the vessel." Other inventors also suggested fitting more than one screw to a ship, and the general idea at one time was to substitute the spiral for the paddle wheel on each side. F. P. Smith both in his model ship and the " Archimedes " fitted a single screw in the deadwood aft ; the Admiralty did the same thing when building their first screw ship, the t( Eattler " ; and con- tinued the practice until 1866, when they built H.M.S. "Penelope." In the mercantile marine the single screw has continued to be the rule down to the present day for these reasons. The hull of the ship is somewhat cheaper when designed for a single screw, and no guards, etc., are necessary on each quarter to protect the screw. There is, moreover, one engine less to look after, which is important in a tramp or coasting steamer, with its very limited staff of engineers. With a single engine one engineer on watch can look after the engines and other machinery. With twin screws, even of comparatively small size, it is practically difficult, if not impossible, to get one man to keep watch single-handed, although with a single screw of the same I.H.P. and appurtenances he would willingly do so. More floor space is occupied by twin-screw engines, and likewise more light and air space overhead, which often means cutting into the 'tween decks. The engine shafting and propellers are more costly and sometimes rather heavier, power for power, than single-screw ones. There are double the number of working parts to attend to, to consume oil, and THE NUMBER AND POSITION OF SCREWS. 141 to require adjusting in port. Finally, there are likewise double the number of parts liable to fracture or hindrance. On the other hand, the liability to break down, that is, to stoppage, is less; about a half that of the single screw. If one screw is so damaged as to be useless, the ship can still steam at a respectable speed with the other ; and if the rudder or steering gear is disabled, the ship can be navigated by manipulating the engines. Moreover, in the case of large high-powered ships the engines may each be of so much more moderate a size as to become less costly to make, and certainly less so to repair or overhaul. Take the case of the s.s. " City of Berlin '' of 1874 with one screw and a low-pressure cylinder 120 inches diameter and 66 inches stroke 5200 I.H.P. ; or that of the " Etruria," single-screw ship 14,500 I.H.P., built in 1884, which had two low-pressure cylinders 105 inches diameter and 72 inches stroke ; while the "Celtic/' with her large power (13,100 I.H.P.), has two low-pressure cylinders, one to each engine 98 inches diameter and 63 inches stroke; and H.M.S. "London," of 15,500 I.H.P., has a low-pressure cylinder to each engine only 84 inches diameter and 51 inches stroke. It is, of course, obvious that a twin-screw ship is safer and handier than, and can be manoeuvred in a way impossible with, the single- screw ship ; and also equally plain that when the draught of water of a ship is such that a single screw of sufficient diameter, etc., for the engine power cannot have proper immersion, two or more screws must be employed. In fact, the first twin-screw ships were so designed because of the shallowness of the water they had often to enter and navigate. Rennie's ships of 1853-4 (see page 28) were required for service on the Nile when it was low; the twin-screw ships built in the early 'sixties of the nineteenth century were for blockade running ; they had to be of large power and of so light a draught that they could get in and out of the Southern ports through channels that were not navigable by the Federal war ships blockading them. After the American Civil War such of these vessels as survived were employed in the Cross-Channel and North Sea express and other services, where their light draught and speed was of advan- tage in competing with paddle steamers. The British Admiralty built a very considerable number of twin- screw ships from 1865 to 1870, mostly small cruisers and gunboats, 1 but the "Captain," of 7672 tons, the "Audacious" class of four ships, 1 Fide Chap. XVIII. [42 MARINE PROPELLERS. 5560 tons, were quite large ships and were the means of fully demon- strating as well as convincing the naval authorities generally of the superiority of the system for naval purposes. For cruising purposes it was at first intended to use one screw only, the other being dragged through the water without revolving. Two of the ships were tried in this way with the results set out below, which show that at about lOf knots on H.M.S. " Invincible " it takes about 427 more I.H.P. to drag the screw than to revolve it, while on the " Vanguard " it took 953 more I.H.P. to drag one screw at 1T36 knots. (See Table XL) Table XL — Trials of certain Twin-screw Ships with one and both Screws running. Length 280 feet, beam 54 feet, draught of water 21 feet, displacement 5560 tons. H.M.S. "Vanguard." H.M.S. "Invincible." Both Screws Full Power. Both Screws Half Boilers. One Screw Full Power. Both Screws Full Power. Both Screws Half Boilers. One Screw Full Power. Diameter of screw . . ft. Pitch „ . „ | Revolutions per minute Slip per cent. ... Speed of ship . . knots Indicated horse-power Indicated thrust . . lbs. Tow-rope resistance of ship ,, Tr. horse-power Efficiency, Tr. H. P. -r I.H.P. Displacement 2/3 x speed* -f I.H.P. 16*17 20'6 mean 73-7 0-21 neg. 14-944 5366 116,600 55,600 2559 0'479 195'3 60*27 4*01 neg. 12742 2752 73,100 40,500 1583 0*575 236'0 70-49 2075 pos. H'356 2903 59,100 32200 1135 0-392 158*4 16-17 17-2 mean 79'0 0-949 neg. 13-51 4562 110,800 45,620 1891 0-415 166 62-18 3-72 neg. 10-926 2438 75,200 29,800 1001 0-410 169-5 79*0 20-2 pos. 10-797 2772 66,500 29,140 967 0-349 139 6 Commodore Melville found by experiment on a twin-screw ship that with one screw running loose and disconnected from the engines for revolving it, when the ship was going 10 knots took 150 I.H.P. ; whereas to revolve it and its engines coupled took 300 I.H.P. Mr William Froude, when experimenting with the " Greyhound/' found the resistance was less with the propeller secured than with it revolving — that is, the power to make it turn was greater than that required to drag it through the water masked behind the stern- post. Table XL gives the results of some trials made by the Admiralty to test this question. Two Screws.— Mr James Howden in 1874 fitted several large tug-boats with two screws, one at the bow and one at the stern on THE NUMBER AND POSITION OF SCREWS. 1 43 the same line of shafting, operated together by one engine. They were said to be good at towing, and probably are handy in the sense that the bow propeller when running " astern " would soon check the headway of the ship, just as the floats of the paddle-wheel tugs do. This enables the tug to run close to its objective at high speed before slowing down. Further, with a tug, the back wash of the forward screw causing pressure on the bow and increase of skin friction is not of much moment when towing. There is another arrangement of two screws which is noticeable and interesting, inasmuch as it is a survival of the invention of Perkins, Ericsson, and others. The Whitehead torpedo, in order that the movement of the screw may not affect the steering in the least degree, is propelled by a pair of screw propellers running co-axially in opposite directions, one before the other, on concentric shafts. Triple screws were first tried on H.M.S. " Meteor" in 1855, with the view of getting a larger disc area and, consequently, more speed. There was, however, only one engine, which drove the centre screw direct, and it was geared to the shafts of the wing ones. The loss from friction and the faulty form of the ship were such that the speed was low and the efficiency very poor. For the same reasons as given for preferring twin screws, especially with shallow-draught ships of high power, the arrangement of three screws naturally commends itself. It also has other advantages, however, inasmuch as by more subdivision smaller engines can be fitted, besides which when cruising there is a greater range of choice in the employment of the engines when comparatively low speeds are required. For example, the middle engine can be run at full power and with consequent high efficiency when the wing engines are standing; or the two wing engines only may be employed at full power, with a corresponding advantage. With the advent of the turbine motor, however, the three screws became necessary in order that the propellers might be of the smallest diameter to keep their peripheral velocity within practical limits. The Italian naval authorities were the first to give the triple screw system a practical trial by fitting it to the torpedo cruiser "Tripoli," 848 tons displacement, 2543 I.H.P. and 19 knots, built in 1886. They followed on with the "Montebello" in 1888, a sister ship with 19 knots speed, and later on some others. Experiments had been made by the French naval authorities so far back as 1884-5, which resulted eventually in the building of the 144 MARINE PROPELLERS. armoured cruiser " Dupuy de Lome" in 1890, by M. Marchal, chief constructor. This ship was very much larger than the Italian, being of 6400 tons displacement, 14,000 I.H.P., and having a speed of 20 knots. In 1895 the French built a much larger ship with three screws, viz. the " Charlemagne," of 11,273 tons, 14,500 I.H.P. and 18 knots. Since then, most, if not all, of the French naval ships of large size have had the three-screw arrangement. Commodore Melville, Engineer in Chief of the U.S. Navy, became a convert to and great advocate of the system, and in 1892 caused the cruiser " Columbia " to be so fitted. She was a comparatively large ship, being 412 feet long, 58 '2 feet beam, 25*6 feet draught of water, 7375 tons displacement, 18,500 I.H.P., and on trial attained a speed of 22*8 knots. The Commodore studied the matter very fully, and was good enough to give his experience and the conclusions he had arrived at in a paper read to the Institution of Naval Architects in 1899. He advocated the making of the midship screw and machinery so that half the full power would be developed in them, and a quarter of the LH.P. in each of the wing engines, and thereby to get a better variation in the division of power for cruising purposes. He estimated that the gain effected by the three-screw system over the twin screws was as much as 11*9 per cent, in the "Columbia, 55 and that a fair average gain by using two screws instead of one screw would be 8 per cent, for speeds from 12 to 20 knots; that at 15 knots a ship will gain, by using three instead of two, 5 per cent. ; while at 24 knots the gain in efficiency would be as much as 12 per cent. These supposed gains were, however, called in question by the British naval authorities, and it was their opinion at that time that the practical objections to three screws outweighed any gain that was to be got, and until the turbine compelled them to depart from it, the twin screw remained the established practice in the British Navy. In 1892 the German Government built the cruiser "Kaiserin Augusta," 6330 tons, 14,000 LH.P. and 22*5 knots speed, and in 1897 some cruisers, somewhat smaller, all with three screws, followed by the " Fiirst Bismark," of 10,650 tons, 14,000 I.H.P. and 19 knots, an armoured ship, and since then several others. Kussia built the armoured cruiser " Rossia " in 1896, of 12,130 tons, 14,500 I.H.P. and 20 knots speed, and followed on with several other ships all having three screws. It may be said indeed that the Russians were the first to use three screws, seeing that the THE NUMBER AND POSITION OP SCREWS. T45 Imperial yacht '' Livadia/' built by the Fairfield Company in 1886, had three screws. Four screws have been fitted to the Cunard steamers " Lusitania " and t( Mauritania " (see frontispiece), because of their great power and their high rate of revolution. It is not unlikely, from the success that has attended the arrangement iu these vessels, that it may be followed by fitting in this way other and even smaller ships. By this subdivision reciprocating units of quite a reasonable size could be employed for very high power for two if not for all the screws, for there might be a reciprocating engine driving its own screw and exhausting to a low-pressure turbine, which in its turn would drive an independent screw on the other side, and vice versa, so that when cruising at half power or less a complete expansion installation could be used without affecting the steering of the ship. The well-known Russian ice-breaker " Ermack " has four screws rather oddly distributed, viz. three at the stern and one at the bow. It is stated, too, that when the bow screw is going "astern" the greatest " ahead " effect is obtained — in fact, more than when it also was going ahead too. This curious result is attributed to the current set up around the bow and at the sides in the direction of the ship's motion by the fore screw when in stern gear. Some of the Mersey ferry steamers have four screws, two at the bow and two at the stern, on a pair of shafts running through the ship on Howden's system. The vessels are double-ended and go either way. The " bow " screws are, as before mentioned, very effective in checking the headway on such ships, as well as in starting them quickly either way. The position of the screw very materially affects its efficiency as a propeller. In the case of the " Archimedes " the propeller was a long one, and placed in a recess or gap in the deadwood (see fig. 4). It seems to have worked fairly well, considering all things, and the conclusion come to is that she had a clear " run " — that is, a very fine line after-body. The after-body of H.M.S. "Battler" (see fig. 7) was all that could be desired for success as a screw steamer, and her high efficiency is no doubt in no small measure due to the position of the screw and the clean lead for the " feed " water. Whether this was not appreciated as it should have been by the naval constructors of the day, or whether parsimony overruled their science, it is difficult to say, but when the conversion of sailing ships into screw steamers was taken in hand some most egregious blunders were made, and 10 146 MARINE PROPELLERS. although rectifying them was a costly business, it was an experience by which they and we benefit, or ought to. It was thought quite good enough to cut a gap in the deadwood immediately on the fore- side of the stern-post, fit a new post, or in some way make a new end to the ship to which a stern tube was fitted and generally an arrangement made for a banjo frame, etc., so that the screw might be lifted up. H.M.S. " Dauntless/' a frigate of 2307 tons displacement, was in 1848 converted into a screw ship as described above, fitted with engines of 580 Nom.H.P. made by Messrs E. Napier & Sons, having two cylinders 84 inches diameter and 4 feet stroke, big enough to have driven her 11 knots, as the prismatic coefficient was only 0'72. As it was, on her official trial she only managed to do 7*366 knots with the engines developing 836 I.H.P. The tow-rope resistance was only 8378 lbs. and the Tr.H.P. 189*7. The efficiency, conse- quently, was only 0'227. A new stern was formed by carrying out the original lines under water so that the ship was 9*5 feet longer and the screw placed that amount of space further aft, which, although not much, was sufficient to effect a marked improvement in the speed, inasmuch as it was then 10'02 with 1388 I.H.P. and an efficiency of 0'350; with an alteration in the pitch of the screw, a better speed was maintained with 1217 I.H.P., and the efficiency rose to 0-432. (See Table XII.) Table XII. — H.M.S. "Dauntless" Steam Trials Before and After having a new stern, compared with those of modern cargo Steamers. Dauntless 2 Dauntless 3 s.s. Z. s.s. F. 1 Dauntless 1 after Altera- after Altera- Ordinary Ordinary as built 1848. tion. Same tion, with New Cargo Cargo Screw. Screw. Steamer. Steamer. Length beam draught . 210x40x16-3 219-5x40x16-6 219-5x40x16-3 250x30x16 240x33x15-0 Displacement prism coeffici- ent Wetted skin 2307 - 0-723 2235 - 0-700 2307-0-708 2270-0-700 2350-0-750 11934 12068 12136 12960 11334 Screw, diameter and pitch . 14-73x16-8 14-73x16-8 14*73X17-72 12-5x14-8 12-5x15-0 Revolution per minute 56 9 71-7 68-28 77 -0 80-5 Slip per cent. 2182 15-63 13-73 12-28 11-69 Speed . . . knots 7-366 10-016 10-293 10*00 1055 Indicated horse-power 836 1388 1217 685 866 i Indicated thrust . . lbs. 27,688 33,170 33,050 19,700 23,890 Propeller thrust calculated 12860 18,420 17,630 13,575 Tr. resistance of ship . lbs. 8,378 15,6S8 16,029 12,960 12*631 Tr. H.P. 291 557 558 416 Tr H.P 189*7 483 525 398 407 Efficiency Tr. H.P.-S-I.H.P. 0-227 9-349 0*431 0-5S6 0-470 Displacements ' 3 x speed :f — 83-4 123-7 156 5 253-0 239 I.H.P. THE NUMBER AND POSITION OF SCREWS. 1 47 Two screws on the same shaft have been tried by Mr Parsons and others where limitation of draught or other causes have necessitated the use of a screw of such small diameter as to preclude the possibility of the necessary surface in one screw only for efficient propulsion. In such cases the screws are not close to one another as fitted by Mangin, but sufficiently far apart to have a feed for the aftermost independent of the stream from the leading one. The efficiency of such a combination must depend largely on the shape of the ship and the distance apart of the screws ; hitherto the success of this arrangement of propellers has been somewhat qualified and not sufficiently pronounced to recommend its adoption in the future. From Millington in 1816 and onward, inventors have not been wanting in taking out patents for so fitting the screw propeller that it shall help to, if not altogether, steer the ship ; but very few of them have carried their ideas into practice. Fig. 45, however, is one of the few instances in which in workaday practice a second screw has been fitted abaft the propelling screw and arranged to help in steering as well as in propelling the ship. The illustration is taken from a photograph of the screw steamer " Stratheden," built in 1882, of 2000 tons and 200 nominal horse-power, and the arrangement of the auxiliary screw with its fittings and connections is in accordance with Mr J. J. Kunstadter's patent. It will be observed that the rudder pintles and gudgeons are of special design suitable for carrying the weight of the auxiliary propeller and its shaft, and the universal joint connecting the latter to the main screw shaft is of a substantial nature, as indeed it should be for the purpose. As the example set in this ship does not appear to have been followed, it is to be presumed that the advantages that the arrangement possessed were found to be outbalanced by the disadvantages, such as prime cost, exposure of auxiliary screw to damage from quay walls and entanglement with ropes and chains, as likewise the extra cost involved in examining and overhauling the propeller shafts, rudder, etc. Scott Russell used to say that there were several places about a ship where a screw might be fitted, but that the worst place of all was the bow, because the stream delivered from the screw impinged directly on the bow and the current flowing about the bow and sides of the ship would be increased in velocity, causing a corresponding augment of resistance. Professor Osborne Eeynolds, and others before him and since his 148 MARINE PROPELLERS. time, have taught and sometimes proved that the further away a screw is from the ship itself, the more efficient it becomes, and that when quite away from "wake" currents its maximum efficiency is reached. Possibly this teaching lias induced the belief in the minds of some engineers that the bow is the place for a screw. This leads thought into another channel, and causes inquiry to be made into the nature, magnitude, and effect of the currents caused by the propeller itself on the ship. 1 Propeller abaft the Rudder. — But long before Professor Reynolds made his researches, others had recognised the ad- vantage of placing the screw as far away as possible from the body of the ship ; for as early as 1855 steamers were built at Hull having the screw behind the rudder on the plan patented by John Beattie in 1850 (see fig. 44) and proposed by Ericsson in 1836. The "feed" to the screw so placed was excellent ; and further, on reversal for " astern " motion, the impact of the stream from the screw was not direct on the body of the ship as is the case with the screw in its usual recess, when the result is that its motion is seriously checked ; it is, of course, obvious that a screw standing out abaft the rudder- post must be exposed more than usual, and so liable to damage from hitting quay walls and pile work as well as to fouling with ropes and hawsers. It was on this latter account that fitting screws in this position was eventually discontinued and some of the ships so constructed were altered. Submersible Screw- — To more than one of the early marine engineers the idea occurred of arranging the screw so that when the 1 Vvh Osborne Reynolds in Trans, of InM. Nara? Architects, vol. xvii., 1876 ; Geo. Calvert, ibid., vol. xxviii., 1886, and vol. xxxiv., 1892. Fig. 44, — Screw outside of Rudder. THE NUMBER AND POSITION OF SCREWS. 149 ship was in open water with plenty of depth it could be more deeply submerged and its efficiency greatly improved thereby. It was then clear away from the broken wake of the bluff part of the after-body, and the possibility of the blades breaking through the surface was quite remote. Fig. 46 illustrates the method of affecting this change of position suggested by Shorter in 1800, by Fig. 45. — Auxiliary Screw outside the Rudder, Trevithick in 1815, Millington 1816, and patented by G. H. Phipps in 1850. The late Sir Edward Harland carried the idea into practice on a big scale when he fitted the s. s. " Britannic," of Atlantic fame, with her huge propeller in a sliding frame, had the shaft connected to the next length by a massive universal joint so that when out of the Mersey and on the open sea it could be lowered down till the tips worked well below the line of keel. So much trouble was experienced, however, and when the risks of serious accident became appreciated, as they did very soon, that this magnificent ship was taken from i5o MARINE PROPELLERS. her station and converted to the usual form of stern and propeller arrangement. Submerged Screw. — Sir John Thornycroft and Mr Yarrow have for many years fitted torpedo boats and other small high-speed craft with the propeller so low down that its blades are far below the boat itself. This, however, was perhaps more in the nature, or perhaps the result, of a development, for, first of all, such fine-lined boats would have a long and deep " deadwood," producing a large amount of skin friction ; it was gradually cut away till the shaft was reached, and, finally, the boat was designed with no deadwood and the shaft just inside the bottom. As these boats never have " to take the Fig. 46. — Phipps' Lowering Screw, 1850. ground," and when on general service are in fairly deep water, there is no risk of damaging the screws, etc. The position of twin screws was at first governed by the practice common with single ones : that is to say, they were placed just abreast of the spot where a single one would have been, that is, in a place just forward of the rudder so that that instrument worked quite free from the tips of the blades. The weight, etc., of the screws and their shafts were taken on A frames or brackets with two legs, one of which was vertical or nearly so, under the ship's counter, and the other horizontal (or nearly so) and attached to the ship's body. Sometimes the stem-frame had a gap, as for a single screw, so as to permit of the two propellers working with their tips nearly touching, and later on, in order to keep the screws as far in as possible THE NUMBER AND POSITION OF SCREWS. 151 so as to be protected by the ship's counters from damage, they were arranged to considerably overlap by placing one a few feet ahead of the other. The first twin-screw ship in the Navy, H.M.S. " Penelope," had double sterns united to one body and formed into one above water. That is, there were two runs each with deadwood and rudder as well as two screws. The ships so built were not at all satisfactory. and most costly to construct. Of late years, especially with large ships, the after-body is so formed that no brackets are necessary. The two deadwoods or fins in this case are placed nearly horizontal and have at their outer edge the cylindrical formation enclosing and carrying the stern tube, shaft, etc. (See frontispiece.) This, no doubt, is for all size of ships a much stronger design for carrying the twin screws, etc. ; and, moreover, the shafts thereby are fully housed instead of exposed, and they can be more conveniently subdivided and handled. The flat formed by the two deadwoods or fins forms an admirable resistance to pitching, and so must tend to render the ship more comfortable in a head sea. On the other hand, a huge amount of surface is exposed to skin friction, and if the fins are not properly designed so as to follow the natural stream lines of the ship, they will form serious obstructions. To some extent also when pitching, even slightly, the feed to the propellers is affected by them. The position of the triple screws differs somewhat from that of twin screws, inasmuch as the wing screws must considerably over- lap the middle one, and as far as possible it is desirable that the latter should not injuriously affect the working of the wing screws or be affected by them. To this end the wing screws are well in advance of the centre one, which, of course, is only possible in ships of very fine lines and large beam. For this reason, no doubt, until the advent of the turbine with its small screw, the three-screw system was not followed in the mercantile marine. It is pretty certain that the screws may with advantage be at a considerable distance forward, so long as the form and dimensions of the ship permit of it. In this case the method of carrying the wing screws, shafts, etc., will be the same as for twin screws. Quadruple screws, as in the " Lusitania," are naturally in pairs (see frontispiece). The aftermost pair will be very nearly in the same plane as the single screw would have been in the case of triple 152 marine propellers. screws, and the other pair will he forward of them, the exact position being determined partly by the disposition of the engines and partly by the shape of the ship. They may, in any case, be placed fairly close to the skin of the ship, and thereby work in the water that is influenced by the motion of the ship, which will there be free from eddies, and fairly constant in the direction and magnitude of its motion relatively to the ship. CHAPTEK XI. SCREW PROPELLER BLADES : THEIR NUMBER, SHAPE, AND PROPORTIONS. Number of Blades. — Various experiments have been made from time to time with model screws, with screws on steam launches and small yachts, and with screws on frigates and battleships, with the object of determining the relative merits of propellers with two or more blades, and, if possible, what number a propeller should have for highest efficiency. The late Mr Charles Sells, for so many years a most successful designer for Messrs Maudslay, Sons & Field, made an elaborate series of trials with screws having two, three, four and six blades varying in pitch, etc., the report of which is given in Chapter XVII. Mr R. E. Froude has for many years devoted much time and thought to the study of the problems involved in screw propulsion, and made so many series of experiments in the tank at Haslar, that little remains for him to discover respecting either the form or number of blades of a propeller. The late Mr Blechynden conducted a series of trials with model screws in a tank, or rather in an oval channel which permitted the water acted on by the screws to How round and round without check or hindrance. The thrusts and turning moments were carefully measured and noted, and were published by him in a paper read to the members of the -North-East Coast Institution of Engineers and Shipbuilders, 1886-7. The following are the conclusions indicated by these careful and conscientious experiments : — The thrusts on propellers of the same diameter, pitch, surface, and shape of each blade, and differing only in the number of blades, are as follows : — 153 '54 MARINE PROPELLERS. Table XIII. Froude. Blechynden. Sells. 1 \ / No of Blades i ^ Screw with four blades . three ,, , , two , , i 1-000 0'850 0-650 rooo '862 0-680 1-000 0-907 0-703 1000 ; 0866 0-707 This really means that the thrust varies as the square root of the area of acting surface and nob in direct proportion to it. The following figures, taken from the trials of H.M. S. " Iris," seem to confirm this. In this case the first set of trials were made with four blades on each propeller ; the second set were made after remov- ing two opposite blades from each so that diameter, pitch, shape and area of each blade was the same in both trials, the aggregate area of acting surface being of course double in the first trial what it was in the second. Table XIV. Revolutions per minute. Thrust with four-bladed screw lbs. two Ratio of second to first . 20,000 14,000 0-700 31,000 ! 45,000 21,500 ' 30,000 0*69 0-67 70 60,000 41,000 0-69 90 78,500 , 98,500 55,000 70.000 0'70 0-71 Mr Isherwood had a series of trials made with a steam launch fitted with screws made by E. & W. Hawthorn over twenty-five years ago. Mr Blechynden made a careful analysis of them and came to the following conclusions as a result : — (1) In screws of equal diameter and pitch, but of different blade area, when the same thrust is developed, the turning moment is independ- ent of blade area, (2) Screws of equal diameter and blade area, but varying in pitch ratios when tried under similar conditions and developing equal thrusts, have turning moments proportional to the pitch. (3) Screws varying in diameter but of equal pitch ratios, develop- ing equal thrusts, have turning moments proportional to the diameter. (4) In any screw, if the total blade area remains constant and the blades are similarly shaped, the propelling effect is the same whether there are two or four blades. SCREW PROPELLER BLADES. 155 This last conclusion is not quite in accordance with the deductions to be made from everyday practice, nor did Mr Walker, 1 who had made a series of trials with his yacht, agree to this proposition as being correct. The third and fourth set of trials of the " Iris " may be appealed to, although perhaps objection may be taken to them, as they were a complete refutation of the above. But inasmuch as the two-bladed screws of the fourth trial with their surface of 56 square feet pro- duced a better speed than the four-bladed ones of the third series with 72 square feet, it would seem to be contrary to what might have been expected with the fourth proposition before their eyes. On the other hand, the trials of screws with different number of blades in H.M.S. "Emerald," while not confirming it, does not dis- tinctly disprove it. Table XV. — Trials of H.M.S. "Emerald" with various Screws. Particulars of Screw. Diameter of screw . ft. Common. Common. Common. 18-0 18-1 18-0 Pitch ,, . ,, 28 '0 26-0 26-0 Number of blades Two Four Six Developed surface 82-8 99-0 103-0 Revolutions 53-83 53-33 51-50 Slip per cent. .... 22-49 12-28 11-26 Speed of ship . . knots 11*530 12*003 11-726 Indicated horse-power 2288 2323 2124 Indicated thrust 48,800 55,300 52,300 It is true that, here, area of surface is not exactly the same in all the cases, but the difference in area is not large, especially between the screws with four and six blades. Mr Froude's experiments were carried out in the large tank at Haslar, with the model of the ship itself in advance of the model screw, and both moving together at the speed corresponding to the relative sizes of model and real ship. The thrust of the screw and the turning moments are most carefully taken by ingenious self- registering instruments, and every care taken to eliminate disturbing elements and keep the records absolutely correct. Such experiments as carried out by these gentlemen are, of course, most interesting and likewise instructive, especially as enabling us to compare one model screw with another, but as a final criterion of the 1 Vide Walker, Trans. Inst. Naval Architects. Also see Chap. XVIII. 156 MARINE PROPELLERS. screws most suitable for ocean-going ships there is good reason for reserve of judgment. It is by no means certain that an accurate comparison can be made between the performance of the actual screw in smooth water with that of its model in water of the same density and viscosity. 1 It is certain that in rough water with the ship sub- jected to a mixed motion of rolling and pitching even when slight, the performance of her screws cannot be gauged by model experiment in a tank. Such performance with the ship moving in any way but that of the straight line of its keel can scarcely be repeated in the tank, nor can the disturbing effects be measured or estimated. The comparative failure of the original screws of H.M.S. "Drake" and her sister ships is an indication of the fallibility of the tank method. At the same time all honour is due to Mr E. E. Eroude and others for the very great — the almost incalculable — work they have all done both in general research and the practical service rendered for which they are primarily employed. An experiment made on the s.s. " Charkieh " with a six-bladed screw of the same diameter, pitch, and surface as the three-bladed service one, and having the same shape of blade, is of interest, and gave the following results : — Three-bladed screw. Coal con- sumption per day . . . 34'01 tons. Speed 10'69 knots. Six-bladed screw. Coal con- sumption per day . . . 33*16 „ „ 10*65 „ Early practice with screw propellers was to follow the example set by E. P. Smith and Bennet Woodcroft, and fit two blades only instead of the multiplicity of Ericsson and others. This was no doubt largely, if not entirely, due to the fact that in H.M. service, as also in the mercantile marine, there existed the desire and need to retain sails as a means of assisting the machinery when at work and to navigate the ship when the wind served sufficiently well to do with- out the engines — perhaps also finally with the idea that the sails would always be a stand-by in a case of engine failures. Otherwise, there seems to have been no good reason for limiting propellers to two blades. In the mercantile marine from the earliest days of steamships there had been a considerable number of them to which sails could be of little or no use, save, perhaps, only to steady them in a beam sea 1 Froude's model screws are usually 8 to 9 '6 inches in diameter. SCREW PROPELLER BLADES. 1 57 and to prevent or quickly damp out heavy rolling. To such steamers a three-bladed screw would be an advantage, and no doubt such were used in comparatively early days, as well as the four-bladed ones were later on. The blades of Smith and Woodcroft were for a considerable time about one-sixth to one-eighth of the pitch in length and the pitch fairly fine, consequently their tips were very broad, with the result that the greater part of the thrust was exerted near them. The difference between the pressure on the top and that on the bottom blade was therefore so great when at or near the vertical position that the vibration was very considerable in amount and trying in character. This was, of course, aggravated when the screws were of large dia- meter and only just immersed. Screws with two blades by Griffiths, Sutherland, and others, having comparatively narrow tips, did not so much distress either ship or passengers. Screws with more than two blades, even when fairly broad tipped, were not so bad as the two- bladed variety. The Admiralty, be it recorded to their credit, made some interest- ing and valuable experiments to assure themselves on these points, chiefly on the best number of blades for naval purposes (vide Chapter XVI.). In 1861 the three-decked battleship "Duncan," of 3985 tons, was tried with a three-bladed screw in place of other two- bladed ones ; later this screw had its leading corners cut off as to reduce the area of its acting surface from 115 square feet to 107*8, as the two-bladed Griffiths had only 77'6 square feet. The speeds varied very slightly, the highest being with the Griffiths, but the speed coefficient with it was only 176*2 as against 187*6 with the three- bladed. A year later further experiments were made with screws differing chiefly in number of: blades on H.M.S. tc Shannon," a frigate of 3612 tons. In this -case the screws having more than two blades were on Woodcraft's principle and had blades of the same size, diameter, and pitch, so that the acting surface of the six-bladed screw was double that of the three-bladed one. The two-bladed one was, however, a common one, but with surface nearly proportional to the number of blades, as were the others. The best speed, 11 '55 knots, was made with the four-bladed screw, the three-bladed running it close with 11 '492 knots. The efficiency as measured by the speed coefficient gives the same verdict in favour of the four-bladed screw. 158 MARINE PROPELLERS. In 1863 a further experiment was made with H.M.S. "Emerald," a frigate of 3563 tons (see p. 155). In this case the propellers were two-, four-, and six-bladed common screws with surface 82'8, 99*0, and 103 square feet, a much more satisfactory proportion. The highest speed, 12'003 knots, was attained with the four-bladed, with a speed coefficient of 171*7 ; the next highest being 11 '726 knots, with a speed coefficient, however, of 175*1 by the six-bladed screw. Follow- ing these trials a considerable number of ships were fitted with four- bladed screws, but with the results already stated. The modern practice as to number of blades on a screw may be summed up by saying that in the mercantile marine the four- bladed screw is almost universally employed ; in the naval service and with all modern swift express steamers, three blades are almost always employed. In both services, however, are still to be found propellers with two blades notwithstanding the prevalence of these rules. Now, as a matter of fact, the right number of blades for a particular propeller cannot be determined by fashion nor be a matter of pre- judice, but must be governed by the conditions impressed on and ruling it only. Two-bladed screws have some special claims for consideration. In the experiments with H.M.S. " Duncan," ' ( Shannon," and " Emerald," the performances of the two-bladed screws were by no means bad by comparison with the others. An inspection of the results of these trials impresses one with the idea that had the two- bladed screws been run at the same revolutions and their surface been nearly equal to that of the other screws, very different verdicts would have followed. Even as it was, with H.M.S. " Duncan," the better designed Griffiths with its two blades gave the best speed. Its low-speed coefficient is, doubtless, due to the high speed of revolution. Griffiths' improved screws with two blades were very efficient and the vibration quite moderate, even if due to them. Now, as a matter of fact as well as experience, the two-bladed screw does not require so much total acting surface as one with more blades, because of the greater efficiency of blades, due to the breadth ; besides which, there are fewer resisting edges. If of the same surface the breadth of blades will be as 50 to 33 of a three-bladed screw. A reduction of 10 per cent, in surface with the two-bladed screw means a substantial saving in surface friction, while the blade breadth is still as great as 45 to 33, and thus possessing a superiority sufficient to produce a marked difference in results. SCREW PROPELLER BLADES. 1 59 It is, therefore, no matter for wonderment that with its superiority in efficiency the Griffiths screw was retained by the Admiralty for all classes of ship. In spite of the claims of other propellers, to the present day, so far as shape is concerned, the Griffiths idea is followed. In the case of the " Iris" the performance of the original screws with two of the blades of each removed was very good, compared not only with that of the screw bearing the whole four blades, but even when judged by that of the new propellers. In fact, had those blades been given a foot or two more pitch there would have been no justification for new ones, except, perhaps, on the ground of vibration. Again, it is noticeable that of the new screws made specially to replace the condemned ones, the two-bladed Griffiths, in spite of its huge diameter, gave the highest speed with the best speed coefficients. To-day we have yachts and other cruisers for waters where coaling stations are few and far between, and other ships with which quickness of passage is not so important as cheapness. Such ships continue to be fully masted and have full sail power, and yet require screws which combine minimum of obstruction with the maximum of convenience for adjustment so as to yield the best combined steam and sail results. They are therefore generally fitted with Bevis' feathering screw (see fig. 47) which allows of this, inasmuch as the two blades can be turned so as to partly fill the stern apertures and yet be masked by the stern-post. The motor boat, which has come to stay, in spite of its non- reversible engine and many noises and smells, requires a somewhat similar screw, inasmuch as, since the engine can only run one way, the propeller must do the same, unless the objectionable wheel gearing is introduced ; the screw blades must then turn round . so as to become a left-handed screw if in " ahead gear " it was right-handed. Such screws will of necessity be two-bladed and have a flat acting surface on the aft side; but probably slightly convex surfaces on both sides will be the most efficient under such circumstances. Maudslay's Feathering Screw and Banjo Frame. — In the early days of the screw frigate it was desirable to have a propeller whose pitch could be altered, and to have an apparatus for lifting it out of the water, so that it ceased to obstruct the passage of the ship. Most of these old ships sailed remarkably well, and in order to use the steam power to advantage when the ship was doing well under sail, i6o MARINE PROPELLERS. it was necessary to increase the pitch of the screw so as to keep down the revolutions of the engine. Several inventors took out patents for the purpose of attaining this end, and one of the best and T I most successful was that fitted by Messrs Maudslay, Sons & Field to H.M. S. "Aurora" and other ships. Fig. 48 shows very clearly the method of accomplishing the end in view, and also the banjo frame containing the screw, and in which it was raised to deck level when SCREW PROPELLER BLADES. 161 the ship was under sail and lowered into position and secured there when under steam. : " . ■ *t [j « The method adopted by Mr Bevis of Birkenhead, which is shown in fig. 47, was, however, so much neater and easier to deal with, that in course of a few years it remained as the only feathering propeller ii 02 CI a Ph I 62 MARINE PROPELLERS. in use, as it still is to-day. It will be seen that both Maudslay's and Bevis' screws could have the blades turned to the fore and aft position so as to cause no obstruction when sailing. Flat blades, that is, those having a constant angle and conse- quently a pitch increasing from root to tip, have been tried on pro- pellers where complete feathering, that is, the blades of which are turned exactly fore and aft, is required for efficient sailing ; or, in cases where the reversing cannot be done by the motor, but must be done by reversing the screw from a right-handed to a left handed-one, such blades would seem to be desirable. But, as a matter of fact, a propeller fitted with them is of very low efficiency, and is equally bad in either head or stern gear ; while a helical blade may lack efficiency when reversed for stern gear, at which times it is not important, as they are of short duration and not frequent. Sir John Thornycroft found that such a screw required nearly twice as much power as a helical screw of the same dimensions. At 11J knots the flat blades required 173 I.H.P. to drive them, whereas 90 I.H.P. was sufficient for the common screw ; and at 10 knots there was practically the same difference. Three-bladed screws are, on the whole, the most satisfactory for general purposes, for they possess high efficiency when working under almost any circumstances ; that efficiency is satisfactory whether the screw be working at a considerable depth or so near the surface as to induce air currents to follow it. The increase of total blade surface over that of the two-bladed screw is not, after all, necessarily very large, while its breadth of blade is respectable, being, as already stated, 33 as against 45, or a reduction of 26-6 per cent. In the matter of vibration the three-bladed screw is superior to either the two or four-bladed variety, inasmuch as no two of its blades are opposite one another so as to form a couple tending to shake the stern, as is the case with both the two- and four-bladed screws. The loss of a blade amounts of course to 50 per cent, reduction with a two-bladed as against 33 per cent, with the screw having three blades; but as the two blades of the one screw have a smaller surface than the three of the others (say by 10 per cent.) the comparison is 33 against 45 per cent. Four-bladed screws have of late years found very little favour either in the Navy or with the designers of high-speed steamers of all sizes. Even in the case of the huge Cunard steamers " Lusitania " and "Mauritania," with their great power and small diameter of pro- SCREW PROPELLER BLADES. T63 peller, three blades only were at first fitted. After several voyages, however, it was found advisable to reduce the pressure per square foot on the blade surface, and for this and other reasons new screws having each four blades have been fitted, and proved advantageous. With practical sea-going engineers of the mercantile marine, young and old, it is an axiom that a sea-going, that is, ocean-crossing, ship, must have four-bladed screws, and that monster with its four great wings has almost become a fetish to many of them. They claim, and rightly, that the loss of a blade is a shortage of 25 per cent, only as against 33 of the three-bladed screw, but, on the other hand, with twin-screw ships and three-bladed propellers the loss of a blade means a shortage of only 16'5 per cent, of total effective surface. A modification of the percentage of loss is requisite here also, from the fact that in practice the surface of the four-bladed screw is usually larger than that of the three-bladed, all other things being the same. It is also necessary to bear in mind that with the larger surface there is the greater loss from friction ; and further, that with the increase in number of blades there is the inevitable increase in the resistance to rotation. It will be seen, then, that for the same reason that the two-bladed is superior to the three, the four-bladed is inferior to that screw. But in rough or moderately rough weather the four-bladed screw is the better for single-screw ships, even when of comparatively small sizes. For twin, triple and quadruple, in which, owing to the screws being of smaller diameter and their centres lower, the propellers are better immersed, the three-bladed screw is the best for all sorts and conditions of service. In small ships, especially those employed in sheltered waters, where the waves are only of small dimensions, even two-bladed screws of the Griffiths type may be used with advantage. In very small vessels two-bladed screws may always be used. The shape of screw blades best suited for each type of ship and condition of work is a problem that may be worked out some day by some one whose leisure equals his knowledge of mathematical science. Judging by everyday experience, however, there does not appear to be much scope for anything but small refinements, and it is very doubtful if such limited differences seriously affect the efficiency of a screw for any other reason than a quantitative one of acting surface. It may be taken as an axiom that no screw is satisfactory that works with excessive vibration, that is, which itself produces vibration 164 MARINE PROPELLERS. in the hull of a ship distinct from the vibration set up by the un- balanced moving parts of the engine itself. It was found in old days that while a screw with broad tips, such as the " common screw/' could drive the ship at good speed and be ■ otherwise efficient, the vibration was always high compared with that of the narrow-tipped screws of Griffiths. Hence it is another axiom that the propeller, to be successful, must have either a rounded end or a narrow tip to the blades. These axioms apply more forcibly to the propellers of excessive diameter so often found in ships of all kinds up to twenty years ago than to the modern ones, whose diameter is generally erring on the side of smallness. It is therefore now the practice with such screws to broaden the tips very considerably, but never to make them square. The maximum breadth was usually at a distance of \ to \ of the diameter from the axis; the breadth at the boss was somewhat less, generally about ^ths of the maximum, the length of the boss being such as to take this. The breadth of the tip would be about five to six tenths of the maximum. A curve then would be drawn through the terminals of these breadths, that portion between the boss and maximum having a much larger radius of curvature than the portion from the maximum to tip. Griffiths' rules for screw blades are very precise and were closely followed for a great number of years ; and bearing in mind that they were and are prescribed for two-bladed propellers, they do not in essence differ from the general practice of to-day. They are (vide fig. 38) as follows : — Diameter of the boss to be from -^ to \ the diameter of screw. Diameter of the flange of blade to be | the diameter of boss. Thickness of the flange of blade to be ^ the diameter of boss. Width of blade at widest part to be ^ the diameter of screw. Width of blade at tip to be \ the diameter of screw, Thickness of (bronze) blade at root to be ^ the diameter of screw. Thickness of (bronze) blade at tip to be \ the thickness at root. Diameter of shank to be -\ the diameter of screw. Curvature of blade forward to be 2 \ the diameter at the tip. The form of blades for a propeller having three on modern lines will be found by the following rules :— S/ is the surface ratio ; the diameter of boss must not be less than three times the diameter of shaft. SCREW PROPELLER BLADES. 1 65 Maximum breadth of blade at a distance of 0'25D from the centre — ~~ of the diameter of screw. I'oo Breadth at tip = 0'45 x maximum breadth. If the propeller has four blades, Maximum breadth =—^- of the diameter of screw. 1'62 The acting surface of a screw propeller may be found by the formula already given for determining the thrust of a screw, viz., Thrust in pounds = Dx ^ xV2 xG, where D is the diameter in feet, A the total area of the acting surface of all the blades in square feet, V the velocity of the screws in feet per second, P r the pitch ratio, and G- a factor depending on shape of the blade. (Vide page 124.) TxP, \ 2 Ride 1.— A = DxV 2 xG As some of the above figures are not always available, and as the amount of surface is modified somewhat by the form of the ship, and also the position of the screw affects its efficiency, the following empiric rules may be followed in calculating the surface, etc., of screw pro- pellers so as to give satisfactory results agreeing with good practice. Rule 2. — Area of acting surface = K* / '. — — — . V Revolution K = Mx prismatic coefficient of fineness of ship. The prismatic coefficient must not be taken at less than 0'55. For four-bladed screws, single screw, M is 20, and for twin screws 15 „ three- „ „ „ 19, „ „ 14 „ two- „ „ „ 17'5, „ „ 13 Example 1. — A twin-screw steamer whose speed is 20 knots ; her engines develop a total of 6000 I.H.P. at 150 revolutions to get this speed; her prismatic coefficient of fineness is 0'600, and her screws are to have three blades. In this case K = 14 x -6 - 8*4. Area of each screw = 8*4. /??°9 = 37'55 square feet. V 150 I 66 MARINE PROPELLERS. Example 2. — A single-screw steamer whose prismatic coefficient of fineness is 072 is steaming at 15 knots with engines running at 75 revolutions and indicating 4500 H.P. Her propeller has four blades. In this case K = 20 x 072 = 14*4. Area of screw blades = 14*4*/ 15rY = 111-5 square feet. Example 3. — A turbine steamer having three screws is to steam at 25 knots with turbines running at 500 revolutions per minute and developing a power equal to 12,000 I. H.P. Her prismatic coefficient is 0-56. Here K = 14 x 0-56 = 7'84. Area of surface of each side screw = 7*84 */ = 22*2 square feet. Example 4. — A steam yacht whose speed is 12*5 knots is to be driven by a two-bladed screw making 250 revolutions per minute and indicating 750 I.H.P. Her prismatic coefficient is 0"58, Here K = 17-5x0-58 -10 -15. Area of screw blades = 10'15/y/ h^.=17'55 square feet. In these rules and examples it is taken for granted that the diameter will not be abnormal, that is, that it will have been deduced from the rules laid down herein for diameter (see p. 171). On the foregoing basis the following holds good: — Hide. — Maximum breadth of blade in inches = N. / - _"2ifj "V devolutions' The value of N for a four-bladed screw is 14, for a three-bladed 17, for a two-bladed 22. (a) The maximum breadth of blade in Example 2 will be Maximum breadth = 14 Jy 1529 = 547 inches. ^ 75 (h) That in Example 4 : Maximum breadth = 22.7 Z5? = 31*6 inches. V 250 The diameter of boss was in the early screws quite small ; in fact, a mere cylinder of metal of sufficient thickness to resist the tear SCREW PROPELLER BLADES. J67 of the blades. Griffiths was the first to recognise the futility of such bosses and to perceive that the blade at its root was so nearly fore and aft that its propelling power was quite small compared with its resistance to rotation. Sunderland and others had recognised the loss from this cause, and proposed to remedy it by cutting away the blade till it became a mere stem or root of the acting surface, which was in their case quite remote from the boss. Griffiths removed the super- fluous and harmful portion of the blades by enclosing it in a huge spherical boss which displaced the water, which otherwise was merely churned about and made to produce eddy currents. The Admiralty tested the effect of a large boss by an interesting trial with H.M.S. "Conflict" of 1750 tons in 1853. Table XVI — Trials of H.M.S. "Conflict" with different Bosses. No. 5. No. 6. No. 7. Number of Trial. Plain Boss. Large Boss. Plain Boss. Speed . knots 8'837 9-425 9*424 Revolutions 73*0 77-0 75 75 I.H.P. 752 784 812 Slip . . . per cent. 19'08 18-18 16-84 SW-fLH.P. 133-4 154*5 149-8 A large spherical wooden boss was fitted to the boss of the common screw with the leading corner of its blades cut away. The 6th and 7th trials were run in smooth water on a calm day, and the 5th in slightly windy weather. It will be seen by these figures that there was a distinct gain with the enlarged boss. This test, together with the adoption of loose blades, whereby a large boss is necessary, caused the Admiralty to adopt the large sphere, while the mercantile marine with its solid cast-iron propellers stuck to the comparatively small egg-shaped boss, and, as a matter of fact, practically do so now, for it is only in the case of loose blades that a large boss is found on a merchant ship. The diameter of the boss with loose blades depends very much on the number of blades ; its size therefore should be determined by the following formula : — Diameter of boss = 0'5\/ diameter of screw x number of blades. Example. — To find the diameter of boss suitable for a loose-bladed I 68 MARINE PROPELLERS. propeller 16 feet diameter, (i.) with two blades, (ii.) three blades, (iii.) four blades. (i.) diam. = 0*5 J 16x2 = 2*83 feet, (ii.) „ -0-5 Jl6x3 = 3*46 „ (iii.) „ =0-5 JlQxl = 4*00 „ Eor the small diameter screws with high power now common in the naval and mercantile express service the multiplier should be 0*7 instead of 0*5, so that had the example been for such ships the diameters would be 4*0, 4*85, and 5*6 feet respectively. This means that the diameter of boss with loose blades should be not less than 2*7 X diameter of screw shaft for three blades and 3*2 x diameter of screw shaft for four blades. The length of boss should not be less than twice the diameter of the screw shaft, and with a solid shaft it should be 2*5 x diameter of shaft. The diameter of boss of a solid screw should be not less than 2*25 times the diameter of shaft; nor the length than 0*2 x diameter of the screw, exclusive of any nutshield in rear or gland cover in front. It should then be 2*2 to 2*5 x diameter of shaft. Elongated Bosses.^It has been pointed out that so early as 1851 one Eoberts suggested putting a conical tail to the boss so that there should be no loss there from eddies, and to prolong the boss forward so as to come in line with the ship's form and so avoid loss. It was very many years after that Thornycroft adopted this method for increasing the efficiency of the screw. CHAPTER XII DETAILS OF SCREW PROPELLERS AND THEIR DIMENSIONS. The Diameter of the Screw Propeller is one of the three features by which its efficiency is governed, the others being pitch, acting surface and shape of blade, or rather the disposition of the surface. Consequently it is easy to understand that there may be a considerable variation in the value of each, without much disturbance of efficiency in any screw running free from the influence of external bodies such as the ship herself, or the sea bottom or river banks near which it operates. For particular cases, such as a tug-boat, where thrust per se is of first consideration and therefore large diameter necessary, and in the case of turbine-driven screws where large diameter is impossible on account of excessive peripheral velocity, this feature of a propeller must be the leading one ; in general cases there may be almost any reasonable diameter of screw without violent variation in efficiency, provided the other features are modified to suit as in the formula : — m , 1 . ., diameter x ./area of surface xv 2 xf Thrust in pounds = - — — — -y~~ : — - — - — - — ^ pitch ratio and the thrust v and /are assumed to be constants. In the past the disposition, due largely to Eankine's teaching, has been to fit a screw of as large a diameter as the ship's arrangements permit; to-day the tendency is quite the other way, and it is now the practice of the expert engineer to try and find which of the many sizes is most suitable for his particular ship on the ground of giving the highest speed to the ship, or of propelling the ship with the maximum efficiency, which is not always the same thing, as may be seen by referring to the various trials whose records are given in Chapters XVII. and XVIII. 169 I70 MARINE PROPELLERS. The following, which are practically axioms, should be considered carefully in making a decision : — 1. Theoretically a screw of large diameter is most efficient. 2. A large screw, especially if of fine pitch, tends to make serious augmented resistance, which means loss of efficiency. 3. It generally runs with negative apparent slip. 4. The skin and edge resistance of the blades is large. 5. The tip of the upper blade is so near the surface as to cause breakage of water on slight disturbance, with consequent aeration of " feed " and loss of efficiency. 6. Liability to hit wharfs and quay walls, resulting in damage to blade tips. 7. A small screw well immersed has a higher efficiency than a larger one near the surface, so much so that even its propelling effect is higher. 8. There is not necessarily any specific relation between the disc area of screw and immersed midship section of the ship ; each may vary in its own way. 9. By special formation of the stern in the neighbourhood of the screw, the diameter may be considerably in excess of the draught of water. 10. The screw propeller is influenced as to efficiency by the position on the ship and the form of the ship. 11. The diameter must be modified to suit each ship. 12. The diameter of a twin screw may be less than that of a single screw, all other things being equal; that is, when driven by an engine of the same size, with the same boiler pressure, and being of the same pitch as a single screw, each twin may be of less diameter and correspondingly less pitch. In the case of quadruple screws, for similar reasons, their diameter may be modified considerably. Taking all these things into account it would seem that the pro- peller must be chiefly governed by the torque, that is, the turning power exerted by the engine, and be modified by the circumstances surround- ing it. It was customary at one time to arrive at the proper diameter of screw by taking the ratio between it and the length of stroke of piston, which ratio was and is now about 3 "5 in fast steamers and 4 - in cargo ships. Again, there is, as a matter of fact, a nearly constant ratio between the diameters of the screw and low-pressure cylinder, viz. 3*0 to 3*25, but the following rule may be taken, as giving a diameter of screw appropriate to the size of the engine and modified to suit the ship. DETAILS OF SCREW PROPELLERS. 1 7 I D is the diameter of low-pressure cylinder in feet ; when there are two L.P. cylinders of diameter D, then D = 1*4 x D. S is the stroke of piston in feet. Pc is the prismatic coefficient of the ship. Z is a multiplier = (2 '4 + Pc) for twin screws, and (2'7 + Pc) for singles. Eule 1. — Diameter of screw in feet = Z x >JD x S. (a) Example. — To find the diameter of screw of a yacht whose prismatic coefficient is 067 and engines with cylinders 18 and 36 inches diameter and 24 inches stroke. Diameter = (27 + 0*67) J3 x 2 - 8*25 feet, (b) Example. — What size propeller should an express twin-screw steamer have whose coefficient is 0*54 and her engines with cylinders 30, 45; 51 and 51 inches diameter and 36 inches stroke ? Diameter = (2*4 + 0-54) ^6 x 3-0 = 12*47 feet. (c) Example, — How large a screw should a tramp steamer have whose cylinders are 25 to 39 and 64 inches diameter and 45 inches stroke, her prismatic coefficient being 0*8 ? Diameter - (2'7 + 0*8) ^5*33 x 375 = 15*65 feet. (d) Example. — The diameter of the screws of a twin-screw cruiser whose coefficient is 0*57 and has engines having cylinders 43, 69, 77 and 77 inches diameter and 42 inches stroke. Diameter = (2-4 + 0*57) Jl'ix 6*4x3*5 = 16*54 feet. The size of engine is not always known in the early stages of the design of a ship, hence another method of getting an appropriate diameter is desirable ; moreover, the above rule is not applicable to turbine-driven screws. The following rule will give such appropriate or suitable size of screw that it may be followed in designing both ship and propeller. /I H.P Eule 2. — Diameter of screw in feet = ^x Pc ' ' K In this case the following are the values of x : — For single screws £ = 7"25, and for ocean-going expresses 7*61 ,, twin „ x=6*55 „ „ „ ,, 6*88 „ quadruple „ # = 6'25 „ „ „ ,, 6*51 Turbine-driven ) c _r c o , >- x = 6 55 „ „ „ „ 6-88 centre screws J ' Turbine-driven ) . -r c n . wing screws j '7 2 MARINE PROPELLERS. With this rule Pc must not be taken at a less value than 0*55, and for ocean-going ships whose normal speed is high and approxi- mating to that of full power, the value of x may be increased by 5 per cent, as shown above. Example (a). — The diameter of the screw of a torpedo boat whose prismatic coefficient is 0*55 and I.H.P. 2000 at 400 revolutions per minute = 3y 7-25 x '55^7^ = 6-8 feet. (b.) — "What is an appropriate diameter for a cross-channel twin- screw express steamer of coefficient 0*60, each engine indicating 3000 I.H.P. at 150 revolutions? 3/3000 Diameter = 6-55 x 0*6. /~^~=. 10-65 feet. V 150 (c.) — A triple screw Atlantic steamer whose prismatic coefficient is 0*65 has engines of 3000 I.H.P. divided equally between the screws, which run at 100 revolutions. The appropriate diameter of the centre is 7"61 x '65^ —^—=23 feet, and the diameter of each wing screw = 6-38 x and/ the limit of elasticity, is 16 51' For mild steel subject to shear, the limit of elasticity may be taken at 27,000 lbs. ; substituting this value for/ T = d 3 x5300. Now this means that a torque T will, if constantly applied, quickly destroy the shaft of a diameter d inches. If L be the maximum total load on the screw, having n blades, the load on each is -, and if the centre or resultant of the forces n composing the load be applied at a distance Y from the axis, the I 76 MARINE PROPELLERS. resistance to turning is LxY, and at danger point = d 3 X 5300. If it be supposed that the blade root is at the axis, the section will be in line with the axis ; and suppose it to be in shape a segment of a circle. If t be its middle thickness and b the breadth of blade, its power to resist bending will vary as b and as t 2 , so that L x Y = nU 2 x F. But Lx Y = d 3 x5300 ; then 7ixbxt 2 xF = d s x 5300; 5300 or nxbxt 2 = d 3 x- anc F 5300 / d* 53C Now, substituting E for ^- , I ~~d? thickness of blade at axis = K / x E V nxb The value of F has been calculated by taking the elastic limit for the zinc and strong bronzes and cast steel and 5 tons per square inch for the cast iron, and assuming the section to be a segment of a circle. Value of F is 1400 for Admiralty bronze, and . E is 3-80 „ „ 2600 „ manganese best and zinc . E is 204 „ „ 2500 „ phosphor bronze E is 2*12 „ 2400 „ cast steel . E is 2'21 „ 1000 „ cast iron . E is 53 N.B. — For a smoothwater or summer service or for screws seldom worked at full power E may be reduced by 20 to 30 per cent. Example 1. — What should be the axis thickness of the three blades of manganese bronze for a shaft 10 inches in diameter and a boss 25 inches long? Thickness = A/^^x2-04= JW2 = 5*22 inches. Thickness of Blade at Root. — The thickness at a distance equal to l£ x a from axis = — ^= — ^r x t\ and as approximately d = D~12 ) thickness at root = 0'75*/ _, X E. v nxb DETAILS OF SCREW PROPELLERS. I 77 Example. — What should be the thickness at the root of the blades of a cast-iron screw with four blades and a boss 42 inches long ? The shaft is 18 inches in diameter. Thickness = 0-75a/-HL x 5.3 ^ 0'75 VT84 = 1017 inches. v 4x42 The thickness at tip should be one-fifth that at root; that is, thickness at tip = 0*1 5 x t. Thickness through the centre section radially should be calculated at one or two points as a check on those obtained in the usual way by drawing a line from the point E at a distance CE equal to t from the centre to the tip A, making thickness at A = 015x*, and lying a curve so as to make the line AE a tangent to it. In the case of blades broad at the middle portion, as in fig. 51, the curve may come within AE as shown. Shape of transverse sections of a screw blade should be " ship-shape/' and designed so as to pass through the water with least resistance (vide fig. 50). For this purpose the maximum thickness or, as it were, the greatest beam, should be nearer the leading edge than the following ; that is, the " fore-body " is comparatively shorter than the (( after-body/' and should be in the proportion of 3 to 7, if possible ; for, judging by experiments with torpedoes, a long after- body is more necessary for high speed under water than a fine entrance. It remains yet to be seen if better results would not be got from true ship sections than by the present half elliptical sections necessary to get the flat acting surface. Such sections certainly would be better near the root of the propeller, as shown in fig. 52, where the plain line section BHAE is as made in accordance with practice ; the dotted line section BKAM is formed with the ordinary face line AEB as a centre and the thickness at corresponding spots the same. The part shown by the dash line HFE is as often done with large solid screws to do away with the shoulder which the back of the blade presents to the water on the passage of the blade through it. All sections near the boss may with advantage be formed so that the leading part may present a wedge form in the direction of motion instead of the round shoulder at the back of the blade. (See fig. 50.) The dimensions as above obtained may be relied on to give a pro- peller which shall be in accordance with good practice and have 12 . 7 8 MARINE PROPELLERS. A JFig. 51.— Solid Cabt-Iron Propeller Blade, DETAILS OF SCREW PROPELLERS. 179 practically the same factor of safety as the shaft. Full advantage, however, cannot be taken of the strength of the strong bronzes, in- asmuch as the blades, while not breaking, would deflect so much as to detract from the efficiency of the screw. Hence for Atlantic work the thickness given by these rules should not be reduced ; but, on the other hand, as already said, the value of E may be decreased by 20 to 30 per cent, for smooth-water ships or those using their full power but seldom. It may be well also to remind designers that blades even of zinc bronze do sometimes disappear and the engines may, especially in twin- and triple-screw ships, continue to transmit full power to the remaining blades for some time before the casualty is appreciated. D if*" — ^ B Fig. 52. — Various Root Sections of a Screw Blade. Fig. 53, — Longitudinal Sections of Typical Screws of Equal Area of Blade A little allowance for this contingency, even if remote, may be made and looked on as so much insurance premium. Fig. 53 shows the true shape of the longitudinal section of a screw blade if it is to have an uniform stress throughout the dotted line for a fantail and the plain line for an oval form of the same area. It will be found that in ordinary practice blades have the weakest part near the root, since they usually break there. It would there- fore be better to take metal from the middle and add it to the root in a general way as shown in these sections AC, BC, which are drawn to a larger scale horizontally than the blades to make its shape and curvature more clear ; it is common practice to draw them of the full thickness. If the size of the shaft has been arbitrarily fixed on, as is the case l8o MARINE PROPELLERS. in the Navy, when the shafts are hollow and enlarged to give stiffness to avoid vibration between the bearings, then d must be calculated by the following rules. Diameter of Screw Shaft. — If this is not given, or is otherwise not available, and if p is the boiler pressure, ^ revolutions Two-cylinder two-crank compound engines, K = 12 J p. Three-cylinder three-crank compound engines, K= 9'8 J p. Three-cylinder three-crank triple-expansion engines, K = 73 J p. Four-cylinder four-crank triple-expansion engines, K = 8"0 Jp Four-cylinder four-crank quadruple expansion engines, K = 8"0 J p. When the boss and blades are in one casting, as shown in fig. 51, the thickness of metal x around the borehole should be determined as follows : — 2^x^x/x(^ + y)-^x5300; 7 5300 3 , 2271 x = cl x — ,- x y = a x ~y~ ■ 9971 For cast iron /= 8000 and —j- =0"284. 0971 For cast steel/= 11,500 and ~„ =0197. For Admiralty bronze /= 10,000 and ^J^ = 0-227. For manganese and zinc /= 12,500 and ^ —= 0*181. 2271 / 2271 For phosphor bronze /= 13,000 and -—- = 0175. Thickness of boss, fore end = l'O x x. „ after „ =0 - 85 xx. „ „ outer shell = 0'74 x x. Solid screw versus loose blades is a subject that has been well thrashed out in the past. The efficiency of the solid screw in the mercantile marine was always, as seemed to be, higher than that of a screw with loose blades; and so it was in nine cases out often, for the blades of the latter were attached to the boss by means of a flange DETAILS OF SCREW PROPELLERS. IOI standing out from the boss and surmounted with the nuts of the studs and sometimes the stud ends. No attempt was made to shield them, or to mimimise the resistance. Then, too, the blades were often of steel and quite out of pitch, besides being rough on the surface, whereas the solid screw was made of cast iron smoothly cast, with the blades very fairly alike in shape and pitch ; the boss an ellipsoid without obstructions, and the blade roots " faired " to the boss. Some of the early naval four-bladed screws were rather rough and ready about the boss ; to-day they are as neat and well finished as the others. The loose-bladed screw as fitted in H.M. ships, and also in high- class express steamers, are as efficient as any solid ones, and possess the advantages that the loss of a blade does not condemn the whole screw, as a spare blade can be fitted by a diver of experience, so that to refit it is not necessary always to go into dry dock. With the ordinary cargo steamer, whose " light " line is generally below the screw boss, there is no need even of a diver, for all the blades can be examined and, if need be, changed without taking the ship out of water. If, however, a diver cannot be obtained and there is no dry dock convenient, the ship sometimes can be " tipped " so as to raise the boss sufficiently near the surface to get a blade off. With loose blades the screw may have the boss of cast iron or cast steel, while the blades are of the very best metal and make ; while the solid one must be all of one metal throughout, thereby adding to the cost if a costly metal is used. On the other hand, when a ship has been a long round voyage, she must go into dry dock or on a slipway to have her bottom cleaned and repainted. The opportunity then arises to examine the stern-shaft and its bearings ; this necessitates the taking off of the propeller, which is also carefully examined, whether it be a solid or loose- bladed screw. When a screw has loose blades they are often made of a different metal from the boss. Formerly in the Navy the boss was always made of Admiralty bronze, while the blades were sometimes of one of the other bronzes. At present the boss may be of any approved bronze. In the mercantile marine blades and boss were made of cast iron; with the advent of steel blades the cast-iron bosses still remained until some of the larger ones split from time to time ; and as the steel makers could make steel bosses, the practice to have blades and boss of that metal became a rule. To keep down 102 MARINE PROPELLERS. cost when the mercantile marine began to use bronze blades, the bosses continued to be of steel or cast iron, except in special cases when both were of bronze, the boss being of a cheaper variety of metal. In a general way there is no reason why the boss should not be of steel on a steel ship or even of cast iron on smaller ships, except that they corrode somewhat from the galvanic action unless care is taken to prevent it. The earlier way of fitting blades to the boss was by forming a shank through which a cotter could be driven, as shown in fig. 38. So long as there were only two blades this method was possible and good for bronze screws. With cast-iron blades it was necessary to have a flange formed at the base of the blade through which screw bolts were fitted, securing it to the boss. With large screws the strain on the bolts was very heavy, and blades sometimes broke away from the boss. It was then found necessary to bed them into the boss either by a central spigot or by dropping the flange itself into a recess formed in the boss to fit it, as shown in fig. 43. All shear was then taken from the bolts, and their function was almost limited to holding the blade down to its seat against centrifugal force ; the front ones, however, do hold the blade against the thrust and resist- ance to turning. The Admiralty engineers have always insisted on having the bolt heads recessed, so that when a cover plate is put over them the sphere is complete and the boss has no obstructions. But not so with the mercantile engineer; for he has no cover plates, even when the boss is spherical and the blades fitted in recesses ; he also often prefers steel studs having capped nuts of bronze fitted to them so that no water can get to them. Finally, when every nut has been " hardened up " and the steel set-screws tightened, a covering of Portland cement will be put on the lot and the semblance of a sphere attempted with this plastic material. It need hardly be said that the screws of express steamships, whether cross-channel or cross-ocean, are treated in a manner more approaching that of the Navy, and the more so since high-speed reciprocating engines and turbines have come into use. The number of bolts in a blade flange should, as a rule, be an odd one, so that there may be one more on the acting side than at the back, as they are always under stress when going ahead, while those at the back take little or none of the thrust load then, and only come on full load when the engine is going DETAILS OF SCREW PROPELLERS. 1 83 astern ; now, as the power developed in stern gear in most ships never exceeds half the full-speed power, there is so need for so many, even then. Consequently small screws have three and two studs or bolts to each blade, while the very largest screws have nine— five in front and four in rear. As a rough guide the following may be taken : — AT , £ , i, , 1 1 1 j diameter shaft in inches + 6 JN umber of bolts to each blade = o By this rule, with shafts up to 9 inches diameter there would be five bolts, above that size and up to 15 inches there would be seven bolts, and from 15 to 21 inches nine bolts. The diameter of bolts when made of steel, manganese bronze, zinc bronze, or naval bronze can be determined by the following rule : — Diameter of bolts or studs = . n Where d is the diameter of shaft as before, n is the number of bolts to each blade and Z is 1*6 for a three-bladed screw, 1*3 for a four-bladed one. Centrifugal force produces in the screw blade at all times some stress, and at high revolutions the stress becomes serious, so much so, in fact, that destruction of blades is due sometimes to this source with screws driven by turbines. Within moderate velocities the forces set up by inertia really tend to balance those by hydraulic pressure on the blade. That is to say, that whereas the hydraulic action tends to bend the blade in a direction opposite to that of revolution, the inertia of the blade tends to make it bend the other way as well as to « throw off." The forces acting on a screw blade due to its velocity can be calculated from the usual formula where W is the weight of a blade in pounds, r is the distance of its centre of gravity from the axis of rotation, g is gravity, and taken at 32, v the velocity in feet per second : — W Then C = — xv\ 9 W v 2 and the tension on the bolts = — x — , 9 r 184 MARINE PROPELLERS. v 2, — being of course the accelerating force, and called usually the centrifugal force. When a propeller is in motion on normal conditions running at E revolutions per minute, v= ^pitch 2 + (27rr) 2 xK-r60. As an example take the case of a screw propeller 12 feet diameter, 15 feet pitch, 200 revolutions per minute ; centre of gravity of blade is 3*2 feet from the centre; it weighs 1600 lbs. Determine the bending moment 011 the root distant 1*8 feet from the e.g. and the tension on the screw bolts screwing it to the boss. v= Jl&xJiirx'fZfx 200^60 = 84 feet per second. C = ^x 84 2 = 352,800 lbs. Tension on bolts = 352,800-:- 3 -2 = 110,125 lbs. If seven bolts, tension on each = 15,732 lbs. Bending moment due to = 352,800x1-8 = 635,040 foot lbs. This is, however, in a plane through the face at the c.#.,and there- fore is resisted by the section at the root longitudinally. Taking circular motion and no advance of the screw, v = 2tt x 3*2 x 200 -r 60 = 67 feet. Then C=~°x67 2 = 224,500 lbs. The bending moment on a plane at right angles to axis = 224,500 x 1-8 = 404,100 ft. lbs. Taking an extreme case of an Atlantic steamer driven by turbines so that each screw receives 18,000 I.H.P. at 180 revolutions, the diameter being 16*6 inches, the pitch 18 feet, the weight of each blade 11,200 lbs., its e.g. being 4*5 feet from the axis and 2'0 feet from the root. Here velocity = x /l8 2 4-( 7 r9) 2 x 180^-60 = 111 feet per second. n 11,200 111 2 in _ C ^^2~ X 2240 = 192 ° tOns - Taking circular velocity only, n 11,200 85 2 , iftft G= ^2~ X 2246 = 1129tons - 1129 Tension on bolts = - 7 -~ F ~ = 251 tons. 4*5 DETAILS OF SCREW PROPELLERS. 1 85 If thirteen bolts to each blade, the load on each =19*3 tons in addition to that due to the pressure on the blade. The weight of a screw propeller can, of course, be calculated ; but to do it accurately is a long and tedious job, and scarcely repays the trouble taken. It can be estimated with a fair decree of accuracy by the following formula : — Weight of a complete screw propeller in cwts.= surface x thickness at root surface being taken in square feet and thickness in inches. When made solid of cast iron K = 4*5 „ „ bronze K = 3'8 When fitted with separate boss and loose blades of cast iron K = 3*0 steel K = 2*8 bronze K = 2-5 CHAPTER XIII. GEOMETRY OP THE SCREW. As already stated, the acting face of the ordinary screw is part of a true helix, and the traces of it on concentric cylinders having their axes common to it enable it to be drawn in plan, and from this plan elevations can be projected on planes parallel to and at right angles to the axis. In fig. 54 AB is the diameter of the screw to be drawn at any convenient scale of which C is the centre. CQ, at right angles to it, is half the pitch. ADB is the half circle of tip, and GN, HM, JP are the ends of the other concentric cylindrical surfaces. This semicircle ADB is divided into equal sectors FAC, EFC, etc., and CQ, into the same number of equal parts AX, XY, etc. Project the point F on the base line AB and produce the line so as to cut the horizontal line through X at K. The point of their intersection K is on the trace of the helix on the outer cylinder ABA 7 . Point 1/ and all the other points on this trace can be found in the same way, and the spiral line ALOA' can be drawn through them. In the same way the trace GOG' on the cylinder whose radius is GC can be drawn ; as also HOH', JOJ', etc. The developed surface of the blade as decided on should then be drawn with C as the centre ; nn' is the portion of the blade inter- cepted by the cylinder HM so that the chord nn is the breadth of the blade seen on looking down on that cylinder. From the corresponding trace HOH cut off a portion hti equal to nn' ; and from the other traces cut off portions corresponding to those intercepted by the other circles GN, AD, etc. Now draw a line on each side through the terminals of these cut- off portions and the outline of the blade in plan is made at hOJi. Now, with Q as centre, draw the circles representing, as it were, GEOMETRY OF THE SCREW. 187 the other ends of the cylinders ; through h and h' draw lines parallel to CQ, then produce them at right angles to QA', cutting cylinder HM end at k and ¥. Then k and Jc are points on the end elevation Fig. 54. —Method of Delineating a True Screw Accurately. or projection of the screw, m and mf are found in the same way, as also any other points. Lines drawn through these points ink, etc., m'k\ etc., will be the outline of the screw as seen looking in line with its axis. I 88 MARINE PROPELLERS. Now t¥ is the longitudinal view of the trace on the cylinder HM, so that by drawing lines hr and h'r at right angles to the axis CQ, tv' will represent in side elevation the line hhf. In a similar way all other points can be found ; and lines drawn through them will give the outline in side view on the projection on a plane parallel to the axis of a blade whose developed form is nDn'. To show the projected width of the blade at mm' it is necessary to suppose the blade to be twisted round till that portion of the blade is parallel to the plane. The points mm will then remain on a line drawn through them, and qq' will be the terminals and equal to ?iM.n in length, but the curvature of qq will not have its centre at Q as being part of an ellipse. In everyday practice, and more especially when dealing with propellers whose blades are comparatively narrow, so that no violence is done to the design or calculation, the portion hh' is assumed to be a straight line. If the screws have very broad blades and accuracy is demanded, then the above methods must be observed. The rough and ready way usually observed in many drawing offices is illustrated by fig. 55. Hence AB is 7r x diameter at tip, and AD is the pitch, both drawn to quite a small scale, say J inch to the foot for moderate size screws, and £ inch to the foot for very large ones. Divide AB into eight equal parts at D, E, F, etc. Join D, E, F, etc., to and produce them beyond ; then AOD is the angle of the blade at £ the diameter, AOG that at |- the diameter, etc, AOH that at f. Draw DjMLM as the developed acting surface of the proposed screw. The breadth on circle H 1 is MM X . Cut off a portion mm of HOM equal to MM 1 . Project this back on circle H p so as to intercept NN X . Carry out this operation at each of bhe other circles LKJ, etc. The result will be that on drawing lines through the terminals, the projected surface D 1 NLN 1 will be formed, which is that of the screw blade on a plane at right angles to the axis. By projecting from the plan horizontally to a series of circles D u E n F u , etc., in the same manner the surface is projected to a plane parallel to the axis. It is, of course, understood that to be correct MNN 1 M 1 should be all in one horizontal line, because point N is GEOMETRY OF THE SCREW. 189 supposed to be the position of M when the blade is twisted to pitch from a transverse position. Varying pitch may mean that in a complete revolution the pitch has changed from a> to y ; so that, at one edge, the entering angle is that due to a pitch x ; while, at the other or discharging A D E F G H J Fig. 55. —Simple Method of Delineating a Screw. angle, it is that due to a pitch y. Woodcroft's screws, when of one convolution, were formed in this way. The method by which to draw the trace of that screw is shown in fig. 29, and described on page 106. Through "the points APS, YXW, the curve drawn is the trace of the blade tip of a screw of one convolution of a screw whose 190 MARINE PROPELLERS. pitch varies from BL to BC as developed on the enwrapping plane. In practice, with screws having a length of from | to I the pitch, it was, and is still, usual to form a leading portion of the blade, say from one quarter to one half of the breadth, with a pitch corresponding to a speed slightly greater than that expected of the ship, and the remainder greater and sufficient to give the necessary acceleration. The blade face can then be drawn as shown in fig. 56, where AB represents ir X diameter and BD the pitch at entry, and BC the pitch at delivery. Join AC and AD, when the angle of the blade at entry is BAD and at delivery BAC. Produce AC and cut off AF and AE in the proportion decided on. Then apply a curve so that AE and AF are tangents to it and the line EAF is typical of the blade surface. Screws are generally made with a rounded back to the blade section, so that it is formed from a part of an ellipse, or an ellipse-like curve as shown in fig. 50. But some- times the back is practically a t ' • t^-j. i. tv, j pair of planes inclined to one -Increasing Pitch Blade. r r another, also as shown in fig. 50. It is claimed for the latter that a stiffer blade is obtained with the same weight of metal as in the round backed ; and, further, that it offers less resistance in its passage through the water, especially at high revolutions. When formed in this latter way it is usual to make the maximum thickness slightly greater. So that in formula (1) (see p. 174) the constant 13 becomes 15 ; and 0"45 becomes 0'40 in formula (2). Hollow Castings. — Messrs J. Stone & Co. have for some time been in the habit of making the root end of large propeller blades hollow. A careful moulder can do this with little or no risk of a waster, or even of a faulty casting ; and as the metal by slow cooling at the very thick part is always highly granular and often spongy, the withdrawal of it means really 110 loss of strength, while it does mean less weight and cost. Root Sections. — It is not an uncommon thing for the leading edge near the root to be shifted from A to F so that the section there GEOMETRY OF THE SCREW. 191 does not present a broad round shoulder in the direction of motion (E to A) as shown by AH, % 52, p. 179. The suggestion that propellers would be better made so that what is now their acting surface, as at AEB, should become the middle of the blade section, is well worth a trial on a ship of fairly good size. In fig. 52 BKAM would be such a section at the root of the screw instead of BHAE as at present, or of BHFE, as has sometimes been adopted. To found a propeller the helical surface is formed on a loam facing to a rough brick foundation, or it may be on common green sand, by a strickle board having an edge of metal formed to the shape of the acting face of a longitudinal section of the proposed screw and made to travel on spiral guides, one just beyond the tip and the other near the boss, while it revolves around a fixed bar standing at right angles to the plane or the Hat cast-iron bed on which the mould is built. The spiral guides are made of triangular pieces of plate, whose bases to height are in the same proportion to the circumference of the circle on which they stand to the pitch of the screw ; they are bent to a portion of that circle. When the bed is moderately dry, a centre line is inscribed on it corresponding to the centre of the blade. The blade breadths are then set out on inscribed circular lines, and to their extremes is applied a piece of square bar iron bent to shape so as to lie evenly on the mould. Lines are then inscribed from it and the blade surface form is completed. Parting sand, quite dry, is then sprinkled on the mould face, and pieces of thin wood cut to the shape of the sections at various points are laid on the face and the spaces filled in with " black " sand ; while over the whole a coating of loam or green sand is laid and the mould completed in the usual way. When dried or baked, as the case may be, the moulds are opened and the black sand and sections removed, and after cleaning up, fairing, and rubbing smooth, the surfaces to be exposed to the molten metal are coated with " blacking " or plumbago mixture ; they are again put together, carefully set, and otherwise dealt with as other moulds are. If a blade only is required, as in the case of a loose-bladed screw, a flange is fixed at the end of a blade mould of the form, etc., suit- able to it constructed as above. If it is a solid screw there are wing boxes, one for each blade, and they are set at the proper angles with a boss mould in the middle of them. Pattern Blades. — In old days it was a common and costly practice 192 MARINE PROPELLERS. to make a pattern blade in mahogany, and as samples of pattern-mak- ing they were very beautiful ; but by the method above described almost as good a pattern can be obtained in cast iron, and certainly a cheaper and more durable one. Having a pattern, any number of blades may be made from it in dry sand. Great care, however, is necessary in the pattern-making as well as in moulding to ensure getting quite true and trustworthy blades and thus to avoid having to chip, file, and grind the acting GEOMETRY OF THE SCREW. 1 93 surfaces to get the best performance on high revolution trials, all costly and slow and unsatisfactory processes, and often ending in having to accept a pitch different from the exact one designed rather than condemn an otherwise good screw to the scrap heap. Methods of ascertaining the pitch of a propeller have been from time to time devised, and special instruments made whereby the approximate pitch is seen at a glance ; but none can exceed in simplicity, accuracy and usefulness the apparatus shown in fig. 57. The radial batten is of dry, well-seasoned wood which is practically unaffected by atmospheric conditions, such, for example, as mahogany or cedar. The metallic fittings may be made, with advantage, of aluminium, now that metal is so cheap (£70 per ton). The dial plate is marked off in degrees, and each angular movement usually is 5 degrees. The horizontal staff or measuring rod is of mahogany, fitted with a metallic tip so that it touches the blade face in line with the edge next the batten and capable of sliding in the guides on the batten. It can be with advantage marked in inches and fractions of an inch, but if the angular movement is always 5 degrees the staff may be marked with one-sixth of an inch as the unit, as then each unit for that angular movement means a foot of pitch. A useful addition to the installation is a metallic scale marked on the four edges with units to suit four different angular movements and arranged to be secured to the measuring rod at the part where it slides in the guides attached to the batten. A reference mark or score is on each guide for this purpose. 13 CHAPTER XIV, MATERIALS USED IN THE CONSTRUCTION OF THE SCREW PROPELLER. The very earliest of the screw propellers were often made of oak staves cut to shape and through-bolted to one another so as to have the appearance of a double spiral staircase. The (( steps " at front and back were, however, " dubbed " away down to the angles so as to form a true helical surface. The patterns for the early screws were also made in this way, for there was then formed a two-bladed screw like those used by Smith and Woodcroft. To employ oak or elm timber was of course quite a natural thing to do, seeing that the floats of the paddle wheel, as also the big sweeps or oars, even then often used in sailing vessels as propellers, were of such wood. The frictional resistance of the surface of these hard woods would be enough to condemn their use, even had not the inherent weakness of their structure already done so, as soon as the diameter of screws was increased beyond the model limits. One good service was rendered by the wooden screw, however, before it was put on one side, when a portion of Smith's early ones broke off, the surface was so reduced as to permit of greater revolution and more speed, thereby teaching engineers that a screw could have too much area of blade, as well as to demonstrate the magnitude of the evil. Naval screws of bronze was the rule from the outset, and this was no doubt due in no small measure to the fact that naval ships were in the ef.rly days of the screw built of wood, copper fastened, and always sheathed with copper. Even to-day ships intended for the parts of the world which have inadequate dry dock accommodation are sheathed with teak planking, on to which the copper sheets are nailed, as they were to the old wooden ships. In the 'sixties of last century, when iron ships were taking the place of wooden ones for naval purposes, the practice of fitting bronze screws still obtained ; and 194 MATERIALS USED IN CONSTRUCTION OF SCREW PROPELLER. I 95 although fears of corrosion of the ships' plating were often expressed, they were never realised sufficiently to condemn the practice. As a matter of fact, there was sometimes corrosion in the immediate neighbourhood of the screw race, which, in some few cases, took the form of pitting and developing pustules under the red lead paint. The fitting of a few slabs of zinc in metallic contact with the hull is, however, a sure preventative of corrosion. The zinc bronzes and manganese bronze do not affect iron in the same degree, so that when the boss as well as the screw blades are of that metal there is no need for the protecting zinc slabs. The Admiralty have continued to use "bronze" with very few exceptions ; but there is not now the same limitation in proportions or ingredients that formerly obtained ; for whereas then the " bronze " or "gun-metal" meant a composition of best selected copper with about 10 per cent, of tin, now it may have in addition to tin, zinc and some other metals, and in some cases be even without tin. The following are the metals, or rather alloys, called " bronze," used for propeller blades, etc., in H.M. Navy, and also largely so in the mercantile marine for express steamships where safety and efficiency are carefully studied. Admiralty bronze as used for propeller blades, bosses, etc., in the Navy is now a mixture of — Copper, 87 per cent.; tin, 8 ; zinc, 5. When carefully alloyed and cast the ultimate tensile strength should be 15 tons per square inch with test bars. The average strength of pieces cut from castings should not be less than 13 J tons with a stretch of 7 J per cent, in 2 inches. Its specific gravity is 8*66. Phosphor bronze, a metal sometimes used for the blades of ships, is an alloy of copper, 82*2 per cent.; tin, 12-95; lead, 4*28; phosphorus, 0'52. It is harder, closer grained, and stronger than Admiralty bronze ; its ultimate strength is 17 tons and sometimes higher per square inch ; but the most important characteristic of this metal is that its elastic limit is very high ; in fact, higher than the ultimate strength of Admiralty bronze. It has also a high resistance to shear. It is therefore very suitable for blades, and especially for the bolts used to secure the loose blades. Aluminium bronze, containing copper, 95 per cent. ; aluminium, 5*0, possesses great tensile strength even in the cast state, amount- ing to as much as 25 tons per square inch with an elongation of 42 per cent., while that with 7'5 per cent, aluminium and 2*0 per 196 MARINE PROPELLERS. cent, of silicon has a still higher tensile strength and an elongation of 25*6 per cent. It has been used for propellers, but only to a small extent, as in cost and physical qualities it cannot compete with what are called zinc bronzes or with manganese bronze. Manganese bronze, one of the zinc bronzes, was very gener- ally used for propeller blades; it was discovered and perfected by the late Mr Percival M. Parsons, and castings of all sizes were made by S.S. "Norman." Fig. 58. — In Dry Dock, after stranding od Coast of Africa, showing bent Propeller Blades of Parson's Manganese Bronze. him, and are now by the company he founded. It has an ultimate strength of 30 tons per square inch with an elongation of 21 '5 per cent., and is tough. It makes very good bolts, and withstands the action of sea-water most successfully. The first blades were made in 1879, and the screw of H.M.S "Colossus/' after twenty-four years 1 use, showed no signs of corrosion. The Parsons Manganese Bronze Company now make a special bronze for screw propellers which possesses in a high degree the qualities necessary for a successful screw blade, and the extent of MATERIALS USED IN CONSTRUCTION OF SCREW PROPELLER, igj punishment such screws can successfully withstand may be seen by examining figs. 58, 60. Stone's bronze is now one of the best known of the " zinc bronzes," which are so called because of the large proportion of zinc in their composition. This alloy has in ingots and cast- ings of any size, as supplied by J. Stone & Co., copper, 56*1 per cent.; zinc, 40*6; tin, 1-05 ; iron, 1*67; lead, 047- Messrs Stone's No. 3 bronze is a very remarkable alloy, having physical pro- perties which make it very suitable for such things as propeller blades. It has the very high ulti- mate strength of 35 tons per square inch with an elonga- tion of over 30 per cent. This metal, therefore, is exceed- ingly tough, and when a blade receives a heavy blow, it generally bends without breaking, and often can be re-set on being heated and carefully handled (fig. 59). A bar of this metal an inch square placed on supports 12 inches apart will sustain a load at its centre of 5750 lbs. without fracture, with a deflection of 1*01 inches ; 6820 lbs. were required to break it down to 2 - 32 inches. The following figures show the progress with similar bars cast in sand from crucibles : — Table XVII. Fig. 59. -Blade made of Stone's Bronze, and bent by accident. Load, lbs. 3000 3500 4000 4500 ! 5000 6500 j 6000 i 6350 Deflection, inches . 0*18 0-26 0-39 0-57 j 0-82 1'18 1-65 Rupt. i 9 S MARINE PROPELLERS. Bars of the same size and length — If of cast iron, give way at about 2000 lbs. „ gun-metal, „ „ „ 2600 to 3000 lbs. „ cast steel, ,, „ „ 6100 lbs. The elastic limit of No. 3 is 17'05 tons, or 47'3 per cent, of the ultimate strength, viz. 35 tons. Tests made by Professor A. K. Huntingdon on the resistance to torsion of this metal are interesting. A bar fths inch diameter and 54 inches long was subjected to twisting, with the following results : — Table XVIII. Torque, Inch lbs. 1195 2170 3250 4340 5425 6510 6980 Angular movement, degrees 1 3'60 i 6-84 1 23 76 64*05 133-9 222-1 Rupt. A bar of this metal I inch in diameter and t inch loner will withstand compression up to 20,000 lbs. with a reduction in length of 0*174 inches. With a load of 5000 lbs. it is only 0'004 inch. It took 25,000 lbs. to break it, or nearly 56 tons per square inch. It may, therefore, be stressed in compression up to 11 tons per square inch under test with safety. Naval brass is an alloy of copper, 62*0 per cent. ; zinc, 37'0 ; tin, 1'0, and is a strong, tough metal often used for studs and bolts. It can be rolled and forged, and has an ultimate tensile strength of 24 tons per square inch with a stretch of 10 per cent, in two inches. Its elastic limit is about 42,000 lbs. ; its resistance to shear, however, is only 55 per cent, of the tensile strength. Bull's metal is very like Stone's in composition, but has no iron ; its character is good. Delta metal is another of the zinc bronzes; it has, however, 2-0 per cent, of manganese, but no tin. It also has a high tensile strength, and is tough. Sterro metal, the earliest of these bronzes, consists of iron, 1-5 per cent. ; zinc, 38*1 ; copper, b'O'O ; and has similar qualities to the others. But all these zinc bronzes, while possessing very high ultimate strengths, have an elastic limit of only about 14 to 15 tons per square inch, and their resistance to shear is also comparatively low. They are, however, comparatively cheap and easy to machine. They MATERIALS USED IN CONSTRUCTION OF SCREW PROPELLER. 1 99 can be smithed when red hot and be re-melted, but great care must be taken in smithing, as is also the case in melting both new and old metals, in order to get the best results. There is no secret in their composition, as it is known and has often been published ; their good physical qualities depend rather on the mixing, melting, and general heat treatment than the exact quantities of: the ingredients. Messrs J. Stone & Co., the Parsons Manganese Bronze Co., and Billington & Newton, have made a special study of these bronzes ; this, together with the great experience they have had in the moulding and casting of blades, ensured their turning out and supplying a thoroughly reliable propeller which has clone much to popularise the adoption and general use of bronze blades, in spite of their high cost compared with that of cast iron. Oast iron was the material of most of the screws of the mercan- tile marine from the earliest days, and has continued to be largely used even to-day. It is, of course, the cheapest of the metals and the easiest of manipulation, and can be got of fairly good quality almost anywhere. It is claimed for it as a virtue that when it receives a serious or damaging blow it breaks clean off and produces no obstruction ; for a propeller blade this is perhaps an advantage, especially in the case of the single-screw ships, for when struck a heavy blow from wreckage, it should part off and sink rather than bend out of place and perhaps prevent the propeller from turning round in the frame, thereby disabling the ship. Admiralty bronze blades have been in the past quite accom- modating in this way, and have performed the feat more than once of breaking off and disappearing. The manganese and zinc bronzes, on the other hand, do bend, and that very considerably, under a severe blow ; and many a blade of these metals has suffered terrible distortion without breaking (see fig. 60) ; but it must be a very rare event for a bend to be so bad as to foul the stern or rudder post of a single screw, and impossible to foul the brackets of a twin-screw ship. Good modern cast iron propellers can be obtained whose ultimate strength is 10 to 12 tons per square inch, and as its yield point is nearly the same, its strength will compare favourably with that of Admiralty bronze, whose yield point is only seven tons; but unfortu- nately cast iron does not stand sudden heavy or continuous shock so well, and does not bend at all. But cast iron has two other characteristics which go far to condemn it as a metal suitable for propellers, especially for those moving at 200 MARINE PROPELLERS. high velocity. It rusts or oxidises quickly when exposed to sea water, especially to the aerated sea water abounding at the backs of the blades ; and when oxidised in this way its surface becomes very rough, and hence causes great frictional resistance after a few weeks' service. It is true the blades may be cast quite smooth and coated with enamel paint, and that when so done the skin resistance is no greater than that of a bronze screw ; it is generally true, however, that in practice it is difficult, if not impossible, to keep the paint on the outer part near the tips, just where resistance is most serious, and still more so on the backs. Further, it is quite impossible to maintain the leading edges of the blades so thin and sharp as is the case with bronze ones, even when the founder manages to cast them so. And, moreover, from the iron blades being so much thicker than bronze, their resistance to turning is much greater. Some four-bladed cast-iron screws were tried by the Admiralty on H.M.S. (( Victor Emmanuel " in 1853, but they were seldom used after. Cast steel was intro- duced to the shipping world as a suitable material for large propeller blades by Messrs Vickers & Sons in 1870, and for many years that eminent firm supplied most of the large Atlantic liners with their products. The Admiralty gave this material a trial on a few ships whose bottoms were sheathed with zinc instead of copper. As, however, the zinc was not found to last long, its use was discontinued ; and as steel did not agree with copper sheathing and bronze castings, blades of it were soon discarded from the Navy. When first placed on the market, steel blades were very costly, almost as much so as bronze ones, especially when their extra weight was taken into consideration, together with the extra cost of machin- ing such hard steel as was then used. Fig. 60. — Manganese Bronze Co.'s Propeller Blade, bent, from s.s. "City of Paris." MATERIALS USED IN CONSTRUCTION" OF SCREW PROPELLER. 201 The cost of steel blades, however, came steadily down after about 1885, and eventually could be bought as castings at a price not very much higher than twice that of iron castings, consequent upon the use of the Siemens furnace for melting as against the crucibles of Vickers. In spite of every precaution the steel blades corroded even worse than the cast iron ones had done ; they were seldom cast of the form or pitch designed, and were nearly always of coarser pitch at the boss than the tip; in fact, no two blades were sufficiently alike for good working. Even those from the best makers were considerably inferior in the matter of correctness of pitch and being helical to the bronze or even to cast-iron ones. The machining and finishing of steel blades was, of course, always much more costly than bronze and cast iron, so that altogether, in the end, they were not really so cheap ; and seeing that, as scrap, they were practically valueless, whereas the bronze blades were worth quite half their original cost, the latter gradually drove the steel out of the market. There are, of course, a considerable number still remaining in use, and steel founders work now more closely to the design than was formerly the case ; they cannot, however, give the fine sharp edges of the bronze blades, or even equal the cast iron ones in this respect or in smoothness of surface, etc. Forged steel blades have been made for the very high speed of revolution on torpedo boats and " catchers " of years gone by, when it was necessary to have blades as thin as possible consistent with strength and stiffness. It need hardly be said that such propellers were very costly, and as liable to corrosion as cast steel and cast iron. When so made they were often fitted to a wrought steel boss by means of grooves planed in the latter and held by wedges, as shown in fig. 61. Bronze blades were also fitted in this way to steel bosses, as such blades could then be cast very accurately, and any want of accuracy was removed by grinding. Moreover, with such a boss quite small propellers may have detachable blades. Fig. 61.- -Blades Dovetailed into Forged Boss. TRIALS OP S.S. CHAPTER XV. ARCHIMEDES " AND R.M.S. 'RATTLER." The screw steamer " Archimedes," built for the Ship Propeller Co., a syndicate of F. P. Smith and his friends, was a sea-going ship designed by Pascoe and of the following dimensions : 106'7 feet long between perpendiculars, 21'8 feet extreme beam, and 13 feet depth of hold. Her tonnage was 237, draught 10 feet aft and 9 feet forward. She was propelled by two engines of 90 H.P., having two cylinders 37 inches diameter and a stroke of 3 feet. She cost £10,500, and in May 1839 steamed from G-ravesend to Portsmouth in twenty hours. The following tables give particulars of her competitors and their performances on the Channel station between Dover and Calais in 1840. Quickest passage from Dover to Calais was made by the " Archi- medes/' and took 1 hour 53£ minutes to perform. Table XIX.- : Archimedes " and her Competitors in the Channel Trials. Name of Ship. Pro- peller. Ton- nage B.M. Dimensions. Area of Mid. Sec. Engines N.H.P. N.H.P. Mid. Sec. Speed on Trial. Cylinders. Ft. Ft. Sq. Ft. 11 Ariel " . Paddle 152 108 long ; 17-3 B.M. 95-0 60 0-632 10-4 " Beaver " ,, 128 102-2 „ 1G-0 „ 84-0 62 0-738 11-2 " Swallow " ,, 133 107 -G „ 14-8 „ 84'0 70 0*833 10-4 "Widgeon" " 1C2 108-0 ,, 17-1 ,, 95'0 90 0-947 10-3 Two 39-inch diameter x 37 inches. " Archimedes" . Screw 237 106-0 ,, 21-8 ,, 143-0 SO 0-559 9-6 Two 37-inch diameter x 36 inches. TRIALS OF S.S. "ARCHIMEDES AND H.M.S. " RATTLER. 203 Table XX. — Eesults of Trials of "Archimedes" at Dover, 1840. Name of Com- petitor. Course. Weather. Conditions. Speed of " Archi- medes." Knots Winner by Time. 1. "Ariel" . Dover to Calais Fair Both using sails 9*75 " Archimedes " by 6 min. 2- „ . . Calais to Dover 9-75 „ 5 „ 3. "Beaver" Dover to Ostend 9-50 4 4- Ostend to Dover 9-25 "Beaver" ,, 9 ,, 5. "Swallow" To Calais and back Steam only "Archimedes" „ 1 ,, 6. "Widgeon" 19-mile course Fine; fair wind „ 8-50 "Widgeon" ,, 6 ,, 7. ,, Head wind . p 8'00 „ 10 „ 8. No wind Quite smooth ,■ 3A „ 9. i-i Fresh breeze Both using sails "Archimedes" ,, 9 ,, Admiralty Trials with H.M.S. "Rattler." — Experiments with the s.s. " Archimedes " having sufficiently satisfied the Admiralty that screws had such advantages for naval purposes as to warrant further trials, it was decided to build two sister ships and fit them with similar engines made by the same engineers, the one set geared 4 to 1 to the screw shaft of one ship, and the other coupled direct to a pair of paddle wheels in the other. The " Battler," a screw steamer, was 176*5 feet long between perpendiculars and 327 feet extreme beam and 15'5 mean draught of water. Her displacement was 1140 tons and the immersed midship section 348 square feet. The " Alecto/' the paddle steamer, was built to the same mould but differed somewhat at the stern from the " Rattier," as the latter ship had to be designed to suit the screw and its appurtenances (see fig. 7). Each ship had a set of Maudslay's twin cylinder engines having four cylinders 40 inches diameter and 48 inches stroke ; the KELP, being 200 and the I.H.P. on trial 437 in the case of the "Battler/ 1 when tried in January 1846 and running at 9'639 knots. The propeller was a Smith's common screw 10 feet in diameter and 11 feet pitch, with an acting surface of 22*8 square feet, running when at full speed 107*9 revolutions per minute. In the competitive trials with the " Alecto " the ships were ballasted to a mean draught of 12"3 feet; the first run was between the Nore and Yarmouth Eoads, under steam only, with the sea smooth and the weather calm. The time taken by the " Battler " was 8 hours 34 minutes, and as the distance was 78-\ nautical miles, the mean speed was 9'2 knots. The " Alecto" took 8 hours 54 minutes, so that her mean speed was only 8*8 knots, but she attained that with only 281-2 I.H.P. as against the 3346 of the "Battler." 204 MARINE PROPELLERS. 10 CO O H CO CO H H I* % o O C i£l t- O l-H — < X 5S o -5 I o c/2 to s ^ o fc- aa C3 M rH £C o Q O i— i < > Pn O 53 3 3 O O CO o o r- OQ 55 r-» H wo as m OS rH .« S3 "*! fei <1 PQ H A Ph 3 i-H ,_ CO C-l CO CO ™ t- > a (U •1- ■I- •1- a '■~ Ph CM Ph W w w c3 -■ fH to Ph P< H H p TRIALS OF H.M.S. DWARF AND OTHER SHIPS. 21 1 Table XXIV. — Analysis of certain Trials made with Screw Propellers in 1847-8 on H.M.S. "Minx," 203 tons, dimensions 131x221 feetx5'2 feet draught. Description of Screw. No. of Trial. Pitch of screw . ft. Surface of blades sq. ft. Pitch ratio Surface ratio Revolutions per minute Slip per cent. Speed of ship . knots Indicated horse-power lbs. Indicated thrust . ,, Dynamometer thrust ,, Calculated „ ,, Tow-i'ope resistance of ship lbs. Pr. H.P. Tr. H.P I. H.P. delivered to screw N.H.P. Pr. H. P. + screw friction . Pr. H.P.-rN.H.P. (per- pendicular efficiency) Tr. H.P. -^I.H.P. (general efficiency) Displacement % x speed 3 ™ I.H.P. All Common Screws with Two Blades 4"5 feet Diameter. No. 1. No. 2. No. 3. No. 4. No. 5. No. 6. 5'0 5-83 5-83 5-83 5*83 5*0 7'20 4-93 5-97 7-10 7*10 7*20 1-110 1-295 1-295 1*295 1-295 1*110 0-450 0-308 0-373 0*443 0-443 0'450 248-9 237 '8 232-0 218-7 256-9 250*1 32-0 417 397 377 38-0 28*0 8-36 7-97 8-04 7*85 9-13 8 54 188-1 178-3 177'7 168-4 252*1 193*4 5000 4240 4335 4350 5600 5110 4713 4458 4437 4372 4282 3416 2808 2302 2089 2360 3170 2836 2169 1963 2002 1909 2576 2262 72*0 56-4 51-5 56-8 88-8 74-5 557 48*1 49-4 46-0 72-6 59-2 126*2 121-6 1237 119-1 185-9 130-6 82-0 64*0 60-1 65-3 102-6 84-7 0'574 0-462 0-416 0*477 0*473 0-570 0-320 0-292 0-300 0-295 0-310 0*331 128'6 123-4 125*8 126*6 112-3 125-4 Table XXV. — Analysis of the Trials made at Stokes Bay, 1857, of Various Patent Screw Propellers on H.M. Yacht " Fairy/ 3 144*7 x 21*13 x 6*1; 210 tons displacement; W.S. 3300 Square Feet. Kind of Screw. Diam. of screw ft. Pitch „ ,, Length ,, ,, Pitch ratio Revs, per minute . Slip per cent. . Speed of ship knots Indicated horse- power Indicated thrust lbs. Resistance of ship lbs. Tow-rope H.P. . Tr. H.P. -4-1. H.P. . Disp. %xspeed*-^ I.H.P. Common, Scott's. Griffiths. both Corners Boome- rang. Lowes. Fisher. Griffiths. Common. Loosey. 6-5 Cut Off. 6-5 6-5 6-5 5-83 5-57 6-0 65 6-3 8-00 9*70 9'82 7-70 14-12 8-00 8-25 7-90 10-00 1-75 1-54 1-50 2-40 1-66 0-76 1-60 1-83 1-71 1-23 1-49 1-51 1-32 2'54 1-33 1-27 1-22 1-59 177-5 178-7 174-0 226-3 187-5 207-5 212-5 205-0 202-5 19-26 26-88 27-21 31-62 =55-37 20-2 23'26 17-15 38-84 11-309 12-463 12-26 11-699 11-658 13-033 1327 13-229 13-216 359 362 349 335 384 410 410 406 416 | 8343 6872 6737 6345 4820 8180 7730 8210 6785 528U 6237 6171 5643 5610 6996 7260 7220 7211 183-4 239 230 203 201 280 296 293 293 0-510 0-661 0-659 0-605 0-523 0-682 0-721 0-722 0-704 142-1 188-9 186-7 168'8 144-8 1S6-5 196 9 198-1 196-0 21 2 MARINE PROPELLERS. making the pitch, say, 10 feet, they made it only 7'25 feet, with results very disappointing. The most interesting feature in this trial is the amount oi slip, which began with 17*68 per cent, negative and ended with 7'24 per cent, positive ; and that with the new screw 13*59 positive slip was experienced. This would seem to point to magnitude of diameter as the cause of negative slip, or it may be that slip ratio is the measure, if not the cause, of this evil, for in later ships where negative slip occurred the pitch ratio was always small, while the diameter of the screw was often by no means excessive. Table XXVI.— Analysis of Trials made on H.M.S. " Archer' in Stokes Bay, 1849, with Screws oe Different Diameters. No. of Trial No. 1. No. 2. No. 3. No. 4. 9-0 1 1 No. 6. I Diameter of screw ft. 12"50 11-62 10-0 9-0 ; Pitch ,, ,, 775 775 7-75 7-75 7-25 ' Surface of blades . sq. ft. 48'24 41*2 31*8 26-4 28-5 Pitch ratio 0-620 0-666 0*775 0-860 0-806 Surface ratio 0'392 0389 0-405 0*418 | 0*448 i Revolutions per minute 71*01 72'99 98-77 110*2 126-17 j Slip per cent. . 1 17*68 10-19 0'45 7-24 ' 13-59 ! neg. neg. pos. pos. , pos. Speed of ship . knots 6-39 6-151 7-520 7-818 7-800 ' Indicated horse-power . 199 206 294 347 J 420 Indicated thrust lbs. 11,940 12,015 12,686 13,400 , 15,200 Calculated ,, . . ,, 6640 5395 6670 6050 1 7670 Tow-rope resistance of ship , , 3785 3508 5244 5587 , 5540 Tr. H.P. 74-2 66-6 119-6 134*1 133-0 Pr. H. P. delivered by screw 130-5 101-8 153-9 145-8 1 182-4 I Pr. H.P. + friction of screw 165 127*8 186 178-2 198-0 1 Pr. H.P.-rl.H.P. 0-660 0-494 0-524 0-423 , 0-434 , Tr. H.P. -T- 1. H.P. 0-372 0326 0407 0-387 0-317 Disp. % x speed 3 ' -r I. H.P. 151-7 130-1 166-0 159-0 ; 133'9 ! In 1856 H.M.S. "Flying Fish," after undergoing some crucial trials, was fitted with a false bow, which not only gave her a liner entrance but made her 18 feet longer. Her best speed as originally built was 11-832 knots with 1362 I.H.P. and coefficient 127'3. After alteration, with the same machinery the speed was 12*572 knots with 1303 I. H.P. and coefficient 165 - 9. A large number of experiments were made with different screws, including those already alluded to with the Mangin screws. In 1858 H.M.S. £ ' Diadem" and " Doris," sister frigates of 3800 tons, were tried with screws of different diameters, viz. 18 feet and 20 feet. The " Diadem " with 18 feet showed an efficiency of 0"420, while the " Doris," with 20 feet, had an efficiency of 0'365. But the " Doris " TRIALS OF H.M.S. DWARF AND OTHER SHIPS. 213 with 18 feet and the same pitch as the 20-feet screw showed an efficiency of only 0*350. ADMIRALTY EXPERIMENTS ON SCREW BLADE SHAPES. Table XXVII. — Analysis of Experiments made with Various Screws in 1859 on H.M.S. "Doris," 240 feet long, 48 feet beam, 20-5 feet draught of water, displacement 3714 tons. Common Screw with Two Blades. Griffiths Screw, Two Blades. Particulars. Cumplete as First Leading Corner only Cut Following Corner only Cut Both Corners Cut Away. Away. Away. Diameter of screw . . ft. 201 20-1 20-1 20-1 20-1 Pitch „ . „ 32-16 31-95 31-95 31-95 31*95 Acting surface of blades . sq. ft. 106-6 105-3 98-0 98-0 90*8 Pitch ratio 1'607 1-597 1-597 1-597 1-597 Surface ratio . 0*341 0*331 0-308 0-308 0-286 Revolutions per minute . 49-83 48-25 5000 49-86 51-83 Slip per cent. . 24-23 22*15 23-46 24-75 26-39 Speed of ship . . knots 11*981 11-826 12-048 11-816 12-012 Indicated horse-power 2822 2784 2880 2846 2916 ,, thrust . . . lbs. 58,150 59,900 60,000 58,900 " 58,400 Calculated ,, . 33,200 40,800 39,500 39,500 38,800 Tow-rope resistance 28,720 27,980 29,040 27,930 28,850 N.H.P. . ... 2174 2157 2230 2200 2240 T.H.P. ... . 1221 1494 1461 1426 1428 Tr. H.P. 1057 1017 1075 1015 1064, T.H.P-fN.H.P 0-562 0-692 0-655 0-648 0*637 Tr. H.P. -r I. H.P. . 0-375 0-365 0-373 0-356 0-365 Displacement % x speed 3 -f I. H. P. 145-6 142-2 145-0 138-8 142-3 Augmented resistance, I. H.P. 164 477 386 411 364 H.P. ---T.H.P. . 0134 0-319 0-264 0-288 0-268 In 1861 H.M.S. ''Duncan/' 3985 tons, was tried with several different screws, and among them two were three-bladed with a moderate amount of surface; the one with 115 square feet of surface gave an efficiency of 0*526 against 0-502 of the two-bladed screw with 77'6 square feet of surface; that is, the blades were of the same size and shape but slightly different in pitch. In 1862 H.M.S. " Shannon," 3612 tons, was fitted and tried with various screws differing in number of blades, surface and pitch. The highest efficiency, 0'445, was obtained with a four-bladed Woodcroft screw as against 0"409, that of a common two-bladed screw. The three- and the six-bladed screws did fairly well, their efficiency being higher than that of the two-bladed. The blades were all of one size and shape. 214 MARINE PROPELLERS. In 1863 H.M.S. "Emerald," 3563 tons, was tried with several screws, including a common two-bladed, a four-bladed, and a six-bladed, but in this case the six-bladed had not such a huge amount of acting surface, having only 103 square feet against 82'8 on the two-bladed. The efficiency of the six-bladed was 0*466, the highest, as against 0*422 of the two-bladed ; the four-bladed was again good, being 0*457, and the highest speed, 12*003 knots, was obtained with it, as also with the four-bladed screw of the " Shannon." Table XXVIII.— Trials of H.M.S. "Shannon," 3612 tons, in 1862, with Screws varying chiefly in the number of Blades. (i) (2) (3) (4) (5) Common Man gin's Wood croft's, Woodcroft's, Woodcroft's, Screw. Patent Pitch Pitch Pitch Screws. Increasing. Increasing. Increasing. Diameter of screw . ft. 18'10 18*00 18*15 18-15 18-15 Pitch ,, . ,, 2375 24-2-26-0 2*75-24-4 22-7-24-7 22-8-24*4 Number of blades . Two Four Three Four Six Surface blades, T. areasq. ft. 71'2 113-6 86-4 115-2 172-8 Pitch ratio 1*312 1-444 1-345 1*367 1-345 Surface ratio . 0-272 0*440 0'334 0*446 0-670 Revolutions per minute 58*58 53-17 56-08 53-17 49-67 Slip per cent. . 17*79 17-42 14-89 10-83 6-34 Speed of ship . knots 11-288 11-330 11-492 11-550 11*208 Indicated horse-power 2057 2033 2055 2023 1956 ; Displacement % x speed 3 164-6 168-4 174-6 179-3 169-5 4- I.H.P. Indicated thrust lbs. 48,750 48,700 50,000 50,800 53,200 Calculated ,, ,, 34,420 31,611 35,930 37,070 39,530 Resistance of ship . ,, 24,225 24,396 24,080 25,346 23,845 I.H.P. delivered to the 1559 1581 1579 1571 1534 screw(N.H.P.) I.H.P. delivered by the 1193 1100 1267 1315 1361 screw (T. H.P.) Resistance H. P., Tr. H.P. 840 848 851 900 821 Efficiency of screw T. H. P. 0765 0-696 0-880 0-837 0-887 -hN.H.P. General efficiency Tr. H.P. 0-409 0-417 0-414 0-445 0-420 -r I.H.P. Tr. H.P.-rT.H.P. . 0-704 0-771 0-676 0-684 0-603 In 1865 H.M.S. (f Pallas," 3660 tons, was found not to be working efficiently. Two of the four propeller blades were removed, thereby reducing the surface to one-half, and the ship tried again, when it was found the speed was 12-824 with 3630 I.H.P. as against 12*56 knots with 3568 I.H.P. on a lighter draught of water; but with a Griffiths screw having two blades a foot less in diameter, the speed was over 13 knots with 3580 I.H.P. See Table XXX. i.vAL Architects," 1879). [To face p. 214. Fourth Series. i 5 1 i 2 1 3 i 5 '8. August 1st, 1878. 15' 8" 20' 6" 700 3290 40-963 21-6 93*25 76-93 59'385 39-15 19-26 606 184 7556 3958 1765 596 ! 163 7-797 18*587, 15-746' 12"475 1 8-321 i 547-5 1 594*9 i 690-5 | 770-0 6767 173-0 188-0 ' 218-2 243-3 213-8 l i i 4 blades. Modified Griffiths, 2 blades. 18' IV 21' Si" 3*36 5*04 2-49 07 1-25 ... 112-0 98-0 ■19 TRIALS OF H.M.S. DWARF AND OTHER SHIPS. 215 Table XXX.— Trials of H.M.S. "Pallas," 225 feet long x 50 feet beam, in Stokes Bay in 1865. (i.) Original Four-Bladed Screw ; (ii.) Original Screw with Two Blades Removed ; (iii.) New Griffiths Screw, Two Blades; (iv.) Griffiths Screw Half Boiler Power. No. of Trial No. 1. No. 2. No. 3. No. 4. Displacement tons 2758 3357 3661 3661 Diameter of screw ft. 19-17 19-17 18*0 18'0 Pitch ft. 19 "2 -21 -2a 20'6 -22 -6a 19-4 19-4 No. of blades 4 2 2 2 Pitch ratio 1*00 -i-io 1-08-1-18 1-08 1-08 Revolutions per minute 64-25 70-37 79-82 63-02 Slip per cent. 4-13-572N 10-8-18-69P 14*53P 8*16p Speed of ship knots 12-654 12-824 13-058 11-078 Indicated horse-power 3568 3630 3581 1906 Indicated thrust lbs. 86,600 75,100 76,300 51,500 Tow-rope resistance of ship lbs. 29,700 32,636 34,932 25,092 Tr. horse power . 1152 1284 1398 853 Tr. H.P.-rl.H.P. 0-323 0-354 0-391 0-448 Displacement % x speed 3 -j- 1 H.P. ' 111-7 130-3 147-7 169-4 The most important experiments made by the Admiralty, outside those made in the tank at Haslar and not published, were those with H.M.S. "Iris," in 1878 (see Tables XXIX., XXXI.). This ship and her sister the "Mercury" were to have made a great advance in speed on any previous ship for naval purposes ; moreover, they were twin-screws. Apparently without any previous experiments, model or otherwise, Sir James Wright, then Engineer-in-Chief, specified the screw to be 18"5 feet diameter and 18'2 feet mean pitch with four blades, as shown in fig. 62, and a surface of 97'2 square feet each. These ships were 300 feet long, 46 feet beam and 181 feet mean draught of water, the displacement 3290 tons; the immersed mid- section was 700 square feet; as the prismatic coefficient was only 0"548, the lines were very fine indeed. It was intended that the I.H.F. should be about 7000, and with this it was not unreasonable to expect 17*5 knots. As a matter of fact the full speed first realised was only 16-58 with 7503 I.H.P., and with 5251 I. H.P. the speed was no more than 15'12 knots. Fortunately Froude had already introduced the pro- gressive trial system, and a few extra runs on the measured mile at reduced speeds permitted the plotting of a curve of I.H.P., a curve of slip, and a curve of indicated thrust. The disclosure made thereby 2l6 MARINE PROPELLERS. caused the removal of two opposite blades from each screw, and on making a fresh set of trials it was found that 15*726 knots could be made now with 4368 I.H.P, and the fearful loss through superfluous surface and diameter with original screws shown most clearly. Now, instead of re-setting the two pairs of blades so that the mean pitch was 20 feet, and having another set of trials, two new four- bladed screws were ordered. The new screws were only 163 feet diameter and 20 feet pitch, while the surface was reduced from 97*2 square feet to 72 square feet, with four blades oval or leaf-shaped. The new trials showed that the ship could do 18*573 knots with 7714 I.H.P., or the original speed 16*56 knots with only 5108 I.H.P. Someone, however, still apparently Original Four-bladed (Nos. 1 and 2). Fig. 62. Final Two-bladed (No. 4). -Screws tried on H.M.S. Substituted Four-bladed (No. 3). 'Iris." hankering after the big diameter, now caused another pair of new screws to be made on Griffiths' plan. These Griffiths screws were with two blades 18*14 feet diameter and 21 -3 feet pitch, and an acting surface of only 56 square feet. They were very successful, for the speed now obtained was a trifle higher than with the last screw although the I.H.P. was less. Now the pity is that the Admiralty did not crop Sir James Wright's screws to 16 feet and try them, or, better still, crop them and increase the pitch to 20 feet; and further, when they were making the Griffiths screws, have pat sufficient surface in them so that they would bear cropping to 16 feet. In order to compare the merits and faults of the four different sets of propellers, the power, revolutions, etc., have been taken from TRIALS OF H.M.S. DWARF AND OTHER SHIPS. 217 H K H rC 3 O K <5J O CD «-J CJ> " „ - *1 Cft FM a CO en 1— 1 4-J CD "4-1 CQ CD .. £ cS O H co w©oo It. XKMCO t* w if) ?o -d - £> - - < •i-i ft rS -,3 . r-H ■a -•.-i - - H Fl Ph - &H +3 03 ., „ . ^Ph r 1 ^ MH (M Oi CO cri CO X) OS 1— 1 r-i — 1 t-I CM * X cc3 O Eh W CO 1— 1 . cd -3 ~ " T, EH co 4-3 TfH <£. 10 m CO ,— 1 H £ 00 00 CO m CO Tf) EG CD ~ >* CJ H3 hJ < rfi * '£ ° 2 ° < !* fs <4H +2 «*H += o3 1 1 rQ M £ = : M 'i X (73 H(NMt(I w »-j ©«...„ PQ te " - - r- rH c CO Tf -« O CW 00 t- CM CO m in on Cl CO CM m m m co OS m 3S 1 OS p VI 0) CO ~t< •* o_ Tf m r- 1- rH Ol CO' 1- p i-H d CO 4i CO CO tr- cm" 1.0 co" -f ■** CO — co rH 00 CM Tf Ol ■-. p CJ O b b CM 33 CO CO Tf OS rH •4-3 O a M p lO z ~- CO CM CM CO OS 1 CO' OS t* ■0 c~ j.- -* m 1- Cl co as ''.' 1 CO X co d rH CO r-i Csl CO CO CN_ TJH ;o Tf to CM cc i-H -1 p r~ p O +3 d 00 0" co 0" co" rH Tf 01 co Tf CO 01 CO Cl 01 01 ■=:■ b CM K O in m co Tf rH Cj >x> m O O ■■"' CO OS CO m CM m 00 O US •^ rH_ -f i— 01 co O CO OS 1 -f p ^T Tf d rH CO CI lfl CO rH co" co" Tf 1^ Tf co OS CO ::■ CO 1 CM b b i-H Z O CTi in Os rH CO CO «s 00 CO rH 00 O CO >rs CO N c t i-H Cl p .-H I— co O rH °°_ Tf co_ O m r~\ Tf J^- ~. 0. d Szi 6 l-H co co" ^0 H CO 1-T CM t-T 00 CN 01 rH i-H b b CM eo K CM O as t- as co OS 5 rH m rH rH rH o> CO m *~- in 1^- CO OS t- CO O fl W rH CO i> -<»" 1— ' CM m Tf - m c Tf p x> 00 Cl d ft O CO rH rH in co co Tf" Cl i-4 CO r-t ■■•-. r-\ b b CM co. O O O as CD CM CO (M -* O -jj ifl co Tf rH m r-i CC' t-~ CO O -f O 7* Ol Cl os l-H^ 1- O m =■ rH Tf to >j p rH «4 6 to CO r^ rH a? cm" CO co r-T co r-H Cl CO T— I CM rH b b CM 55 CO l- co ■OS co eft O -* co l-H m O Tf O co 1-- 10 cc 01 rH p C~ CO -*_ OS ci CM CM -: 1 ■■/:■ f Tf t— t- m d ib CO Cl CN as" CO co" Tf rH CO r-< '-.-. 90 00 OS b w 00 0" co" i^" cm" l-H CO m m Tf CO b b ■* co CM rH M p O O CO Tf Tf X 01 00 O \: -rH - Tf CO CM CO -f 01 CO O rH iO O CO CM m t^ O :: ; -i :.:■ CO CO CO CO p at 3 d lb CSI CO CO c;i as" rH i-T rH" : 1 m rH Tf m Tf O b Cl » 00 O O r-- 1— 1 I~ CM fr- I> m m O CO os DO Os p Vj 10 co in Tf ■;:-. y. 1- l-H OS 00 OS i- t~ Tf d Tf rH - 00" -*" c^" l> rH co i-H t— CO f b O l-H :o (M rH in a ^1 cd = IS r-j CO a> ft a Eh ft ft -d cu CD ft 32 ft H a £n •1-1 O P CD ai CO fl fcj' H CD c; r-H -1. O j. :.■ O |S H=! 1 1 c fl 4J r-. CU 1) m ca [-1 -_ CU ■d CD CD CU s a CD ft co fl .3 fl CD > ft -C +2 O! - CU CJ CJ (O 'to CJ CU = ■■■:: 'co 0) t. m O s 'to a fl +CJ to to -t-i ■1- ;-, Oj -> CD •_" CD fl" O '*J O 'C 1-1 fl O fl CD > cd 3 "cci CJ a m CD t- CD ft O Fh i ft W •i- ft pq "A ft w H ■1- ft in X c£ ■W fl cp 3 (H 0) cc! -u rt ft G a) -d "it. •m -u ft i ft CO ft S Fh a P4 ft ft ft ft ft M ft a et} > ft 5 S 1— 1 '•B - O OS w" W a ! w w M PH w ft m P5 C« O H Z to w »— i H PC w C4 PL. w < (n 03 Pi J P J ro w e/j Q CO <4 W <3 CO D i— i IT] H O <1 M S H Uh —i O Ph CO „ Ed CO hH in ecj rn w H C/J fc i— i i— i o fc 05 el oj m CD cj tf£ H PQ H ■I- ci iQ C6 CO CI © O >o CO O o 1—1 CI CI CO id o i-H r- r-l l-H 1— 1 fell PH H ft S OS vO, o in -# & s.= CM (M 0.°H CuPM ■— '■ & « i— i OS CO cd os Cl co OS CO co ITS CO Cl Cl I— 1 I— 1 1— ( CD i— i CO CO o CO CO OS Cl CO CO CO '■:. CO Cl TO Cl 02 h " 1 © © © o © © © O 0) o a Cl Cl Cl Cl — 1 TO f— i © p Cl to Cl Cl ,d >c. o TO © to o TO Pm a Ttf CO J>. os © i— i ■—1 CC i—i 5 llj © © o o © © © S ■- CD CO © to CO © CO Number of Experiments. CxO DoO DcO C»0Co0 CoODoO ANALYSIS OF MR CHARLES SELLS EXPERIMENTS. 221 < W H O < Ik Pd P m H P P5 H D O ffi O P o >— < >H W tf 3 -aj 03 t> o CO Ed H W s © M m H f* o i~~ -rP CC' 10 r-H ta oo CO to OS CO OS o v •1- w CI CO T* r-H CI CI "& .fc'S -4J a ei Q 1— 1 r-H r-H r-H r-H r-H 1-t 4)'o d \a ^H lO to 1ft O OS PhcS -(■■ £*• T*l -* OS 1-^ O CO W o o Oi CO "* 7"l CJ CO CO CM f-H r-H T-t rH r-H r-H r-H -* Ol OS os O to to oj a ^ fe •^ C^ o O OS t^ S BO & S^ (N CM CN Cl CI r-H l-H ht3 opH S3 a C o Ph - £ ° ** o -rts ■*< 1ft O "J*M M iO CN t^. t-- 10 -# O « " Oft lO ift -* -^ "* ^ t* SPo Png 5*» H Eh - ° a Jg r>- (0 QJ C3 to no CO CO r^ ' ° "-£ 5 P i— i. o o 10 !"■ -m CO CO CO d CI CI CM j i> o^S Shm-i a n ^2 CO -X5 CO OS to CO O O 4> ■H ft » 5 c a OJ CO I— rs OS CO P4^ 1— ' r-H r-H CM r-H I— 1 '3*» P OJ *= o ■H V>l -tH CO CTj tSH lift OS fr- rt io M CO CO 1ft 1-- CO CI r-< .£ o « J CI CO ~T to ^S o O 00* .=5 6 o o o O rH

en OS OS OS 2 w p p p O p S a r-H r-1 K r-H B W M r-H > & r> p\ t — ) o r \ // / > \ / \\ f~y a V 1H 0> c ^o c=o=ac ^f 0oC )CXC c^o= nc^r^ri V- \\ \^^^^ ^^ 1\ ^rr ^-J >— — V-X^ — -J \ \ \ \ / \ a> \^> \ > / \ tf vy /J cy \> 22 2 MARINE PROPELLERS. By comparing the work done when running at a uniform speed of 250 feet per minute by 1 lb. of thrust with the power used in producing it ; that is, assuming 250 foot lbs. as the amount and dividing it by figures in column C, the comparative efficiency of the different screws can be made. It will be seen on comparing columns G- and H that the results are practically the same, and that the most efficient screw is No. VII. and the least is No. I. Strange to say, Mr Sells himself put the values of the screws as propellers in the reverse order, as he took into account only the amount of thrust in each case. The second series of experiments were made with screws of the same diameter and pitch, but with blades differing only in number ; he also made two groups by having Nos. VIII. , IX. and X. propellers all of one surface ratio, and Nos. XI., XII., XIII., and XIV. all of one surface ratio, double that of the former. Judged by the same criterions as before, No. X., having three blades and a surface ratio of only 0*143, is the most efficient screw, and No. XIV., with six blades and a surface ratio of 0*286, follows close on it. By Sell's own criterion No. XIV. was the most efficient, No. X. coming next; he also found No. VIII. to be the least efficient but nearly as good as No. VII., whereas by the criterions here used No. VII. stands as high as No. X., and No. VIII. is better than No. XL, which is a two-bladed screw of 0*286 surface ratio. It will be seen that at uniform speed the thrust is practically the same for the screws of the same surface ratio (column A) whatever be the number of blades ; that it was only the screw with the single blade that differed appreciably from the others of its groups, and the six-bladed which differed from the members of its group. Comparing the performance of screws which differ only in surface ratio, it will be observed on looking at column A that the thrust of No. IX. is 1-52 lbs. and that of No. XL is 2*27 lbs. ; that is, they are in ratio of 1 to 1*49. Further, the thrust of No. X. is 1'51 lbs., and that of No. XII. is 2*22 lbs. ; they are therefore in ratio as 1 to 1-47 ; that is, the thrusts vary very nearly as the square root of the surfaces. The blades of No. XIV. must have been very narrow, seeing there were six with a low surface ratio. They were, however, exactly the same in size as those of the screw No. X. Notwithstanding this, o o Tfl OS -* -** CO CM CO -* CO CO CO CN CM CM r-( o »£S -# CO CO r-( rH i-H rH i-H rH rH l-H rH I— 1 i-H O rH Tt< O ■* CO CO O CO cd r- CO CO CM CO CM rH o lp T* 1 CO CO to co cm o CI -tf o \a o 4-. 1-- OS o i— 1 iQ I-H CO irs CM uo co CO i-H CM CM CM CM CM CM CM CM CM CM lO CO O iO O O O O O O O ■^ CO CO CO CO ■** t— l-H !>. kO trs ■** ■* >* lO -^ ITS »0 lO iO CO CO CO oo CO CM CO f- CO l-H CO to o \a co CO CM CO co o ■"* 10 CO OS CM CM CM CO CM CN CM CO CO CO CO CO i-H CM CO CM rH CM CO *0> T* CO OS OS O OS O CM •^ CO I~- CO CO i-H rH CM rH CM CN CM ^H rH rH i-H OS CO cm m CM o O CM CM r^ o _ t-^ -^ CO "* CM <* -* kO CM <-( i-H rH CM CO ■**» I>- -* OS OS CM O CO CO CM OS O CO O Oj rH CM rH OS OS cm 7-t CM rH CM i-H cM rH r-\ o o Tfi HO uo ITS lO \Ci iO iO i-H i-H i-H r— 1 I-H rH rH rH rH l-H rH o o O O o O o O O o o os a OS OS os OS OS. OS OS OS OS o o o o O o o O O o o CO CO CO CO CO CO CO CO CO CO CO **- CM -#i CO J— 1 i-H OS CO rH C— C- co \a *3 1— i I-H > X X X r— I h- 1 1— 1 r> X I-H X X 1 i ' H HH l-H k> 1—1 XX XI X! X rtf f\ T3 o f ) jO dpp i 0 W <*3 CO CD \J EQ TJ T3 •a ■r-l t3 0> CD £3 CO CO r&- ) " r^oo T3 i — i S3 O CO en o H rC CO CO c=o> 14-5 18-125 2 31*5 0-19 1*250 Half of above. C Modified Griffiths' (mere. ) 14'5 15'0 4 63-0 0-38 1*075 D ;t ) ) J J 14-5 16'0 4 63-0 0-38 1-103 E 14"5 17*0 4 63*0 0-38 1172 F )' a >» 14-5 18*125 4 63*0 0-38 1*250 Motive weight 2-589. G 14-5 19-625 4 63'0 0-38 1*353 H 14'5 21*00 4 63*0 0*38 1-449 J 14-5 23*20 4 63-0 0*38 1*600 K 14*5 29-0 4 63'0 0-38 2*00 L Proportional to fig. 4. 9-625 12-06 4 277 0-38 1*254 Motive weight 1-695. M ») ») 2175 27*19 4 142-0 0-38 1-250 Motive weight 3-810. N 2175 1813 4 142-0 0-38 0*833 Parallelogram 14-5 1813 4 92*0 0-556 1*250 Q Griffiths' (Naval) . 14-5 1813 4 63-0 0-380 1-250 p Fantail, broad tip . 14'5 18-13 4 63-0 0-38 1-250 A very interesting paper communicated by Mr Blechynden to the North-East Coast Institution of Engineers and Shipbuilders in 1886 was practically confined to the enunciation of seven propositions and to the establishment of them by means of the results of the experiments made by Mr Isherwood as already given, and those made by himself for Messrs E. & W. Hawthorn with propellers as per schedule above. Mr Blechynden states that these screws were tried at twenty-three different speeds, the result of which was to indicate that — 1. Thrust = C x x W; 2. Thrust = C 2 x R 2 ; 3. Revs. =C 3 x V w ; R being the revolutions, W the motive weight causing the revolutions, and C v C 2 , and C 3 are coefficients which are constant for each screw, and any one of which is deduced from the others. For purposes of comparison they were all reduced to a common thrust of 5 *6 lbs., which includes in the indicated thrust an allowance 232 MARINE PROPELLERS. for the effect of fluid friction on the blades. The following are the propositions as enunciated by Mr Blechynden as deduced from the above : — 1. The turning moment in any screw is independent of the quantity of surface or the mode in which it is distributed, friction and edge resistance being excluded. 2. Screws of equal diameter tried under similar conditions have turning moments directly proportional to their pitch ratio for equal thrust. 3. Screws with equal pitch ratio have turning moments propor- tional to their diameter when indicating equal thrust. 4. Screws tried under similar circumstances have turning moments proportional to their pitches. 5. In any screw, TxP=2ttxM = thrust x pitch = 6'28 X turning moment, which simply means that TxPxK-=- 33,000 = net H.P. 6. The thrust of any screw working with velocity of advance V and slip S can be approximately determined by the following: — Thrust = CxAxVxSxy-f#, A being the disc area of the propeller and y the density of the fluid ; and from experiments made by Mr Blechynden he deduces the value of C as 0'946. The corollary of this is that with constant slip ratio the thrust varies as E 2 or as the square of advance. 7. The effect of surface is the same irrespective of the number of blades into which it is divided, so long as it is similar in distribution. KENNIE'S EXPERIMENTS EFFECTS ON THRUST OF IMMERSION OF SCREW Mr George Rennie in 1856 made an interesting experiment with the object of ascertaining what was the exact benefit to be derived from submerging a screw well below the surface of the water. The records, as reported by him to the Institution of Naval Architects in 1878, while demonstrating that the gain obtained by so doing was great, are not easy to follow as set forth by him in detail. On carefully analysing the figures, however, and dealing with them graphically, the following have been deduced and may be taken as a fairly correct statement of what really took place The Screw employed was a two-bladed ''common '' one such as then used in the Navy, 1 foot 9 inches diameter, of true pitch, made of bronze, and highly polished. A tank or caisson was placed and EXPERIMENTS MADE BY ISHERWOOD AND OTHERS. 233 secured on the bank of the Thames close to Rennie's works; the screw shaft projected horizontally through the side and carried the screw at its end about 3*5 feet from it; it was supported close to the screw by a V bracket. Inside the tank the shaft ran in a bearing and was fitted with a belt pulley, by which it was turned at a constant velocity from connections with the factory engine. The thrust was taken on one end of a bell-crank lever ; its other end was connected to a balance scale above it, in which weights were placed so as to measure the thrust for the time being. Continuous observations were made from the time the screw was just immersed to that when there was 5 feet of water over the tips. How far the tidal currents affected the screw is not stated, nor can one now say with certainty whether the proximity of the tank was the cause of the erratic records, but is more than probable. The following may be taken as the real variations in thrust : — Table XL VII. Distance of Axis below Surface. Water over Tips. Thrust. Distance of Axis below Surface. Water over Tips. Thrust. Feet. 0-875 1-375 1*6*25 1-875 2'375 2*875 Feet. o-oo 0-50 0-75 1-00 1*50 2*00 Lbs. 46 95 125 160 280 340 Feet. 3-375 3-875 4-375 4-875 5*375 5-875 Feet. 2*5 3-0 3-5 4*0 4-5 5'0 Lbs. 360 372 383 392 400 405 It is remarkable that the thrust should have been more than doubled by an immersion of 6 inches, and shows how very necessary it is under any circumstances to have the blade tips always quite under water. It is perhaps still more remarkable that with a foot of water — that is, the addition of another 6 inches over the tips — there was an increase of thrust of 68 per cent. Most engineers would have thought that the limit had then been reached, and almost useless to go further ; but Eennie went on, and found that with 2 feet of water over the tips, equivalent to a little more than one diameter, the thrust had risen to 340 lbs., or more than double that with the 1-foot immersion. After this the thrust continued to increase, but only slowly, so that at 5 feet it was 405 lbs., a pressure it had very nearly reached at 4 feet, for it was then 392 lbs. Now the acting surface of the screw is not named, but it was 234 MARINE PROPELLERS. probably rather less than one square foot ; the pitch also is not known nor are the revolutions stated, but it is obvious that to get a thrust of 400 lbs. with so small a screw they must have been high, perhaps as many as 500 per minute. Now under the circumstances and conditions above named, it is more than likely that this is the first recorded case of cavitation, for no screw properly fed with water could produce such differences in thrust as this one had, nor could a screw which exerts 400 lbs. of thrust fail to produce 50 lbs. fully immersed at the same rate of revolution if fed with water to even quite a moderate amount. The sudden rise in thrust after the immersion reached 1 foot points to cavitation, due to the high rate of revolution and the obstruction of the caisson. ADMIRALTY EXPERIMENTS TWIN VERSUS SINGLE SCREWS A review of the performance of the early naval twin screwships will not disclose any convincing superiority as regards speed and efficiency. Table XLVIIL gives the trial results of five sister single- screw ships of the "Nassau" class and five sister twin-screw ships of the " Swallow " class. The older ships were 15 feet longer and 9 inches less beam and of slightly more displacement ; in fact, 849 tons as against 716 tons. The twins did a higher speed consequent upon the higher I.H.P., but the Admiralty speed coefficient was only 122 against 143*3 of the " Nassau, 5 ' and general efficiency was 0*345 against 0-410. It was, however, at half boiler power and cruising speeds that these twins showed to advantage, for with higher speed than the "Nassau," the "Swallow" showed an efficiency of 0"434 and speed coefficient of 154*7 against 0*425 and 147*5 of the "Nassau." The " Penelope," twin screw of 1867, was of 4368 tons displace- ment, and did only 1276 knots with 4703 I.H.P. with a speed coefficient so low as 118, whereas the "Galatea," single-screw frigate of 1860, of 4270 tons, did 13"004 knots with 3516 I.H.P., giving a coefficient of 165. The former ship, however, was handicapped with a double stern. The "Vanguard" performance was so very much superior to the rest of her class, that it is necessary to also take of "the same class the "Iron Duke," of 5563 tons. It is seen that her speed was 13*855 with 4789 I.H.P,, while the single-screw ship "Triumph," of 6840 tons displacement, required 5114 I.H.P. to do 13*522 knots. But the "Vanguard" herself did 14'94 knots with 5366 I.H.P. and speed coefficient 195 (vide Table XLIX.). EXPERIMENTS MADE BY ISHERWOOD AND OTHERS. 235 & w « co h h-5 O £ 1— 1 t/J so c c? ftj ^ ■— & £ H r/5 W W rn GU H p te W W O a ►h 1 — 1 J & <; N £ Ph CO Ch- EZ! >H «l H CO i-h h-l <3 &3 i— i CO £ t> ft in <4 H <1 Ph 02 ft O ►h M Ph H •a "a to 03 O CO *# CO CO co CO lO co O O tH O To Ph" CO CO to ■:::. m l-H 00 CO - — ITS CM O O ■* EG 1— 1 rH rH hH rH ■* -# rH T) Ph co" cc CO rH s O CO CO "* ■-i O 10 CO rH O t- O tfi -^ Ol Ifl i-t CO CO m Cft Mi rH O CO O Ph I-H rH rH ^ Ol CO IH 03 > O O CO c- CO o-l CO O CO O TH CO a H O l— rH t- co rH rH Ci CO cN rH O O CI O O CO Ph co CO H CO Ci m lA fq" m .;: 0) CO CI -« r~ =s O ■JTi : 1 CO CI — CI <-< pi CO CI rH in rH rH H a H 3 O co CS3 O ■:.-.■ CO I— tH O CO O CO Ifl ■::. eo i-H =y EC O CO H CO CI CD ■■::-.■ CO 01 uO CO CO co rH m ■: S rH p4 CD O CO o 22 CJ TH H W r D a Ph Ph" PH W W w i-H r-i CI) 'I- I" PH -r a X PQ Q £ W 5 H Eh fl H § Ph PH M s ■1- pn -r. W i-T. H P H M CO 236 MARINE PROPELLERS. CO CO o H CO CO CO U2 J 1— i H H < I-H p u o h-H m GQ CO PI O CO 3=° 3 aT m . t* 2 i-h co ?> S ^3 *- t* rH «» 71 r-( ** ,-H "3 ™ urs 1-1 S » O a) CI CO CI CO so ^O of CI O 01 .—1 CO CO to 1— ( 1— 1 K !> — O 00 O 1— 1 CI CO CO 1— 1 rH CM O O CI I— 1 CM CO~ of I-H CO to I-H 1—1 I-H 'O SO CO CM CO CO O J>. 1^. O 00 as OS CI O O id CO i-H ■0 gee in CO CI CO CM ■^ CO 1^ 1— 1 to CO rH i-H H a a, 02 O CN CO CI CO 00 CI CM HO CO •■# 'O 1— 1 CI CI 1— 1 CI 10 ~i CO 00" T— 1 10 to t— CO CO CO Oi CM Oli x. OO r— 1 OC0 OS ^ -" 2 S •-< ^ s M P5 H -4-> Ph q a ri4 CO T3 ai O -t-3 ^ 02 .-_ Ch 4J cn y O ^ ri-l -(J np 'n fH 0) +i -r-J HH h-C EH Pm W r— 1 •1- Cm a Oh Ph w H 1-H t^ "I" tH G C/J ffl *o X (-1 &a •N H w « EXPERIMENTS MADE BY ISHERWOOD AND OTHERS. 237 It is rather to the merchant ship running without top hamper that one has to look for confirmation that the twin screw ship is more efficient, and this may be seen by referring to Table LIX., where various types of ships are compared with one another. Table L. — Analysis of Experiments made by Mr Geo. R Dunell with an Ordinary Screw and one of Dickinson's Six-bladed on the S.S. "Herongate." 186 feet long x 251 feet beamx 19*3 Displacement, 1110 tons, W.S. 7500 square feet. Particulars of Screws, etc. Ordinary C.I. Mercantile. Dickinson's C.S. Screws. Diameter of screw . ft. 1075 10*5 Pitch * ; * 15*0 15-0 Acting surface sq. ft. 32-0 30-0 Number of blades 4 6 Pitch ratio . 1*395 1-430 Surface ratio 0*355 0-346 Revolutions per minute 75'3 73-3 Slip per cent. 20-8 13-0 Speed of ship knots 8-82 9*43 Indicated horse-power 328-6 389-3 Indicated thrust . lbs. 9600 11,680 Calculated , , . 5 ) 6938 6734 Resistance of ship }> 5850 6675 Tr. H.P. . 158*5 193-3 D 2 / 3 xS 3 -rI.H.P. 223 232 Tr.H.P.-rI.H.P. 0-482 0*497 T.H.P. -rN.H.P. 0-619 0-534 State of bottom Fresh paint. Foul. T.H.P 188-0 194-8 Analysis of some experiments made by Mr Yarrow in 1883 with a 60-ton torpedo boat : — 1. Propelled by her own engines, which were carefully suspended so that the exact thrust could be accurately taken and measured. 2. Towed by another boat at similar speeds after removal of her screw, when the tension on the tow-rope was carefully taken and registered. The indicated horse-power below is 10 per cent, higher than that registered by the indicators, and is probably nearer the actual gross power developed. 2 3 8 MARINE PROPELLERS. Table LI. — Yarrow's Experiments. Speed in Knots 9-0 lO'O 110 12-0 13-0 14 15-0 255*2 Indicated horse-power, gross . 38-5 49-5 67*1 99-0 143-0 193*6 Engine friction loss H.P. 9'0 10-1 11-7 135 15*4 17-4 19*5 N.H.P., being power delivered 29'5 39'4 55-4 85*5 127-6 176-2 235*7 to the screws Screw friction loss H.P. 1*2 1*4 3-4 12-5 26"6 40*2 65*7 Thrust H.P. by dynanometer 28 '3 38-0 52-0 73*0 101*0 136'0 170*0 Augmented resistance H. P. . 6*3 8-0 11-0 13*0 18-0 23*0 24-0 Tow-rope horse-power 22-0 30*0 41*0 60*0 83 '0 113 146*0 Speed 3 -r constant (288) . 25-4 34-5 46-2 60-0 76-3 95*3 117-2 Tr. H.P. 4- 1. H.P. (general effi- ciency) T.H.P.-rN.H.P. (screw effi- 0-571 0-606 0-611 0'606 0*586 0-584 0*573 0*960 0*964 0'938 0-854 0792 0*766 0*720 ciencyj N.H.P. H-I.H.P. (engine effi- 0-766 0-796 0-827 0-864 0-892 0-910 0-923 ciency) Augmented R.H.P. -f I.H.P. . 0-213 0-203 0-199 0-152 0*141 0-130 0*102 Screw friction -r N. H. P. 0-041 0*036 0-061 0-146 0-208 0-228 10-278 Two-bladed and Four-bladed Screw Propellers. — A series of trials made in 1887 by Mr J. Brucker Andreae of the Koyal Dutch Navy on two ships belonging to their Indian rleet are interesting and instructive, as will be seen by carefully examining the analysis of their results as now given. The " Ceram " and " Flores " are sister ships by different builders, of the following dimensions : — Length . 152 feet. Displacement . . 566 tons. Beam . . 25*6 „ Immersed mid section 189 sq. ft. Mean draught 10*17 „ Wetted skin . 4875 „ Engines with three cylinders 20 inches, 29 inches, and 46 inches diameter, 27 inches stroke; working pressure 120 lbs.; three screws tried were of Griffiths' form. No. 1 screw had four blades and was 9 feet diameter, 13 feet pitch, 30 square feet of surface. No. 2 screw had two blades and was 9 feet diameter, 13 feet pitch, 15 square feet of surface. No. 3 screw had two blades and was 9 feet diameter, 11 feet pitch, 17 square feet of surface. It was noted on the trials that up to 12 knots vibration with the two-bladed screws was not objectionable; over 12 knots it was bad, especially with No. 2. No. 3 screw was tried on the "Flores," which was trimmed more by the stern than was the " Ceram/' and hence her screw had better immersion. EXPERIMENTS MADE BY ISHERWOOD AND OTHERS. 239 Table LIL— ANDREW'S EXPERIMENTS. Results of Trials at Eull Speed in Each Case. Screw. Revolu- tions. Speed. I.H.P. Slip per Cent. Indicated Calcu- Th ™ Bt - ThJust. Tc.H.P. Tc. H.P. I.H.P. D2/3 x S3 T7h;p7 ' No. 1 „ 3 128-6 132'0 148-3 12-78 12-23 12-94 804 728 811 22-5 27*7 24-2 Tons. 7-08 6-25 6*91 Tons. 4-76 3-54 4'25 418 298 378 0-520 0-409 0*467 166 161 171 At Ten Knots when the E.H.P. was Estimated BY MODEL AT 185. Screw. Revolu- tions. I.H.P. Slip per Cent. Indi- cated Thrust. I.H.P. Estimated Thrust. Tc. H.P. Estimated. Tc. H.P.-h I.H.P. No. 1 No. 2 No. 3 91*3 ioo-o 107-0 307 298 332 16-0 22-5 17-5 3-82 3-39 3-96 221 205 190 Tons. 2-39 2-03 2*25 165 140 154 0-537 0-470 0-464 At Speed : Knots 8-70 9-70 10-60 11'35 12-00 11-80 Estimated Net H.P. . 118 169 230 298 372 348 Actual I.H.P. with No. 1 Screw . 200 300 400 500 600 565 9 180 275 375 480 625 565 ,, ,, of difference . 20 25 25 20 25 I.H.P. No. 2-i- I.H.P. No. 1 0-90 0-92 0-94 0-96 104 1*00 E.H.P.--- 1. H.P. No. 1 screw 0*590 0*563 0*575 0*596 0'620 0-616 E.H.P. -f I. H.P. No. 2 screw 0'655 0-617 0-613 0-621 0-595 0*616 At Revolutions 80 90 100 108 116 123 1*28 800 Actual I.H.P. of No. 1 screw . 200 300 400 500 600 700 >i )) jj ■* )i 140 210 300 375 465 545 625 ,, ,, of difference 60 90 100 125 135 155 175 I.H.P. No. 1 screw -r I. H.P. No. 2 screw 1*430 1*430 1-333 1-333 1-290 1-285 1*280 VsoWlS .... 1-414 1*414 1*414 1-414 1*414 1-414 1-414 240 MARINE PROPELLERS. Experiments made by Mr W. G. Walker in 1891 with the S.S. " Ethel," her Dimensions being 55 feet long, 9 feet beam, 3*25 feet mean Draught, Displacement 18*5 tons, and Wetted Skin 733 square feet. The seven screws experimented with had each an acting surface of 275 square feet. No. 1 screw, made up of two halves, each with two blades so set as to form a two-bladed screw. No. 2 screw, made up of two halves, each with two blades, the blades being parallel as in Mangin style. No. 3 screw, made up of two halves, each with two blades set with the after one leading the forward. No. 4 screw, made up of two halves, each with two blades set at right angles. No. 5 screw, made up of two halves, each with two blades set with the forward one leading by 30°. No. 6 screw, made up of two halves, each with three blades so set as to form a three-bladed screw. No. 7 screw, made up of two halves, each with three blades at 60° so set as to form a six-bladecl screw. The speed on each trial was 8*0 miles or 6'94 knots, and the resistance 500 lbs. Table LIII. — Walker's Experiments. No. 1. No. 2. No. 3. No. 4. No. 5. No. 6. No. 7. Diameter of screws ft. 3-21 3*21 3-21 3-21 3-21 3-21 3 21 Pitch ,, ,, 5-35 5-35 5-35 5-35 5-35 6-00 6-00 No. of blades 2 4 4 4 4 3 6 Pitch ratio 1-667 1-667 1-667 1-667 1-667 1-870 1-870 Revolutions per minute . 209-9 212-0 203-0 208-0 218-0 192-2 187-3 Slip per cent. 37-3 37-9 35-2 36-7 39'6 33-7 37-2 Indicated H.P. 3012 30-45 29-06 29-79 31-34 30-26 29'45 ,, thrust . lbs. 885 888 844 859 862 866 865 Calculated ,, . ,, 678 692 605 665 732 629 598 N.H.P. delivered to screws 26-91 27-20 25-95 26-61 28-00 27*32 26-58 Tc. H.P. delivered by screws 14'43 14-72 12-87 14-15 15-58 13*40 12-72 ; Dispm.%x Speed 3 -=-I.H.P. 77-87 77-03 80-71 78-73 74-80 77-52 79'51 ! Te. H.P.VN.H.P. 0-537 0*541 0-492 0-532 0-559 0-491 0-403 H.M.S. " Prince Consort."— In 1863 a series of trials was made with this ship in order to test the efficiency of the different groups of boilers and finding at what speed each set could supply steam to drive the engines. About this time some other ships had a trial at a lower speed than that with half the boilers, but none had so many or any at so low a power as this ship; they thus form a group equivalent to the progressive trials of a modern ship. Table LIV. is EXPERIMENTS MADE BY ISHERWOOD AND OTHERS. 24 1 an analysis of them, which shows how rapidly the efficiency of the screw rose as the speed fell till 10 knots was reached, when it fell back suddenly ; it was never very high, and no doubt accounts for the low general efficiency of this ship at low speeds. The " Prince Consort" was 273 feet long, 58*4 feet beam and 24*2 feet draught of water, 6430 tons displacement, copper sheathed, and propelled by a single screw 21 feet diameter, and pitch varying from 23"84 feet to 25*7 feet, having four blades and an acting surface of 100 square feet with broad tips; the pitch ratio was therefore 1*22. It is very obvious from the following analysis that the engine friction and resistances did not with these large horizontal jet con- densing engines decrease pro rata with the revolutions. If the power required to drive the ship be deducted from the gross I.H.P. it will be found that at 8 knots it is 549, at 7 knots 485, at 6 knots 450 ; deducting from these the screw friction, there remain 463, 426, and 420, which looks as if the resistance was practically con- stant, while the revolutions varied from 34 to 265. At 9-Q knots it is only 510 H.P, 16 242 MARINE PROPELLERS. CO ZD CO H cd o 02 o o O p-l 02 W ft o ffl . ^-. few O O b b b b . V . TR. H.P Tow-rop Resistan of Ship m CO 10 -# in CN CO CD m Cl T* 1— 1 CO OJ t- cz> CO ~* H CO CO ~# Cl i—< i—i .— t 1— 1 1— 1 .{3 Ph ^ i>- O r— CO CO m m j>. O O i^- CI •^ CN O CO CO Oi to -* TO Cl l-H tin* CM 1— 1 T-H i ^ M H a fcDoj TB. H.P. Kemainin to Produc Thrust. CN O t~^ CO l-H CO CO CO 1— 1 O O -* -* ■^ O Cl CN -H ^g O O O CO CO CO O O *# in CO CO CT, CI -.;: CI 00 m CO I— 1 fcW fi CO C"! CI t— 1 I— 1 CO fH .' L 8 tf Ph S3 ^ QJ W O J>- Cl 1—1 CO i>- t^ hi CS £ Ol CT> Cl I— 1 CD CO CD m CO lC CO t^ CO CT> 1>- m -# CO CO CN I— I r-H Q-M was H«3 10 O CO CO Cl I>» CO 00 W CD cx> CO 01 -rf !>. CO w ■* CO CO CI CN Cl I—l f-H SAh O CO O m O m fcM in O Cl 10 in CO r^. CN *-~ I— 1 CO CD CN OJ i>- CO H# 0_; -^ cs CI r-H i—l G .2 « 13 si 00 CN © 00 p p in CO !&l to CO CI t^ ^ O CD I— ( I— 1 EXPERIMENTS MADE BY ISHERWOOD AND OTHERS. 243 H W CM H M CM — 1 a? P P PR P < to P OS >— 1 CO 02 P 1 CO P Ph M >H l-H CO 02 CD P- P P to < O CO OS CO T— 1 l-H t— 1 CO £ h-1 a >-H p PQ H H M <1 H £ii ri CO *^. t^ o> i-H -* CD CO iO ^H -+* rtt -* -* W aj *-» I>. i>- t^. CO CO on CO rr> CO CO rr> CO CO O O O en O CO 03 1^ O O O O O An Ph 1—1 CO CO -* t^ i-H -H t— iC CO en CD W t>- I-H CM CM CO ■* CO CI CI i-H co O CD a ■* m >0 ITS lO >o IO iO 10 m 10 -tH -# CO pi HH O O O O O O Ph Ph" "* iO CO I-H CD 10 IO OS "Hi i-H CO CO O T-H w W U3 CO iO r^ O Ol 01 00 ■* CO l>. 00 CO CO CO CO OJ OS cn> en CO CO CO l>- ^ O O O O Ph Ph 1— 1 CN -* CO CO CM CO on ro m i-H r-H CO a CM i-H CJ5 CO r— 10 ^H i-H of) -^ O CD W o> OJ o> CO CO CO CO CO CO !>. t^ r^ CO CO to HH O O O O i-H O O O O O Ph' Ph Th CO CO CM C7i CO on on CO w Ph l-H Ir^ t^. CO 10 en OJ c~- on 00 CO CO 00 CO 01 Oi Ol CD OS 01 en en cn l>- H O O O * O O . « - l-H jr>- 10 10 t^ r-H l~~ -* O W -t< ■-H Ol l^. CO Tjl CO Ol i-H i-H CM i-H r— 1 i-H HH^ Tc. H.P. Thrust Power of Screws as Calculated. CO O O OS Oi CM CM O CM CO CO CD l-H CO CO i-H i—l i-H CO CO 1^ CD CO CO -^ CO OO CO CM CO in T-H r-H -^ H.P. Avai for ust. «3 O OO IO Ol i-H l-H Oi CM -* ro i-H 01 co O CD CO IO CO CO <-> 10 ■* CD O W CD CO CO CO *tf 1— ' co CD id CO CI r-H I-H CO CM CN i-H H r— 1 Ph P-j Ph.. .DC Boo ■«- CN O t— O i-H CD rH ■^H CO «*1 CO m O 1>« 01 -fi OJ CO CO O aj CO ■* CO CM r-H CM CN i-H I-H l-H l-H x 3fe nS 0) N.H.P. Power ansmitt to the Screws. CO O IO IO ^H IO 10 ITi 10 -* ID ro cn CO CO O O CM i-H CD CO "^ 10 CO CD l^- i-H !>. OS i-H IO O CD CN CO r- CO CM r-H T-H CO CO CM CM l-H l-H EH 3 » — ■ a fl sis fM ^ W O CO CD Oi CM CO CD CO r-H -* CO CM I-H O O 01 CO 0. !>. CD ID -Hi -rP CO to rH O O O 1ft CO r^- t-^ on ■<* -H O 00 CO O h- CO i-H 10 71 CO I~^ CO if5 O CO CM CO *tf r- CM CO •* i-H 0(J CO Tf< CO CM T-H l-H ^hH 1$ CO CM CM i-H r-H l-H -a 3 0) O O O O O O O O O O CO i^ CO 10 ■tti CO CI i-H O (71 on !>. CO m ^M 244 MARINE PROPELLERS. M m Q < o o o sz; t-H t> M O CO o cq o w ►J Ph H H 3M So So rH S 0> OJ ,_, rO 10 -Q <1 I— 1 <5 aj C3 P. r^ o r& S3 "* rH . - pq r^> CN -£ H O H3 ^ +> t» &0 pS •g W) •u o 1 1 PQ H EXPERIMENTS MADE BY ISHERWOOD AND OTHERS. 245 O 1— 1 > H W H w o P3 P H P hJ JZi o 3 3 o o 03 w Ph O ft! o o M •4 r-l o cc o Ph XI 5 m PQ CM m th I— t-H ■—5 CM 01 rH ■HH OS fr— m rH 1— 1 s 00 OS in CM i-H O CM CM CO m i-H CO i-H m tH CO CO rH in CO h4 ip fr— ^H in CO 1* T-H in Cl r-l CO OS GM 4t< r-l m rp CO fr- fr— rH CO CO I— 1 CM rH CO CO r-i CO O m M 00 >n "HH CM ■HH T* rH CI GO m t- m "<*< rH O fr— rH tH CM 00 t— |— 1 rH CO Cl CO ,_, CO CO in f 5 m T* CO CI -HH CO in O CO ■-& T-H "HH ^H rH O r-i m CO fr- CO i-H 1— 1 r-l CO Cl rH in O m eo GO CO M lp fr— 1* OS CM ■"# CO O fr— CM in m 1— 1 O fr— rH CO t— i—< 01 1— 1 i-H r-i "^ C] r-t CO CO CM in O in w m O O T^ CM ■HH "■# fr— CM Cl p Cl t— c!j . OS rH 4* CSI I>- 00 CM CD ^n rO a O ,08 jH. -4-a O ,«a Fh 3 Ph* ^5 CD H O HH 1 ■6 CD r-H eS -(J O Q £ fc CQ £ GQ 1— i w H hJ Q H 246 MARINE PROPELLERS. o H m •— i t— < H o P O Q -J H 02 O O pp o g <5 02 M &a H Cm O M M O O 02 Cm -1 hi N ►J PQ H CO O 60 r- rH O iO SD Cl HH m O H^ CN O C3 £3 in Oa j— 1 Cl 1— 1 O CO CO CO CN OS hO £i 33 EH H CO - CO t>( O -* CO 1— 1 O 00 H* in C) i-t CO I-- XI i-H CN H F— I 1— I T-H -H H O CN S3 rH to CN 1^. CI in m n O CD* p; m m in O CO CO m CO Cl CO Cl 1^ j>- a*M as CO CO CO ~H CN "H CO -0 Cl 01 H .—i 1— 1 r-H CO r-< (J Oi tQ ■-H CD -rf f 1 CD in m *>. CD O O O aifc CD Cl Cl to CO CI CO CO *-- O in m O O CO ^H O OS P O m O HH H rH 1— I CN CO CO 1-1 U CN tq r-H t>- co-> O CO p CO as CO i-H CO CO CO CO in CO p m O CO CD N PI m in 01 C>) Pi Xi CO in to i-H CN 1-1 rH Cl m CO HH i-H S] t- SJ HH CO m PJ « O in ■Hi OS Cl 1—1 T-H r-H p 01 O Cl O m M co HH HH HH CN CN 1— ( CN HH m ■* i-H O 1>- O co !>. CO c P, CO 1— 1 in °3« CO to O CO Ml CO CO rH CO rH m 00 HH co CO in CO p O CO CO OS CN O 1— 1 CN to CN HH CO rH in O Cl r-H m CN CD* m UO P S« O as O r-H HH 1— 1 CO CN in m O Cl CO O CO w rH CN in m 1-1 * CO rH m m P3 ^ in ■* CO CO 1— 1 rH CO CO O CO CN O CO CO P P3 p W 7-H ■—) *>. HI HI f— 1 < CO csl CN O CO 1— 1 CO m m Cl O O > c3 O O CD CD o3 bX3 CD S (3 pq CD F4 O IS H -a CD hH 1 CD a Q a CO O PP a CO ■I 1 8 Bfl" CD 8 CD la O ^ ^ S ° X.8.S. Bellona. Full Power, 410' x 50*7' 7245 26,800 16-5 22-0 3 69 1-334 0*323 89*5 10*90 17-33 126,500 99,422 0*786 80,400 7545 535 7010 464 5287 4278 1009 0'928 0*754 0-567 258 S.S. Larnica. Full Power. 460' x 50' 9471 32,] 90 19-5 21-0 4 118 1-076 0-395 79*8 11-63 14-62 112,100 71,885 0-642 68,560 5694 453 5241 407 3304 3072 232 0-919 0-631 0-539 246 S.S. Don. Full Power. 388' x 43' 4315 19,750 19'5 26-5 4 112 1-36 0'377 72-0 1070 16-82 80,597 61,470 762 55,890 4660 410 4250 376 3175 2871 304 0-912 0-747 0-616 270 SS. St Konans. Full Power. 402' x 43' 5670 23,600 19-0 23-0 4 96 J 54 12 •21 ■339 ■5 ■31 -28 55,300 35,060 0-634 31,050 2101 165 1936 120 1321 1235 86 0*922 0-682 0*588 280 S.S. Cazengo. Full Power. 340' x 41' 4120 18,900 17-0 21*5 i 90 1-26 0-397 71 '0 2-50 13-75 65,010 37,380 0-580 35,721 2884 212 2672 282 1588 1513 75 0-926 0*588 0-525 245 S.S. Lutter- worth . 240' x 32' 2040 11,200 13-5 17-0 4 57 1*26 0*329 91-5 13-8 13-23 33,600 27,000 0-800 19,600 1584 126 1458 131 1096 797 301 0*920 0-752 503 235 [To face page 246. EXPERIMENTS MADE BY ISHERWOOD AND OTHERS. 24? Table LX. -Values of V and V • V. 1 V 1 ' 83 - V 2 ' V. V 1 " 83 - v 2 - V. 41 v 1 ' 83 - V s - V. v 1 ' 83 - v 2 - 1-00 1 21 262-7 441 894 1681 61 1842 3721 2 3*56 4 22 286-1 484 42 934 1764 62 1900 3844 3 7-47 9 23 310-3 529 43 976 1849 63 1958 3969 4 12*66 16 24 335-5 576 44 1017 1936 64 2017 4096 5 19-01 25 25 360*7 625 45 1060 2025 65 2077 4225 6 26-50 36 26 388-6 676 46 1103 2116 66 2137 4356 7 35-30 49 27 416-3 729 47 1147 2209 67 2197 4489 8 4470 64 28 445*0 784 48 1193 2304 68 2258 4624 9 55*70 81 29 474*0 841 49 1238 2401 69 2319 4761 10 67'70 100 30 504-7 900 50 1285 2500 70 2380 4900 11 80-60 121 31 536-0 961 51 1332 2601 71 2442 5041 12 94-20 144 32 568-0 1024 52 1380 2704 72 2504 5184 13 109*0 169 33 601 1089 53 1429 2809 73 2568 5329 14 125*2 196 34 635 1156 54 1480 2916 74 2633 5476 15 142-0 225 35 670 1225 55 1531 3025 75 2700 5625 16 159*7 256 36 704 1296 56 1582 3136 76 2767 5776 17 178-3 289 37 740 1369 57 1634 3249 77 2834 5929 18 198-2 324 38 778 1444 58 1687 3364 78 2901 , ,6084 19 218*8 361 39 816 1521 59 1740 3481 79 2969 6241 20 240*4 400 40 855 1600 60 1795 3600 80 3038 6400 INDEX. Acting surface of screw propeller, Seaton's formulae, 165. Actual thrust of screw propeller, 116. Adjusting pitch of screw blades, 28. Adjustment of pitch, Maudslay, 27. Admiralty bronze or gun-metal, 195. experiments, twin versus single screw, 234. gun-metal or bronze, 195. Air resistance, 51. " Alecto" versus "Rattler," 203, 205. Allen, John, proposals for propulsion, 4. system, hydraulic propulsion, 22. Aluminium bronze, 195. American Navy triple screws, 144. Analysis of ship model, Kirk, 43. Froude, R.E., 118. Froude, Dr W., 121. of the performances of sundry steam- ships on trials, 246. of trials, H.M.S. " Archer," 218. H.M.S. "Dwarf," 210. H.M.S. "Fairy," 211. H.M.S. "Minx," 211. H.M.S. "Rattler," 205, 207. first-class cruisers, 218. Ancient galleys fitted with paddle-wheels, 3. Andrese's trials with two- and four-bladed screw propellers, 238, 239. Angle of entrance of ship's model, Seaton's rule for, 44. Apparent slip, 57. negative, 59. "Archer," H.M.S., experiments on, dia- meter of screw, 209. analysis of trials, 212. " Archimedes," screw steamer, 17, 21. 6.3,, trials of, 202. Area of paddle floats, 82. Arms of paddle-wheels, 85. Atkins, multiplicity of propellers, 33. Augmented resistance of ship, 50, 109- 128. Axioms relating to screw propellers, 170. Bacon, Roger, observations on ship pro- pellers, 3. Barber, J., application of turbine to propulsion, 6. Beaufoy's experiments and formula, 39. Belfast and Glasgow, steam communica- tion between, 14. Bell, Henry, built " Comet," p.s., 1811, 10. Bending moments of propeller blades, curve of, 174. Bernouilli methods of propulsion, 5. Bessemer, hydraulic propeller, 97. turbine propeller, 23. Bevis' feathering screw, 159, 161. Billington and Newton, bronzes, 199. Blade area, ratio of thrust to, Sells, etc., 224. flange, number of bolts in, 182. Blades, cast iron, in H.M. service, 200. cast steel, 200. dovetailed forged bosses, 201. forged steel, 200. hollow, 190. loose — bronze screws, examples of, 244. of screws, Griffiths* improvements, 29, 30, 34. propeller, effect of surface on thrust, Blechynden, 154. Sells, 154. Froude, 154. effect of number on thrust, 154. number of, 151. screw, corrugated ribbed, etc., 27. various number of, 157, Blasco de Garay, steamship, 3. Blechynden, experiments with various screw propellers, 231. propeller blades, effect of surface on thrust, 154. proposals, 232. Boss of propellers, Roberts' improvements, 26. screw, length of, 168. Bosses, different experiments with H.M.S. "Conflict," 167. elongated, 168. paddle-wheel, 84. Bourne's internal turbo-motor, 28. Bow screw, Wakefield Pirn, 24. Howden, J., 33. Bramah's propeller, 7. Brandon, s.s., compound engines, 28. INDEX. 249 Brass, naval, 199. "Britannia," paddle steamer, 12. " Britannic," s.s., submersible screw, 149. Bronze, aluminium, 195. manganese, 195, 196. phosphor, 195. screws, loose blades, examples of, 244. Stone's, 195, 197. Bronzes, Billington and Newton, 199. Brown, Samuel, chain-driven ferry-boat, 15, gas engine, 15. special prize for screw propulsion, 1825, 15. Buekholz, triple screws, 26. Bull's metal, 199. Butterly Co., engine builders, 12. Callaway and Purkiss, hydraulic propul- sion, 24. Carpenter, lifting screw, 21. plan for double screws, 26. Cast iron, 199. blades, in H.M. service, 200. screw propellers, examples of, 245. Cast steel blades, 200. Cavitation, QQ, Centrifugal force, effect on propellers, 183. wheel, guide-blades to screws, 24. "Charkieh," s.s., experiments with three- and six-bladed propellers, 156. "Charlotte Dundas," experiments with steamer, 10. "Chelmsford," t.s.s., trials of, analysis, 243. Chinese use of paddle-wheels, 2. Church, William, double wheel propellers, 14. "City of Glasgow," s.s., 25. Claudius, use of paddle-wheels by, 2. "Clermont," paddle steamer, 1807, 10. Coaxial double screws, Captain Smith, 19. "Comet," H.M.S., paddle steamer, 12, 15. loss of s.s. "Comet," 12. trials of, 11. Common screw, 105. improved form of, 133. surface of, 105. Compound engine on screw steamer, 28. " Conflict," H.M.S., experiments with dif- ferent bosses, 167. Conoidal screw, Rennie, 20. Construction of propellers, material used in, 194. Corrugated screw blades, 27. Cruisers, "Drake" class, trials with dif- ferent screws, 218. first-class, analysis and trials, 218. Cummerow, Charles, screw propeller, etc., 16. Cunard steamers, four screws, 145. Curve of bending moments of propeller blades, 174. of thrust, Dr W. Froude, 122. Cylindrical casing to screw, Rigg's proposal, 32. " Dauntless," H. M.S., experiments with, 146. fitted with new stern, 146. Dawson's service of steamers, London to Gravesend, 1818, 14. Delta metal, 199. Depth of water, influence on speed, 51. Deschamps, diving boat, 28. Destroyers, resistance, total, of, 53. varying, 53. Details of modern naval screw, 138. Developed surface, screw propeller, 105. " Diadem," H.M.S., trials of, 212. Diameter of bolts of propeller blades, 183. of paddle-wheel, 80. of propeller boss, 3 66. of screw boss, rule for, 167. of screw propeller, 105, 169. rule for, 171. of screw shaft, rules for, 180. Differentiation by Froude, Dr W., 123. Disconnected paddle shafts, Wilkinson, 17. Diving boat, submersible, Deschamps, 28. "Doris," H.M.S., trials of, and experi- ments, 213. Double propellers, Church's proposals, 14. screws, Bennet Woodcraft, 25. Carpenter's plan, 26. Howden, J., 33. Reed, Sir E. J., proposal of, 35. Droop and Martin, steering steamships, 33. "Drake'' class of cruisers, trials with different screws, 218. "Duncan," experiments with H.M.S., screws of different number of blades, 213. experiments with, 157. Du Quet, further experiments, 4. propeller, 3. submarine helix, 5. "Dwarf," H.M. S., analysis and trials of, 208, 210. Dynamometer experiments, 117. Effect of centrifugal force on propellers, 183. of trim on speed of ship, 52, Effective horse-power, 58. Rankin formula for, 58. Efficiency, total, of machinery and screw, 116. of engine shafting, 116. of propeller, 116, 126. of propulsion, 49. of ship and machinery, 116. Elongated bosses, 168. 250 INDEX. "Emerald," H.M.S., experiments with, 155. experiments with different number of blades, 155, 214. "Enterprise," paddle steamer, Calcutta, 1825, 15. Ericsson, John, paddle screws, 18, 19. "Ermack," I.R.S., four screws, 145. Experiments on immersion of screw, Rennie, 232. s.s. "Charkeih," with three- and six- hladed propellers, 156. Isherwood's, 1 54. onH.M.S. "Duncan," 157. with screws different number blades, 213. with different number blades, H.M.S. "Shannon," 213 "Dauntless," H.M.S. , 146. "Flying Fish," H. M.S., 212. Hirsch screws, 137. model screws, Charles Sell's, 219, 220, 221, 222, 223, 224. propeller bosses, 167. propellers having one, two, three, four, and six blades, each same size, Sell's, 225. screws of various surface and number of blades, Walker, 240. "Shannon," H.M.S., 157. twin screw ships, 142. twin versus single screw ships, 235, 236. various propellers, Blechynden, 231. Isherwood, 226. various screws, s.s. "Ethel," "Walker, 240. Yarrow's, 117, 237. Failuke of H.M.S. "Dauntless," 146. "Fairy," H.M.S., analysis and trials of, 211. trials of, 209. Feathering gear paddle-wheels, 87, 88. paddle-wheel, Galloway, patent, 13. Morgan, 13. Wright's method, 14. paddle-wheels, 72. screw, Bevis, 159, 161. Maudslay, 159, 160. Ferry-boat, chain-driven, 15. First-class cruisers, analysis and trials, 218. Fitch's experiments with steamers, 7, 8. Fitting loose blades to boss, method of, 182. Fittings, paddle float, 89. Flat blades for screws, 162. Float, mean velocity of, 55. pitch of, 82. Floats, number of, 83. propulsion of, 83. rules for area, 82. steel versus wood, 83. Floats, thickness of, 83. Flow of water to paddle float, 58. " Flying Fish," H.M.S., experiments with, 212. experiments with Mangin propeller, 133. Forged bosses, dovetailed blades, 201. steel blades, 200. Formulae, Seaton's, acting surface of screw propeller, 165. diameter propeller bosses, 168. Four-bladed screws, 163 Four screws on the Cunard steamers, 145. in I.R.S. "Ermack," 145. of " Lusitania," 145. of Mersey ferry steamers, 145. Fourness and Ashworth paddles, raising, 7. French Navy, triple screws in, 143. Frictional resistance of common screw, 111. of Griffiths' screw, 112. of screw, rule for, 113. Froude, R. E., analysis of screw trials, 118. formula for thrust, 119. Froude, Dr W., analysis of ship trials, 121. curve of thrust, 122. differentiation, 123. index values for various lengths of ships, 40. for various materials, 40, 41. propeller blades, effect of surface on thrust, 154. Fulton, trial experiments, 10. Fundamental principle of propulsion, 2. Galloway, Elijah, patent paddle-wheel, 13. Gas engine, Brown, Samuel, 15. GemmeH's patent, 19. Genevois' method of propulsion, 5. Geometry of screw, 186. German Navy, triple screws in, 144. " Great Britain," s.s., 1845, 32. "Great Eastern," s.s., 1858, 29. "Great Western,'' paddle steamer, New York, 1838, 20. Green, hydraulic life-boat, 97. Griffiths, improvements in screw blades, 29, 30, 34. screw with four blades, 31. Griffiths' screw patent, 1860, 131. frictional resistance of, 112. self-adjusting blades, 24. Guide cylinders to screws, Maudslay, 27. Haddon, J. C, ribbon blades, 20. Hale, method of propulsion, 15. system hydraulic propulsion, 93. Hamer, turbine propulsion, 21. Helix, angle of, 105. pitch of, 104. surface of, 105. INDEX. 251 "Herongate," s.s., trials of, 237. Hirsch, Herman, improvements in screw propellers, 1866, 32, 36. in propeller shafts, 36. patent screw, 134. Hollow screw blades, 190. screw shafts, 27. Hook's windmill compared to propellers, 3. Horse-power imparted to propellers, 127. Howden, J., bow screws, 33. double screws, 33, 143. sheet-metal propeller blades, 35. two-screw system, 143. Hull, Jonathan, patent steamship, 5. Hunt's steering screws, 20. Hydraulic life-boat, Green, 97. propeller, Bessemer, 97. calculations of, 99. efficiency of, 101. propulsion, 91. Callaway and Purkiss, 24. Hales system, 93. Ramsay system, 92. Ruthven, 21, 24, 93. Thornycroft system, 95. t{ Hydromotor,"s.s., 91. Immersion of screw, effect of, 232. Increasing pitch propeller, proposed by Bourdon and Tredgold, 17. Woodcroft's, 17. Index value, Froude, 40, 41. Indicated thrust, 114. Influence on speed of tidal currents, 52. of depth of water, 51. " Iris," H.M.S., propeller blades, effect of surface on thrust, 154. trials of, 215, 216, 217. Iron, cast, 199. Isherwood's experiments, 154. experiments with various propellers, 226. Italian Navy, triple screws, 143. Johnson's patent, two pairs of paddle- wheels, 32. JoutTroy's method of propulsion, 6. " Kingston," paddle steamer, 12. Kirk, analysis of ship model, 43. method of estimating wetted skin, 42. proposal for steering steamships, 33. Krupp, Alfred, forged steel screws, 31. Kunstadter's patent screws, 147. Laird, screw shaft tunnel, 21. Lang, Oliver, designed H.M.S. "Comet,'' 12. Length of screw propellers, 157. Lifting screws, Carpenter's proposals, 21. Maudslay's plan, 22. Lignum vitse, Penn's, 28. Liquid fuel, 32. Longitudinal sections of propeller blades, 179. Loose blades versus solid screws, 180. Lowering screw, Phipps, 1850, 150. Low's patent screw, 19. " Lusitania," s.s., four screws, 145. Lyttleton, William, propeller, 8. Macintosh, reversible blades, 24. Manby, Aaron, Staffordshire engineer, 12. Manganese bronze, 195, 196. Mangin screw, 132. Marine propulsion, Trevithick patents and proposals, 13. Marquis of Worcester claimed as inventor, 3. Materials used in construction of propellers, 194. Maudslay feathering screw, 159, 160. guide cylinders, 27. lifting screws, 22. method of adjustment of pitch, 27. self-adjusting screw blades, 24. Mean pitch of screw, 125. velocity of float, 55. Mercantile four-bladed screw, 136. "Mercury,"H.M.S.,trialsof,215,216,217. Mersey ferry steamers, four screws, 145. Metal, Bull's, 199. Delta, 199. Sterro, 199. " Meteor," H.M.S., triple screws, 1855, 143. Method of delineating a true screw accur- ately, 187. fitting loose blades to boss, 182. forming surface of screw blades, 106, 107, 108. founding propeller, 191. measuring pitch of propellers, 192. propulsion, Brown, Samuel, 15. Hale, William, 15. Miller, Patrick, double-hulled boat, 7. experiments with paddle-wheels, 7. Millington, John, method of propelling vessels, 13. steering screw, 14. "Minx," H.M.S., analysis and trials of, 211. trials of, 209. Modern naval screws, details of, 138. practice, number of blades, 158. Multiplicity of propellers, Atkin's claim, 33. Multipliers for Seaton formula for wetted skin, 45. Mumford, estimation of wetted skin, 44. Naval brass, 199. Napier, David, arrangement of two pro- pellers, 21. Negative apparent slip of screws ; 59. slip of screws, examples of, 60, 61. explanation of, 62. 252 INDEX. Niepce, J. C, hydraulic propulsion, 14. internal combustion machine, 14. Number and position of screws, 140. of blades, modern practice, 158. of bolts in blade flange, 182. of floats, 83. of propeller blades, 151. One-paddle-wheeled ship, 76. Oscillating engines, 1830, Church, 16. 1827, Maudslay, 16. Paddle blades, ribbed or corrugated, 19. boat, on Bridgewater Canal, 1793, 8. float, flow of water, 56. fittings, 89. reaction of, 81. wheel arms, 85. bosses, 84. diameter of, 80. feathering, 72. feathering gear, 87, 88. rims, 86. ship with wheels, 3, 76. ship with one wheel, 76. speed of, 56. steamers, example of, 79. steering of, 84. wheels, advantage of, 69. two pairs of, Johnson patent for, 32. "Pallas," H.M.S., trials of, 214, 215. Pappin's proposals to Prince Rupert, 3. "Paragon,' 1 paddle steamer, 10. Pattern of propeller blades, 191. Pearce, perforated blades, 27. Pearson, Richard, shipbuilder. Yorkshire, 12. Penn's lignum vitse, 28. Perforated blades, Pearce, 27. Performance of screws, of same surface and pitch, 140. of screws varying only in diameter, 139. of sundry steamships, 246. Phipps' submersible screw, 1850, 150. lowering screw, 1850, 150. Phosphor bronze, 195. Pirn, Wakefield, bow screw, 24. Pitch, increasing screw propeller, 106. of float, 82. of propellers, method of measuring, 192. ratio, influence of, 64. screw propellers, 108. true, screw propeller, 105. variable screw propeller, 105. varying screw blade, 190. Position of quadruple screws, 151. of screw, 145. of triple screws, 151. of twin screws, 150. Practical method of drawing screw, 189. Primitive methods of propulsion, 2. "Prince Consort," H.M.S., trials of, 240. " Prince of Coburg," paddle steamer, 12. Principle of propulsion, fundamental, 2. Progressive trials of t.s.s. "Chelmsford," analysis, 243. Projected surface screw propeller, 105. Propeller, abaft rudder, 148. blade bolts, number and size of, 183. blades, bending moments, rule for, 125. dovetailed to boss, 28. effect of number on thrust, 154. of surface on thrust, 154. Blechynden, 154. Froude, 154. "Iris," H. M.S., 154. elliptical section, 31. longitudinal sections of, 179. number of, 153. pattern of, 191. sections at root, 190. Sells, 154. sheofc metal, 33, 35. thickness at root, rule for, 177. thickness of, 173. at axis, rule for, 175. radially, 177. at tip, 177. transverse section, 177. boss, diameter of, 166. experiments with, 167. bosses, thickness of metal, 180. Bramah, Joseph, 7. Propellers, efficiency of, 126. Hirsch's improvements, 36. racing of, 68. two, Napier, D., 21. Proportions of screw blade surface, 164. Propulsion of floats, 83. hydraulic, 91. Allen, 92. Ruthven, 92. Niepce, 14. theories of, 54. Propulsive powers, test of efficiency, 49. Pumphrey, proposals for twin screws, 16. Purkiss, Jacob, angle of screw blades, 15. patent coaxial screws, 15. oblique floats to paddle-wheel, 16. Quadruple screws, position of, 151. Racing of propellers, 68. Radial versus feathering wheels, 71. Ramsay, system of hydraulic propulsion, 92. Rankin, formula for E.H.P., 58. Ratio, pitch, influence on 64. of thrust to blade area, Sells, 224. "Rattler," H.M.S., 21-23. trials of, 203, 207. versus " Alecto," 203, 205. Reaction of paddle float, 81. Real slip of screws, 57. Reed, Sir E. J., double screws, 35. INDEX. 253 Rennie, conoidal screw, 20. early screw steamers, 28. experiments on immersion of screw, 232. Residual resistance, rule for, 48. Resistance, augmented, 50, 109, 128. frictional, screw propeller, 110. of screw, to calculate, 111, 113. of destroyers, 53. of screw propeller, 109. of ships, 39. of ship, total, 109. of ship, tow-rope, 109, 129. of torpedo boat, Yarrow, 237. tangential, of ship, 40. through air, 51. through water of surface of various materials, 48. Reversible blades to screw propeller, Macintosh, 24. Ribbon blades, Haddon, patent, 20. Rigg's proposal, cylindrical casing to screw, 32. Rims, paddle-wheels, 86. Roberts' improvements, propeller bosses, 26. Robertson, oblique floats to paddle-wheels, 16. t( Royal William, ,J paddle steamer, Liver- pool to New York, 1838, 20. Gravesend, 1838, 20. Rudder before propeller, 148. Rudders, double, 21. Rule for bending moments of propeller blades, 175. calculating frictional resistance of screw, 113. diameter of screw shaft, 180. diameter of propeller bosses, 167. screw blades, 164. thickness of propeller blades at axis, 175. at root, 177. Rumsay, James, method of propulsion and experiments, 8. Russian Navy, triple screws, 144. Ruthven, John, hydraulic propulsion, 21, 24, 93. "Salamander," H.M.S., 12. "Savannah," paddle steamer, 1819, 14. Savory's engines in ship on river Fulda, 4. suggestion for steamships, 4. Schiele turbo-motor, 28. "Scotia," paddle steamer, built 1862, 31. Screw blade, surface proportions of, 164. with varying pitch, 190. blades capable of adjustment, Wain's method, 28. methods of forming surface, 106, 107, 108. movable in boss, Woodcroft, 22. rule for, 164. Screw blades, shape of, 163. boss, length of, 168. coaxial, Furkiss, 15. common, 105. Screw propeller, 104. actual thrust, 116. arrangements for working, Cummerow, 16. axioms relating to, 170. developed surface, 105. diameter of, 105, 168. rule for, 171. experiment with, Hirsch, 137. flat blades for, 162. four blades, 162. Griffiths, 31. frictional resistance, 110. geometry of, 186. Griffiths' patent, 131. Hirsch patent, I860, 29, 135. improvements, Hirsch, 32. Thornycroft, 34. increasing pitch, 106. length of, 157. Mangin, 132. mean pitch of, 125. mercantile, form of, 136. method of founding, 191. Millington's method of, 13. patent, Kunstadter, 147. performance of, varying in diameter only, 139. pitch of, 105. pitch ratio, 108. position of, 145. practical method of drawing, 189. projected surface, 105. resistance of, 109. surface ratio, 108. three blades, 162. useful effect of, 128. variable pitch of, 105. velocity of, 108. weight of, 185. Woodcroft's increasing pitch pro- posals, 1832, 17. propellers, forged steel, Krupp, 31. mercantile marine, examples, 245. Navy, examples, 244. two blades, 158. four blades, 162. six blades, 237. various forms of, 130. shaft, diameter of, rules for, 180. tunnel, Laird's patent, 21. shafts, hollow, 27. submersible, 148, 149. "Britannic," s.s., 149. Phipps, 150. Screws, same surface and pitch, perform- ance of, 140. with increasing pitch, Tredgold, 1827, 17. 254 INDEX. Screws, Woodcroft, 1832, 17. rectangular blades, Taylor's patent, 19. Seaton's formulae for acting surface of screw propeller, 165. for calculating thrust, 123, 124. for diameter propeller bosses, 167. for thickness of screw blades, 173-177. method of estimating wetted skin, 44. multipliers for formula for wetted skin, 45. rule for angle of entrance of ship model, 44. friction al resistance of screw blades, 110. Section of propeller blades, elliptical, 31. Sections at root of propeller blades, 179, 190. Self-adjusting blades, Griffiths, 24. Maudslay, 24. Sells, Charles, experiments with model screws, 219, 220, 221, 222, 224. experiments with propellers with one, two, three, four, and six blades, 225. propeller blades, effect of surface on thrust, 154. relation of thrust to blade area, Sells, 224. Shafts, propeller, improvements, Hirsch, 36. Shallow - draught screw ships, stern, Thornycroft, 35, 36. steamers, Yarrow, 36, " Shannon," H.M.S., experiments with different number of blades, 213. trials with different kinds of screws, 214. Shape of screw blades, 163. Sheet-metal propeller blades, 33, 35. Howden, 35. Ships, resistance of, 39. Shorter, Edward, portable screw propeller of, 1800, 9. " Sirius," p.s., New York, 20. Six-bladed screw propeller, trials of, 237. Skin resistance of ships, 40. Slip of propellers, apparent, 57. real, 57. negative, examples of, 60, 61. explanation of, 62. Smith, Captain, coaxial blades, 19. Frank Pettit, patent screws, 1836, 17. Smooth water necessary to high efficiency, 51. Solid screw v. loose blades, 180. cast-iron screws, examples of, 245. Speed of paddle-wheels, 56. Steamer service between Glasgow and Belfast, 1818, 14. Steering of paddle-wheel, 84. screws, 147, 149. Hunt, 20. Millington, 14. "Stratheden,"s.s., fitted with, 147. Steering steamships, Droup Martin, 33. Kirk's proposals, 33. Stern, new one fitted on H.M.S. "Daunt- less," 146. of twin-screw steamer, Dudgeon's pro- posal, 32. Stern-wheelers, 76. Sterro metal, 199. Stevens, screw propeller turned by rotary engine, 10. Stone's bronze, 195, 197. " Stratheden," s.s. , fitted with steering screw, 147. Submersible screw, 148, 149. Surface ratio proportions of screw propeller, 108. Symington's experiments with p.s. " Char- lotte Dundas," 9. patent, 7. Symons' twin screw, 29. Taylor, James, and Patrick Miller, 7. rectangular blades, 19. "Thames," steamer, made voyage from Glasgow to London, 12. Theories of propulsion, 54. Thickness of floats, 83. of metal in propeller bosses, 180. of propeller blades, 173. at root, rule for, 177. at tip, 177. radially, 177. Thornycroft 's hydraulic propulsion, 95. improvements to screw propellers, 34. stern for shallow-draught screw ships, 35. Three-bladed screws, 162. Three-paddle-wheeled ships, 162. Thrust, calculations of, 57. experiments on, Yarrow, 237. Froude's formula, 119. Seaton's formula for calculating, 123, 124. indicated, 114. of screw, 114. Tidal currents, influence on speed, 52. Torpedo boat, experiments with, Yarrow, 117. resistance of, 237. Torsion meter, Wimshurst, 1850, 25. Total resistance of destroyers, 53. of ship, 109. Tow-rope, resistance of ship, 109, 129. Towing vessels, Chatham Yard, 3. Transverse sections of propeller blades, 177. Trevithick, Richard, proposals for marine propulsion, 13. Trials, analysis and performances of sundry steamships, 246. and experiments with H.M.S. "Doris," 215. of s.s. "Archimedes," 202. INDEX. 255 Trials of H.M.S. " Diadem," 212. of s.s. (( Herongate,"237. of H.M.S. " Prince Consort," 240. with different screws, " Drake " class of cruiser, 218. H.M.S. "Dwarf," 208. H.M.S. "Emerald," different number of blades, 155. H.M.S. <( Fairy," 209. H.M.S. " Iris," 215, 216, 217. H.M.S. "Mercury," 215, 216, 217. H.M.S. " Minx," 209. Mangin screw, on H.M.S. "Flying Fish," 133. H.M.S. "Pallas," 214, 215. H.M.S. "Rattler," 203, 204. H.M.S. "Shannon," different kinds of screws, 214. steam, " Waterwitch," H.M.S., 95. Trim of ship, effect on speed, 52. Triple screws, American Navy, 144. Buckholz, 26. French Navy, 143. German Navy, 144. Italian Navy, 143. H.M.S. "Meteor," 143. position of, 151. Russian Navy, 144. Turbine propeller, low-pressure, Bessemer patent, 1846, 23. propulsion proposed by Hamer, 1843, 21. Turbo-motor, Schiele, 28. Bourne, 29. Twin screw on one shaft, 147. position of, 150. ships, experiments with, 142. steamer, stern for, 32. steamers, early, Rennie, 28. Twin screws, interlocking, Taylors' patent, 19. Symons' arrangements, 29. versus single, Admiralty experiments, 234. Two-bladed screw, 158. screws, Howden's system, 143. Useful effect of screw, 128. Values of V 1 " 83 and V a , 247. Various forms of screw propellers, 130. number of blades, 157. Velocity of screw propeller, 109. Walker, experiments with various screws, s.s. "Ethel," 240. "Water slip, loss by, 126. H.M.S. "Waterwitch," 93. calculations of efficiency, 102. steam trials, 95. Watt, James, suggestion for screw, 6. Weight of screw propellers, 185. Wetted skin, Kirk's method, 42. Mumford's method, 44. numerous examples of, 47. Seaton's method, 44. Wheels, position of, 74. without outer rims, 74. Wilkinson, disconnected paddle-shafts, 17. Wimshurst torsion meters, 1850, 25. Woodcroft, Bennet, double screws, 25. blades movable in boss, 22. patent screw propeller, 17. Worcester, Marquis of, claimed as inventor, 3. Wright, Richard, patent feathering paddle, 14. two double- cylinder engines, 14. Yarrow, experiment with torpedo boat, 117, 237. experiments on thrust and resistance, 237. shallow- draught steamers, 36. " Yorkshireman," paddle steamer, 12. PRINTED BY NEILL AND CO., LTD., EDINBURGH