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/cu31924032191938 Cornell University Library arY91 Reports of experiments on the properties 3 1924 032 191 938 olin.anx REPORTS OF EXPERIMENTS OX THE PROPERTIES OF METALS FOR CAB'ON, iSD THE QUALITIES OF CAMON POWDER; WITH AX ACCOUNT OF THE FABRICATION AND TEIAL OF A 15-INCH QUN. BY CAPT. T. J. RODMAX, OF THE ORDNANCE I3EI»^!tIlTIwIE3SrT, XT. S. A.ttl>/L~Y. Ms ^atjoritj af tk Bttxttm of <89ar. BOSTON MASS.: CHARLES H. CROSBY. 1861. J.C.F /f St /s LA/ n '. ■' t.. g_L INlVirlhoiT 1 V LiERA.T/ PREFACE. The following Reports might have been considerably abridged, and their contents rendered less repulsive to the general reader by re-writing them after the experiments had all been completed ; but for the use of the prac- tical student, it was deemed preferable to publish the Reports as they were originally made. In these Reports I have endeavored, in every case, to record results exactly as they were evolved by experiment, accompanying them by such explana- tory remarks as appeared necessary to a clear understanding of the circum- stances under which they were evolved. Few, if any, of the subjects experimented upon are regarded as exhausted, the experiments having in most cases only contracted the limits of uncer- tainty, without having arrived at absolute truth. My letter to the Colonel of Ordnance, (p. 93,) is re-published, reference to it having been found necessary in writing these Reports. T. J. RODMAN, Copt, of Ordnance. Apkil 17, 1861. CONTENTS. Report of Experiments made foe the purpose op further Testing the Relative Merits op Capt. Rodman's Mode of Cooling Cannon, as compared with the Ordinary Method ...... Preliminary Trials ...... Table of Mechanical Tests for Preliminary Castings Charge of Metal for Casting 10-inch Trial G-uns, No. 331 and 332 Mode of Manufacture of Iron used Description of Air Furnace Casting Condition of Pits Cooling Inspection of Guns Proof of Guns Endurance of Guns Of the Powder used in the Proof Enlargement of Bores by Firing Enlargement of Chambers Table of Mechanical Tests of Metal from different parts of Guns Table comparing Head with Muzzle specimens . Comparison of Tensile Strength of Radial with Tangential specimens from the same Gun . Table comparing the Compressibility of Metal from like parts of Solid and Hollow Cast Guns .... Table showing the Change of Form by Compression Table showing Compression, Restoration and Set of Radial specimens from each Gun ........ Table showing Compressibility of Iron from both Guns Table showing Extensibility of Iron from both Guns . Summary of all the Tests to which the Metal of these Guns was subjected Table comparing Endurance, &c, of Trial Guns made up to Date of this Report 3 3 5 5, 6 9 12 13 13 15 15 16 18 19 20 20 21 22, 23 24 ' 24 26 27 VI CONTENTS. II. Initial Velocity or 10-inch Shot and Shells PAGE 29 III. The Velocimeter 31 IV. Bursting Gun Heads 35 Indentations in Coppee and Pressure due to Falling Weight . . . 37 Table comparing Actual with Computed Pressures producing equal Indentations in Pure Copper ........ 38 VI. Of the Effects of Different Eates op Application op Straining Forces upon the Bodies to ■which thet are applied ..... 41 VII. Op the various kinds op Strain to which a Gun is subjected at each Discharge Tangential Strain ...... Longitudinal Strain ...... Crushing Force ...... Transverse Strain ...... Expressions for Tendencies to Eupture different kinds of Eesistance 43 43 44 46 47 51 VIII. Bursting Eppects op dipperent Weights op Powder and Shot, in Guns op dif- ferent Calibre ........ Position of Shot when Maximum Pressure is attained 52 54 IX. Subjects requiring Investigation 55 X. Experiments made for the purpose op determining the Eelative Endurance of Guns made from the same Iron, but melted in Furnaces of different con- struction ; also, that of those made from the same Iron, melted in the same Furnaces, but differently cooled, one Gun being Cast Solid, and cooled from the Exterior, and the other Cast Hollow and cooled from the Interior Selection of Iron for six 10-inch Columbiads for Experimental Purposes Charges of Iron for Casting the first set of Triplicate 10-inch Columbiads Of the Flasks, the Pits, and the Furnaces Casting Cooling Inspection and Proof . Of the Powder used in Proof Mode of Discharging Guns Endurance of Guns . Tables of Interior Measurements Tables showing the Diameters of Vents 57 59 GO Gl G2 63 G4 64 65 65 68-77 78 CONTENTS. Vll Meteorological Observations, and number of rounds fired each day Table showing Tenacity and Density of Metal from the heads, and from different parts of the Guns ...... Table showing Extension, Restoration, and Permanent Set of Inner Specimen from West Point Gun, No. 983 . Table showing Extension, Eestoration, and Permanent Set of Outer Specimen from West Point Gun, No. 983 Table showing Extension, &c, of Inner Specimen from Fort Pitt, Solid Cast Gun, No. 335 Table showing Extension, &e., of Outer Specimen from Port Pitt, Solid Cast Gun, No. 335 . Table showing Compression, &c, of Inner Specimen from West Point Gun, No. 983 Table showing Compression, &c, of Outer Specimen from West Point Gun, No. 983 . Table showing Compression, &c, of Inner Specimen from Port Pitt Gun, No. 335 Table showing Compression, &c., of Outer Specimen from Port Pitt Gun, No. 335 . Recapitulation ........ PAGE 79 80 81 82 83 84 85 86 87 88 89 XL Report on the Causes of- Difference in the Endurance op Canxon, when Cast Solid, and Cast Hollow, cooled from the Exterior and the Interior . XII. Beport of the Fabrication and Proof, up to 2450 Service Charges each, of two 10-inch Trial Guns; one Cast Solid, and Cooled from the Exterior, and the other Cast Hollow, and Cooled from the Interior Of the Iron ...... Charges and Distribution of Metal Condition of Pits ... Preparation of Moulds Casting .... Cooling ... • • Temperature of Pits ... Ordinary Proof of these Guns Proof Charges ..... Service Charges ..... Of the Powder ..... Difference in Velocity of Shot, due to difference in Enlargement of these Guns Difference in Maximum Pressure of Gas, due to difference in Enlargement Mechanical Tests of Metal in these Guns . Enlargement of Bores by firing . . • • • 112 " 93 10L 101 101 102 102 102 103 104 105 105 105 105 108 109 111 -120 Vlll CONTENTS. Tables of Enlargement of Vents .... Meteorological Observations during the period of firing these Guns Endurance of this pair of Guns . . . • Comparison of Solid Guns of 1857 . Comparison of Eort Pitt Guns of 1857 . Comparison of Fort Pitt Guns of 1857 and 1858 . . • • Table comparing Endurance of all the pairs of Solid and Hollow Cast Guns made up to date of this Eeport ....•• Eecapitulation .....••• Eatio of good and bad Guns ...•■• XIII. Eepokt of Experiments made at Alleghany Arsenal, by Capt. T. J. Eodman, U. S. Ordnance Department, in the Years 1857 and 1858, for determining THE PROPERTIES OF GUN METAL, THE EeSISTANCE WHICH GUNS CAN OFFER TO A Bursting Eorce, the Actual Pressure per square inch due to different Weights of Powder and Shot, &c, &c. XIV. Transverse Eesistance of Guns XV. Deflection of Bars under loads equally distributed along their whole lengths .... XVI. Transverse Eesistance of Hollow Cylinders Table of Eesistances of Hollow Cylinders to a Central Eorce acting on different Lengths of Bore XVII. Effects of Chambers on Endurance of Guns XVIII. Thickness of Metal in the Breeoh XIX. Tangential Eesistance of Hollow Cylinders Tables showing Bursting Pressures of Hollow Cylinders XX. Extensibility and Compressibility of Gun Metal Tables of Extensibility .... Tables of Compressibility .... Table showing Effects of Bepetitions of a Strain of Constant Intensity Deductions and Conclusions Compression ..... Intermittent Force of Constant Intensity . Capacity for "Work .... XXI. Of the Absolute Pressure of Gas in the Bore of a Gun Pressure per square inch due to Proof Charges in the 42-pdr. Gun 122- 126- 121 124 125 128 129 129 133 135 137 139 141 143 143 145 152 153 154 155, 156 157 15S-161 162-165 166, 167 168 169 170 171 174 176 CONTENTS. IX PAGE Preliminary Trials with Accelerating Cartridges .... 176 " " with Grained Powder . . . . .176 Pressure of Gas at different points along the Bore . . . 176 XXII. Constant Weight of Projectile and Increasing Chaege . . . 177 Constant Weight of Charge and Increasing Weight of Projectile . . 177 XXIII. Effects of Windage in the Cartridge ..... 178 XXIV. Pressure in Epbouvette Moetar . . . . . .179 XXV. Table of Pressures of Gas at Bottom of Bore, and at two Calibres from Bottom, in a 42-pdr. Gun ....... 180 XXVI. Effects of Sabots ........ 180 XXVII. Greater Uniformity of Pressure from Acceleeating Chaeges . . .181 XXVIII. Time of Combustion of Chaeges . . . . . 182, 183 XXIX. Bursting Tendencies of different Charges in Guns of different Calibee 184—186 XXX. Eepetition of Constant Strain . . . . . .186 XXXI. Bepoet of Experiments made by Captain T. J. Bodman, at the Watertown Arsenal, in the second half of 1859, for the purpose of determining the proper Qualities of Iron, Exterior Model, &c, for Cannon, with Special Beference to the Fabrication of a 1 5-inch Gun . . . .191 XXXII. Experiments to determine the Belation between the Thickness and the Tangential Besistance of Hollow open-ended Cylinders, by Bursting them with Powdee ........ 191 Table of Eesults . . . . . . . .192 Table comparing these Besults with those deduced from the Hypothesis that the Strain from a Central Eorce diminishes as the square of the distance from the axis increases . . . . .193 XXXIII. Difference in Pressure due to Equal Columns of Powder behind Equal Columns of Metal when fired in Guns of different Calibre . . 195 Table of Besults 196, 197 Discussion of Besults ...... 197, 198 XXXIV. Difference in Pressuee of Gas and Velocity of Shot, due to Equal Weights of Powdee of the same Quality, in all respects except in Diameter of Grain, and fired from the same Gun . • • • .199 X CONTENTS. Tables of Eesults ..... Proof Eange of Towder used .... Table showing the Velocity of Shot and Pressure of Gas FAGE 200, 201 202 203 XXXV. Determining the Pressure exerted by Exploded Gunpowder, when the Products or Combustion occupy a certain number of times the Volume OCCUPIED BY THE POWDER BEFORE COMBUSTION .... Table showing the Eesults ...... 204 206 XXXVI. Pressures of Gas due to Unequal Charges of Powder when Burned in Spaces bearing a Constant Eatio to those Volumes . . . 207 Table of Eesults ..... .208 XXXVII. Of the Absolute Pressure of Powder when Burned in its own Volume 208 Table of Eesults . . . . . . .209 XXXVIII. Tables of Properties of Iron in 42-pdr. Gun, No. 336, Cast Hollow at the Fort Pitt Foundry, of Bloomfield Iron, (No. 2 Pig,) and Burst at the 491st fire, with 10 lbs. of Powder, and one solid Shot . . 211-216 XXXIX. Determination of Exterior Models of Guns XL. Effects of Compressibility .... XLI. Termination of Bore ..... XLII. Preliminary Castings for 15 -inch Gun Table showing Extension, &e., of Cylinder A, 0. . " Extension, &c., of Cylinder A, I. . " Number and EiFects of Eepetitions on A, " Extension, &o., of Cylinder on B, 0. " " Extension, &c., of Cylinder on B, I. Number and Effects of Eepetitions on B Extension, &c, of Cylinder C, 0. . Extension, &e., of Cylinder C, I. . Number and Effects of Eepetitions on C, " Extension, &c, of Cylinder D, 0. . Extension, &c, of Cylinder D, I. . Number and Effects of Eepetitions on D, Compression, &c. of Cylinder A, 0. Compression, &c, of Cylinder A, T. Compression, &c, of Cylinder B, 0. Compression, &c, of Cylinder B, T. 217 219 224 225 227 228 229, 230 231 233 234 23.". 237 238 239 240 241 242 243 244 2JC CONTENTS. XI Table showing Compression, &c, of Cylinder C, 0. " " Compression, &c, of Cylinder C, T. " Compression, &c, of Cylinder B, 0. " " Compression, &c, of Cylinder B, T. " comparing Extensibility of Outer Specimens from Cylinders A, D, C and B " comparing Kestoration from Extension of Outer Specimens from Cylinders A, B, C and D . " comparing Permanent Set from Extension of Outer Specimens from Cylinders A, B, C and !>...•• " comparing Extensibility of Specimens from near Axes of- Cylinders A, B, C and D " comparing Restoration from Extension of Specimens from near Axes of Cylinders A, B, C and D . " comparing Permanent Set from Extension of Spesimens from near Axes of Cylinders A, B, C and D " comparing Compressibility of Outer Specimens from Cylinders A, B, C and D . " comparing Eestoration from Compression of Outer Specimens from Cylinders A, B, C and D . " comparing Permanent Set from Compression of Outer Specimens from Cylinders A, B, C and B " comparing Compressibility of Specimens cut transversely 13 Inches from lower ends of Cylinders A, B, C and "D . " comparing Eestoration from Compression of Specimens cut transversely 13 Inches from lower ends of Cylinders A, B, C and D • " comparing Permanent Set from Compression of Specimens cut trans- versely 13 Inches from lower ends of Cylinders A, B, C and D '• showing the General Properties- of the Iron in Cylinders A, B, C, and » " of Repetitions of Strain Water Test and Tangential Eesistance XLIII. Fabrication of 15-inch Gux . Casting . Cooling . Cooling Table Temperature of Pit Rate, Extent, and Effects of Internal Cooling Mechanical Tests XLIV. Or the Rate op Api'UCAtion or Eoece PAGE 245 246 247 24S 249 250 251 252 253 254 255 256 257 258 259 260 261 262 262 263 203 264 235 260 266 267 268 Xll CONTENTS. XLV. Of the Difference in Effect due to Difference in the Times of Action of a given Force ....... XLVI. Experiments with Powder of variable Diameter of Grain . Tables of Eesults ...... Consolidated Table of Eesults ..... XLVII. XLVIII. XLIX. Trial of 10-inch Guns Nos. 362 and 363. Eeport of the Inspection, Transportation. 15 -inch Gun Inspection [Continued from page 126.] Mounting and Trial of the Transportation Mounting Trial Table of Eesults Detail of Board for Trial of 15 -inch Gun Table of Eesults of Firing for Accuracy m Table of Eanges, &c, at 10° Elevation Trials for Bicochet on Water Loading and Manoeuvring . ' Opinion and Eecommendation of Board Table showing Initial Velocities of Shells, and Maximum Pressure of Gas due to different Charges of .6 in. Grain Powder Table of Eanges of Shells fired from the 15-inch Gun at different Elevations Explanatory Eemarks Of the Perforated Cake Cartridge Manner of forming the Cakes L. Projectiles for Guns of very large Calibre . LI. Of the Internal Pressure Gauge ..... Table showing the Eelation between the Pressures and Corresponding Lengths of Indentations in Annealed Copper from 100 to 9000 lbs. LII. Standard Qualities of Iron for Cannon .... Letter to Colonel of Ordnance explaining Method of determining a Standard of Qualities for Gun Iron ...... LIII. Smelting of Iron for Cannon ..... LIV. A 20-inch Gun .... 270 272 273 274 277 281 281 282 282 283 283, 284 285 286 287 287 288 289 290 290 291 291 294 297 299 300 301 301 304 307 REPORT OF EXPERIMENTS MADE FOR THE PURPOSE OF FURTHER TESTING THE RELATIVE MERITS CAPTAIN EODMAN'S MODE OF COOLING CAMON, AS COMPARED WITH THE ORDINARY METHOD. EEPOET EXPEEIMENTS MADE FOE THE PUEPOSE OF FUPvTHEE TESTING THE EELATIVE MEEITS OF CAPT. EODMAN'S MODE OF COOLING CANNON, AS COMPARED WITH THE OEDINAEY METHOD. In undertaking these experiments, the first object to be attained was to procure iron of a suitable quality for casting into guns of the size of those to be made. With a view to the attainment of this object, the following preliminary trials were made : — PEELIMINAET TEIALS. July 18th, 1856, three tons of No. 2 Greenwood pigs were melted in an air furnace with bituminous coal, and run into 5-inch square pigs, dry sand moulds. July 19th, forty-five hundred of the above 2d fusion pigs, and fifteen hundred of No. 2 Springfield pigs, were melted together, and a cylinder, 20 inches diameter, and 24 inches high, was cast from it in a dry sand mould. Two specimens were taken from the lower end of this cylinder, one 3 inches, and the other 7 inches from the axis. The first gave Density =7.273, and Tenacity = 42.884. The second gave Density = 7.272, and Tenacity = 38.993. These tests were of the highest order, both for density and tenacity ; but the metal appeared to be stubborn, harsh, and inclined to brittleness, and gave evidence of a strong tendency to separation into different compounds of carbon and iron, collections of free graphite being quite conspicuous on the fractured surface ; and, as it was known that this tendency would be increased, and its effects more fully developed, by the slower rate of cooling to which the iron would be subjected in the larger masses of the guns, it was rejected ; 4 PEOPEETIES OF METALS FOE CANNON, and, on the 12th of August, 3000 of 2d fusion No. 2 Greenwood pigs, and 1500 of No. 3 Greenwood pigs, were melted together, with 1500 2d fusion No. 1 Salisbury pigs. A portion of this heat was run into a cylindrical mould of dry sand, 20 inch diameter by 24 inch high, cut longitudinally into four parts, by thin cores, thus : — From this cylinder two specimens were taken ; one from the centre of the lower end of one of the quarters, axis parallel to that of the cylinder ; the other from the middle of the length of same quarter, equi-distant between axis and exterior of the cylinder, its axis being perpendicular to the face of the quarter from which it was taken, and parallel to a tangent to its circum- ference. The first of these specimens gave Density = 7.137; Tenacity = 33.268. The other gave Density = 7.159 ; Tenacity = 36. 373. These tenacities were considered quite satisfactory, though the densities were rather low ; but it was thought that this quality, as well as the tenacity, would be improved by bringing the iron up a little higher, and that the metal would not be rendered brittle by so doing, as there was scarcely any appearance of mottle in the fracture, and the iron appeared to be very tough. Accordingly, a sufficient quantity of No. 2 Greenwood pigs, and of No. 1 Salisbury pigs, were re-melted, and prepared for casting the guns, samples being taken from one of the 5-inch square pigs, into which each heat was run. Preliminary samples are all marked, 1st, A, and then by numbers and letters, to designate the number of heat, and other characters of the specimens. AND OF CANNON POWDEE QUALITIES. Table showing Mechanical Tests of Preliminary Castings. Marks. Description. Density. Tenacity. A. 1 B, A. 1 B. D, A.2B, A. 3 P, A. 2P, A. 1 c. 3, A, 1 c. 7, A. 3B, A. 4P, A.5P, A. 2C, c. A. 2 c, A. 6 P, A. 7P, A. 8 P, 13 c, From small bars, cast from No. 2 Greenwood pigs, . Duplicate of A. 1 B From small bars cast from 2d heat, 3 parts 2d fusion, No. 2 Greenwood, and 1 part No. 2 Springfield pigs, From 5-inch square pig, made from No. 2 Greenwood pigs, From 5-inch square pig, made from 3 parts 2d fusion Greenwood No. 2 pigs, and 1 part Springfield No. 2 pigs, From cylinder 20 inches diameter and 24 inches high, made from 3 parts 2d fusion Greenwood No. 2 pigs, and 1 part No. 2 Springfield pigs, taken from bottom, 3 inches from axis, From same cylinder as A. 1 c. 3, but 7 inches from axis, . From small bar made from No. 1 Salisbury pig, From 5-inch square pig, made from No. 1 Salisbury pig, . From 5-inch square pig, made from 2 parts 2d fusion Green- wood pig, 1 part No. 3 Greenwood pig, and 1 part 2d fusion No. 1 Salisbury pig From central part of largo cylinder, 20 inches diameter and 21 inches high, made from same iron as A. 5 P, axis of speci- men as perpendicular to that of cylinder, and parallel to tan- gent to its circumference, ....... From end of same cylinder A. 2 C, axis parallel to that of cylinder, . . . . " . From 5-inch square pig, made from 2 parts No. 2 Greenwood pig, and 1 part No. 1 Salisbury pig From 5 inch square pig, made from 3 parts No. 2 Greenwood pig, and 1 part No. 1 Salisbury pig From 5-inch square pig, made from 3 parts Greenwood No. 2 pig, and 1 part No. 1 Salisbury pig, From 10-inch cylinder, made from No. 1 Salisbury pig, axis parallel to that of the cylinder, ...... 7.184 7.198 7.307 7.099 7.304 7.273 7.272 7.219 7.210 7.172 7.159 7.137 7.106 7.100 7.109 7.191 33079 31334 35486 23776 31317 42884 38993 25372 22547 28518 36373 33268 22290 22179 22888 23873 August 21. The mould for the solid gun was made and placed in the drying oven, and the first coat of clay put upon the core barrel for the hollow gun. August 22. The mould for the hollow gun was made and placed in the drying oven, and that for the solid gun pitted, and the metal for casting the guns charged as follows, viz : — Furnace No. 1 received . Furnace No. 2 received . Furnace No. 3 received . 12000 lbs. 16000 " 18000 " Total charge for both guns, .... 46000 lbs. Seven-ninths of the whole charge was Greenwood iron, and two-ninths Salisbury iron. 6 PEOPEETIES OE METALS EOE CANNON, The Greenwood iron consisted of the following varieties, viz : — 5722 lbs. 2d fusion, No. 2 pigs. 5085 lbs. 2d fusion, No. 2 pigs, melted with No. 1 Salisbury, in the propor- tion of 2 parts Greenwood to 1 part of Salisbury. 12372 lbs. 2d fusion, No. 2 pigs, melted with No. 1 Salisbury pig, in the proportion of 3 Greenwood to 1 Salisbury. 4429 lbs. 3d fusion No. 2 pigs, melted with 1476 lbs. No. 1 Salisbury pigs. 8172 lbs. No. 3 Greenwood pigs. 35780 lbs. Total Greenwood iron. 2077 lbs. 2d fusion No. 1 Salisbury pig. 2543 lbs. 2d fusion No. 1 Salisbury pig, melted with No. 2 Greenwood pigs, in the proportion of 1 part of Salisbury to 2 parts Greenwood. 4124 lbs. 2d fusion No. 1 Salisbury, melted with No. 2 Greenwood, in the proportion of 1 of Salisbury to 3 of Greenwood. 1476 lbs. 3d fusion No. 1 Salisbury, melted with 2d fusion No. 2 Green- wood, in the proportion of 1 Salisbury to 3 of Greenwood. 10220 lbs. Total of Salisbury iron. 46000 lbs. Total charge. The Greenwood iron in these guns was made from magnetic ore, smelted in a furnace of eleven feet diameter of boshes, with charcoal and a warm blast, the temperature of the blast being kept regularly at about 300° Fahrenheit, and at a very low pressure, — say |- inch of Mercury. The Salisbury iron was made from brown haematite ore, smelted in a furnace of nine (9) feet diameter of boshes, and 30 feet high, with charcoal and warm blast ; temperature of blast 450° Fahrenheit ; pressure of blast not known. Blast enters the furnace at two tweyers, the average yield of the furnace being five tons per day. The different kinds of iron were distributed among the three furnaces in the same proportions, so that each one would contain the same composition of iron. The preliminary melting was all done in one of the furnaces in winch the metal for casting the guns was melted. These furnaces are all what are Scale of lee I J 1 1 1 1 I 1 I I L__i '8 " 72 '3 /• fS Veo""Srae J-LATE 1- ll ( I L I \ F \ N ~ ' 1 -1 ( I \ x / ( 1 AND OF CANXON POWDER QUALITIES. 7 termed air furnaces, having a sufficient height of chimney to create the requisite velocity of air through the fuel ; which in all the meltings, both preliminary and for the guns, was bituminous coal of the best quality. The furnaces, are so constructed that the metal, as it melts, flows towards the flame. The accompanying drawing (Plate 1,) shows a vertical longitudinal section of one of the furnaces, in which (a) is the ash-pit, (/) fuel chamber, (p) metal pool, (t) tap hole, (c) charging door, and (F) the flue. With a given charge of metal its treatment may be somewhat varied by the manner of dressing the bottom of the furnace, a broad, shallow pool exposing the iron more effectually to the action of the flame, and consequently making it hotter than when it is' collected in a deep pool, with a small surface exposed to the flame. The drawing represents the furnace as dressed for melting the iron from which the guns were cast ; the line (m e) showing the surface of the metal when melted. The mean depth of the metal was about seven inches ; that in No. 3 being something less, it being desirable that the iron in this furnace should be very hot, as it had a considerable distance to run — greater than that of either of the other furnaces. Furnaces Nos. 1 and 2 were charged up to their capacity. No. 3 could have received, perhaps, two tons more. The flasks in which these guns were cast are in two parts, which open longitudinally, one-half of the gun, and one trunnion, being moulded in each half of the flask. Their cros3 section, when clamped together ready to receive the metal is hexagonal, which gives an unequal thickness of sand around different parts of the same section of the gun, which, it is believed, is calculated to produce irregularity in the rate of cooling, of the different wedges or staves of which the gun may be supposed to consist. And, as perfect homogeneity, in each of the concentric cylinders of which the gun may be supposed to consist, is believed to be of the utmost importance to its endurance, it would seem that a flask of circular cross sections, which would give a uniform thickness of sand around every part of the same cross section of the gun, would be more suitable for casting cannon. The casting for the hollow gun was cylindrical from about the middle of the chase to the muzzle, while that for the solid gun corresponded more nearly in shape with the finished gun. 8 PEOPEKTIES OF METALS FOE CANNON, Casting. August 23, 1856. Two 10-inch Columbiads were cast from the metal charged on the 22d. One, No. 331, was cast hollow, and cooled from the interior by circulating water through the core barrel ; and the other, No. 332, was cast solid, and cooled from the exterior The furnaces were lighted at lOh. 50m. A. M., and Nos. 1 and 3 were tapped at 3h. 20m. P. M., the metal in all the furnaces having been 30 minutes in fusion. After the flow of metal from these furnaces had sufficiently decreased to admit of so doing, No. 2 was tapped, and the flow of metal kept nearly uniform during the whole time of casting. The metal from all the furnaces was received into one reservoir, from which it flowed in a single stream to a point equi-distant from the two gun moulds ; from this point it reached the mould for the solid gun in a single stream, and that for the hollow gun in two streams, one entering the mouth of the mould on each side of the core, the metal in both cases entering the mould directly, and not through " side runners." The hollow gun mould was filled in 8 \ minutes, and that for the solid gun in 12 minutes ; the great heat of the metal preventing the workmen from so regulating its flow as to fill both moulds simultaneously. Condition of Pits. Both pits were closely covered at the time of casting ; that in which the hollow gun was cast was a new pit with an earthen bottom, recently walled up, and in which no castings had yet been made. On this account this pit had fire kindled in it on the evening previous to the day of casting, and had been previously dried, to some extent, by fire. The pit in which the solid gun was cast was an old one, with a wrought iron bottom, and in which heavy castings had been recently made. The pits were both in good order for casting, and well adapted to slow cooling ; both being provided with means for keeping up a hot fire around the flasks, for as long a time as might be desired. Fire was lighted in the pit of the solid gun at 3h. 55m. P. M. Tempera- ture of hollow gun pit at the same time, 415°. AND OF CANNON POWDEE QUALITIES. 9 Cooling. Water circulated through the core barrel of the hollow gun at the rate of two cubic feet per minute ; entered at 71°, and left at 111°. Temperature of Pits, at the undermentioned numbers of hours after casting, in Fafirenheit degrees. Tempebatuhe. Hours after casting. 331 332 Remarks. Hollow. Solid. 0, 2, 4, 17, 25, 40, 48, 64, 88, 94, 115, 119, 207, 415° 500 530 550 550 1000 300 175 240 150 120 84 80° 300 350 530 600 600 900 350 200 300 150 190 144 Thermometer broke ; temperature judged of. n Jilerafioj? ofjireech *'•■'■>,,. c.» ■ :;-c/h v Zi>h AND OF CANNON POWDEE QUALITIES. 13 that they should be accurately finished, as the guns were intended to be broken. Proof. October 7th, 1856. Both guns were laid on the ground and proved in the ordinary mode. 1st fire, 20 pounds powder (proof range 314 yards), 1 shot (125 lbs.), 1 sabot, and 1 10-inch wad. 2d fire, 24 pounds powder (proof range 314 yards), 1 sabot and 1 shell, 99| pounds. The guns were then suspended, each in its own frame, for extreme proof, and fired alternately, with charges of 18 pounds of powder, and one solid shot and sabot; average weight of shot, 125 pounds. The guns were fired with friction tubes, the shot passing through a pen filled with earth, and lodging in the face of a vertical hill beyond, or more generally rolling down, and being found behind the pen ; were recovered and used again till broken, only 8 shots being used in the extreme proof of this pair of guns. Endurance. October 9th. The solid gun No. 332 burst at the 26th fire, including proof charges, into two main pieces. The plane of fracture was nearly horizontal, and about one inch below the axis of the gun, passing through the left trunnion. The lower and smaller piece broke off midway between the trunnions and neck, a crack extending forward in the larger piece to the middle of the neck ring. Several longitudinal cracks were found in the chamber and taper, but none in the cylinder of the bore. Figures 1, 2 and 3, Plate No. 2, show the hues of fracture very accurately. The firing was continued with the hollow cast gun, the same charges of powder and shot being used up to the 103d fire, including proof charges, when five rounds were fired with the same charge of powder, but a shell of 101.75 pounds, instead of a shot, for the purpose of determining the initial velocity of shells from this calibre of gun. One shell broke in the gun. The firing was then continued with the same charges as before, with solid shot, up to the 213th fire, when a crack was observed, beginning about midway of the taper, and extending into the chamber. After the 270th fire another crack was discovered on the opposite side of the chamber, beginning 14 PKOPEKTIES OF METALS EOE CANNON, nearer the junction of the taper and chamber than the other, and extending to near the bottom of the chamber. The shot recovered after the 214th fire was found to be cracked, and marked on opposite sides by a smooth, bright band, indicating that it had rubbed hard against the bore of the gun. This gun broke at the 315th fire, including proof charges, into three main pieces, splitting through the breech in a nearly horizontal plane ; the lines of rupture running forward, above the right and below the left trunnion, giving a slightly warped surface of fracture, to a point about 12 inches in rear of the neck ring, where the breech pieces broke off from the muzzle piece, leaving it hanging in the suspenders. One of the breech pieces was thrown to a very considerable height, and fell but a short distance from the intended bomb and fragment proof chamber ; and it is believed to be lucky for those in it at the time, that it did not fall on it. This remark is intended for the benefit of those who may hereafter be engaged in similar experiments. Figures 1, 2 and 3, Plate No. 3, give an accurate view of the lines of rupture of this gun. The annexed tables show the enlargement of the bores, chambers, and vents of these guns ; also the mechanical tests of the metal of which they were composed. Inspection after rupture of the interior of gun No. 331, showed great numbers of small cracks in the chamber, more resembling in appearance a piece of netting than anything else, with occasional larger longitudinal cracks, no cracks of any kind being discoverable in any part of the bore. Nor did the fractured surfaces exhibit any cavities or appearance of unsoundness in the metal, except the appearance of a slight draw or " soakage " in the chase ends of the fracture of the breech pieces from the muzzle ; these were at the same distance from the surface of the bore as the cavities before mentioned, and doubtless are attributable to the same cause. From the nature of the fracture of both these guns, it was quite evident that they first gave way by splitting through the breech, the gas afterwards acting as a wedge to extend the fracture forward ; and this has been the case in every Columbiad I have ever seen broken, and indicates the breech as the weak point hi these guns. PLATE JH 10 Inch Columbia cl Hollow ^° 331. -Rff 1. Fig. 2. AND OF CANNON POWDEE QUALITIES. 15 Of the Powder. The powder used in these experiments was made by the Messrs. Dupont ; was rather fine grained, for cannon powder ; the proof range by the eprouvette varied between 295 and 318 yards; it was new, and is believed to have been a very quick, and therefore a very severe powder upon the gun. The velocity of the 10-inch solid shot of 125 pounds, and sabot 1.5 pounds, with 18 pounds of this powder, was 1308 feet per second by the gun pendulum, and that of the 10-inch shell of 101.75 pounds, and sabot of 1.5 pounds, with 18 pounds of the same powder, was 1427 feet per second by the same pendulum. This powder is unquestionably too quick for guns of large calibre ; nor is it believed that the increase in velocity and range at all compensates for the heavy strain to which the gun is subjected, by the use of the excessive charges now assigned to heavy guns. Table showing the enlargement of the bores above their original diameters, after the under- mentioned number of fires, in thousandth of an inch. 2d Eire. (Proof.) 17th Fire. 47 th Fire. 70th Fire. 103d Fire. 153d Fire. 213th Fire. 270th Fire Distance from 331 332 331 332 331 331 331 331 331 331 Muzzle. Hollow. Solid. Hollow. Solid. Hollow. Hollow. Hollow. Hollow. Hollow. Hollow. 10 in. .002 .001 .002 .001 .002 .003 .003 .003 .003 .003 20 2 1 2 2 2 2 2 2 2 30 1 1 1 1 1 1 1 2 3 40 1 1 1 1 1 1 1 3 3 50 1 1 1 1 1 1 1 2 3 60 2 2 2 2 2 2 4 7 70 1 1 1 1 1 2 3 4 80 2 1 2 2 2 2 2 2 4 81 1 1 1 1 1 1 1 1 2 4 82 2 3 2 3 2 2 2 2 3 6 83 1 2 1 2 1 1 1 1 3 5 84 1 2 1 2 1 1 1 1 3 5 85 2 2 3 2 3 4 4 4 5 6 86 3 4 3 4 3 4 4 4 5 7 87 4 4 4 4 4 4 4 5 7 7 88 4 7 4 4 4 5 5 5 7 9 89 3 7 5 15 5 5 6 6 9 12 90 6 12 13 39 13 13 13 13 18 19 91 8 19 15 72 16 16 17 21 36 35 92 10 20 19 52 26 26 29 33 48 60 92i 12 17 23 52 26 28 39 44 61 66 J. Diam eter in plane of ax is, and interior of vent. 35 55 58 75 80 16 PEOPEETIES OP METALS EOE CANNON, Table showing the enlargement of the chamber of Gun No. 331 {east hollow), after the under- mentioned numbers of fires, in thousandths of an inch. Distance from Muzzle. 2d. 17th. 47th. 70th. 103d. 153d. 213th. 270th. H. 99J 3 5 7 8 8 8 14 18 10 103 2 5 6 6 6 6 7 11 8 106* 4 7 7 7 7 7 7 9 8 Table showing the enlargement of the chamber of Gun No. 332 {cast solid), after the under- mentioned numbers of fires, in thousandths of an inch. Distance. 2d. 17th. 991 9 14 103 5 9 106J 4 9 In both guns the greatest enlargement was in a plane containing the axis of the bore, and the interior of the vent ; it has always, heretofore, been in the vertical plane through the vent. These facts seem to show that the position of the interior of the vent exercises an influence over the plane of greatest enlargement ; and, if so, might not the enlargement caused by the indendation of the shot be prevented by igniting the charge in the axis of the bore at the bottom of the chamber ? Elongation per inch of metal in bore of hollow gun after 270th fire = .0066 in. " " " " solid " " 17th fire =.0072 in. chamber of hollow " " 270th fire = .00225 in. " " chamber of solid " " 17th fire = .00175 in. AND OF CANNON POWDEE QUALITIES. 17 Table shoiving the enlargement of the vent of Gun No. 331 (cast hollow), after the undermen- tioned numbers of fires, in hundredths of an inch. Distance from Exterior. 2d. 17th. 47th. 70th. 103d. 153d. 213th. 270th. 2 2 4 5 10 3 2 4 6 11 4 2 4 8 12 5 2 2 5 10 15 6 2 3 5 7 15 7 2 3 6 12 15 8 2 2 4 7 12 18 9 2 2 4 10 16 28 10 2 2 4 7 14 20 11 2 2 4 10 21 30 12 2 3 5 8 20 22 13 8 Though the instrument showed no enlargement of the interior of the vent till after the 270th fire, yet it was evident, from inspection after the gun was broken, that enlargements had been effected at this point at a much earlier period ; but it was principally in a longitudinal plane, the appearance after fracture being thus : — The vent of gun No. 332 (cast solid) was not perceptibly enlarged, as shown by inspection after the gun was broken. 18 PEOPEETIES OF METALS EOE CANNON, Table showing the tensile strength and density of specimens from various parts of the two guns, there being two specimens for each number, one from each gun, and from corresponding positions. Densities. Tenacities. Marks and Numbers G. 331 G. 332 G.331 G.332 of Specimens. Hollow. Solid. Hollow. Solid. 1,. 7.205 7.154 30062 29179 2,. 7.205 7.159 31831 28389 3,. 7.191 7.150 28740 30770 4,. 7.194 7.143 29311 24846 5,. 7.205 7.147 34510 28383 6,. 7.225 7.127 27850 28306 7,. 7.235 7.178 30070 31369 8,. 7.233 7.179 30941 31227 9,. 7.212 7.145 32720 28614 10,. 7.225 7.189 32292 30947 11,. 7.223 7.134 31050 27192 12,. 7.216 7.192 32647 30747 13,. 7.218 7.193 30733 30770 H.I.. 7.221 7.155 33712 31831 H.2,. 7.217 7.166 33620 31389 H. 3,. 7.217 7.166 33590 32811 M. 1,. 7.217 7.156 29720 29187 M.2, . 7.200 7.151 30290 29797 11. 3,. 7.220 7.154 31671 29869 Mean, 7.214.}f 7.159-Jf 31.334|$ 29.769|f Figures 1, 2, 3 and 4, Plate No. 4, show the parts of the guns from which the specimens were taken, the numbers on the plate corresponding with those in the table. Specimens 6, 7, and 8 of gun No. 331, are of greater density and less tenacity than the general average ; the cause of this is believed to be that the metal in this part of the gun was subjected to greater pressure of gas, and to the balloting of the shot ; the density being thus increased, while the tenacity was diminished. plate rr Dlitzz7r of (run Section, on ^ L/j. Fir/ 4 £~ - , i CH Crosiy I.i : AND OF GANNON POWDEK QUALITIES. Comparison of Head with Muzzle Specimens, from loth guns. 19 Guk No. 331 (east hollow) . NO. OF SPECIMENS. Densities. Tenacities. Head. Muzzle. Head. Muzzle. 1 2 3 7.221 7.217 7.217 7.217 7.200 7.220 33712 33620 33590 29720 30290 31671 Mean Difference, . . 7.218 .006 7.212 33641 3081 30560 . Gum No. 33! ! (cast solid). 7.155 7.166 7.166 7.151 7.151 7.154 31831 31389 32811 29187 29797 29869 Mean Difference, . . 7.162 .010 7.152 32010 2392 29618 The above results show a difference in both density and tenacity between the specimens taken from the gun heads, and those taken from the muzzles of the guns ; these specimens were only separated in the casting by the width of the necking tool, say .75 of an inch. The discrepancies cannot, therefore, be ascribed to difference of position in the casting, but are believed to be due to deterioration of the qualities of the metal, caused by the violent series of vibrations to which the gun is subjected at each discharge in firing. And this view seems to be sustained by the fact that the difference in tenacity is greafer in the gun which has been oftenest fired. 20 PEOPEETIES OF METALS FOE CANNON, Comparison of tensile strength of radial with tangential specimens from the same gun. Gun No. 331 (cast hollow) . Gun No. 332 (cast solid). Number of Specimens. Radial. Tangential. Number of Specimens. Radial. Tangential. 5 and 1, 8 and 7, 12 and 10, 13 and 11, 34510 30941 32647 30733 30062 30070 32292 31050 5 and 1, 8 and 7, 12 and 10, • 13 and 11, Mean Difference, 28383 31227 30747 30770 29179 31369 30947 27192 Mean Difference, 32208 1340 30868 30282 611 29671 These results show the radial specimens to be something stronger than the tangential, from the same cross section of the gun, and are in confirmation of the theory of Mr. Eobert Mallet, viz., that the major crystalline axes correspond in direction with that of the passage of heat in cooling ; or, that they are perpendicular to the cooling surface, and that the metal is strongest in that direction. These results further sustain this theory, in the fact that the difference between the strength of the radial and tangential specimens is greatest in those from the hollow cast gun, which was most rapidly cooled. Table showing the compression per inch, in length, caused by tlie undermentioned weights per square inch (radial specimens). Gun No. 331 (cast hollow). > Gun No. 332 (cast solid). Number of Specimens. Weight per square inch. Compression per inch. Number of Specimens. Weight per square inch.. Compression per inch. 1 2 3 4 5 6 72571 71730 72290 72855 73140 73140 .03537 .03560 .03375 .03281 .02299 .02617 1 2 3 4 5 6 73428 73140 72855 73140 72855 73140 .07017 .06825 .06406 .07480 .09422 .10080 The specimens from which the results recorded in the above table were obtained, were taken from radial specimens opposite the middle of the chamber. AND OF CANNON POWDEE QUALITIES. 21 No. 1 being cut from the end next the chamber, and the others numbered from the inner to the outer end of the specimen, each number having two specimens, one from each gun. These results show from two to four times as much compression in the specimens from the solid gun, as in those from the hollow one ; and this property of resistance to a crushing force will be shown further forward in this Report, to exercise an important influence on the endurance of guns. Table showing the change of form produced by a crushing force of 1 5.000 pounds on the undermentioned specimens, and their ultimate resistance to a crushing force. {Same speci- mens as described in the preceding Table.) GUN No. 331 [cast hollow). Diameters. Lengths. Number of Ultimate Specimens. Before. After. Before. After. Crushing Force. 1 .513 .525 1.244 1.200 25400 2 .516 .525 1.264 1.219 23700 3 .514 .523 1.274 1.231 22900 4 .512 .522 1.280 1.238 22200 5 .511 .518 1.261 1.232 24400 6 .511 .519 1.261 1.228 20900 GUN No. 332 (cast solid). 1 .510 .533 1.254 1.166 20700 2 .511 .533 1.260 1.174 19800 3 .512 .534 1.280 1.198 19600 4 .511 .535 1.270 1.175 19900 5 .512 .546 1.263 1.144 19800 6 .511 .544 1.259 1.132 18600 It is believed that specimens Nos. 1 and 2, from gun No. 332, should have been Nos. 6 and 5, and vice versa, as it is thought that the workman misunder- stood the directions given for marking, after turning. The results indicate that No. 6 should be from the interior instead of the exterior of the gun, and a comparison of the ultimate resistance of No. 6 gun 331, with that of No. 1 gun 332, adds strength to this belief, as there should be less difference in quality between the outer portions of the two guns, than in any other corresponding parts ; the circumstances of cooling being more nearly identical in these than in any other portions of the guns. 22 PEOPEETIES OE METALS FOE CANNON, Table showing the compression, restoration and set, of radial specimens, one from each gun, and from corresponding parts (inner ends of radii). GUN No. 331 (cast hollow). GUN No. 332 (oast solid). Original Diameter = .688 in. Original Diameter = .685 n. Original Length = 5.52 in. Original Length = 5.52 in. Weights. Compression. Restoration. Set. Compression. Restoration. Set. 1000 lbs. .0005 .0005 .0000 .0020 .0020 .0000 2000 .0011 .0011 .0000 .0025 .0025 .0000 3000 .0020 ^0020 .0000 .0030 .0030 .0000 4000 .0025 .0024 .0001 .0035 .0035 .0000 5000 .0033 .0031 .0002 .0040 .0039 .0001 6000 .0047 .0044 .0003 .0044 .0042 .0002 7000 .0058 .0054 .0004 .0060 .0057 .0003 8000 .0065 .0058 .0007 .0080 .0075 .0005 9000 .0076 .0066 .0010 .0085 .0077 .0008 10000 .0090 .0078 .0012 .0095 .0082 .0013 11000 .0105 .0091 .0014 .0110 .0090 .0020 12000 .0115 .0097 .0018 .0130 .0100 .0030 13000 .0133 .0103 .0030 .0143 .0098 .0045 14000 .0143 .0105 .0038 .0190 .0080 .0110 15000 .0168 .0118 .0050 .0230 .0101 .0129 16000 .0217 .0127 .0090 Began to bend. 17000 .0267 .0137 .0130 18000 .0333 .0133 .0200 19000 .0523 .0183 .0340 20000 .0623 .0163 .0460 Began to bend. Diameter after 20000 lbs. = .691. Diameter after 15000 lbs. = .686. AND OP CANNON POWDER QUALITIES. 23 Table showing the compression, restoration and set, of radial specimens, one from each gun, and from corresponding parts {outer ends of radii). GUN No. 331 (oast hollow). GUN No. 332 (cast solid). Original Diameter =.692 ia. Original Diameter = .682 in. Original Lengte [=5.52 in. Original Length = 5.5 in. Weights. Compression. Restoration. Set. Compression. Restoration. Set. 1000 lbs. .0007 .0007 .0000 .0018 .0018 .0000 2000 .0016 .0016 .0000 .0028 .0027 .0001 3000 .0024 .0024 .0000 .0039 .0032 .0007 4000 .0033 .0033 .0000 .0049 .0041 .0008 5000 .0040 .0040 .0000 .0060 .0051 .0009 6000 .0048 .0048 .0000 .0070 .0061 .0009 7000 .0054 .0053 .0001 .0084 .0073 .0011 8000 .0064 .0060 .0004 .0097 .0082 .0015 9000 .0074 .0066 .0008 .0122 .0101 .0021 10000 .0085 .0074 .0011 Not reliable.* 11000 .0097 .0084 .0013 .0212 .0104 .0108 12000 .0108 .0086 .0022 .0220 .0110 .0110 13000 .0130 .0090 .0030 .0236 .0113 .0123 14000 .0153 .0113 .0040 .0300 .0110 .0190 15000 .0188 .0129 .0059 .0460 .0135 .0315 16000 .0224 .0143 .0081 Specimen began to bend it 13000 17000 .0277 .0127 .0150 lbs. ; only reliable above 10000 lbs. 18000 .0388 .0128 .0260 19000 .0551 .0150 .0401 20000 .0652 .0157 .0495 Began to bend a t 20000 lbs. Length after 20000 lbs. = = 5.470 in. Length after 15000 lbs. =5.486 in. Diameter after 20000 lbs = .693 in. Diameter after 15000 lbs. = 683 in. Specimen shifted while measuring set. These specimens were turned with collars or projections left near each end, and the lengths given in the table are. the distances between the collars. The space between the collars was surrounded by a cast iron sheath about .75 inch less in length than this space, and the compression was measured by inserting a graduated wedge between the upper end of the sheath and the lower side of the upper collar, before any pressure was applied, and again after the application of the force whose effects were to be measured ; the difference of these readings gave the amount of compression which that portion of the specimen between the collars had undergone, there being no pressure upon the sheath at any time. The taper of the wedge was .01 inch to 1 inch. 24 PEOPEKTIES OE METALS EOK CANNON, Tabus showing the compression and set of a cylinder 5.5 in. long and .508 in. diameter, cut longitudinally from trial cylinder {A. 1. C.) Weight. Compression. Set. Weight per square inch. Compression per inch. Set per inch. 100 lbs. .0000 in. .0000 in. 493 lbs. .00000 in. .00000 in. 200 .0005 .0000 986 .00009 .00000 500 .0010 .0000 2465 .00018 .00000 1000 .0015 .0000 4930 .00027 .00000 1500 .0020 .0000 7395 .00036 .00000 2000 .0025 .0000 9860 .00045 .00000 3000 .0040 .0005 14790 .00072 .00009 4000 .0055 .0007 19720 .00100 .00013 5000 .0065 .0010 24650 .00118 .00018 6000 .0085 .0012 29570 .00155 .00022 7000 .0098 .0018 34510 .00178 .00033 8000 .0124 .0030 39440 .00225 .00054 9000 .0160 .0055 44370 .00291 .00100 10000 .0197 .0098 49300 .00358 .00174 11000 .0384 .0250 54230 .00698 .00454 12000 .0550 .0493 59160 .01000 .00896 Diameter after 10.000 pounds = .512. Diameter after rupture (12.000 lbs.,) = .513. Broke by bending. Specimens began to bend at 8000 pounds ; therefore results from weights above 8000 pounds, are not strictly reliable. A fragment of this specimen 1.25 in. long and .510 in. diameter, After having borne 14000 lbs., was 1.235 " " " .515 " " " " " 20000 " " 1.180 " " " .526 " " « " " 25000 " " 1.074 " " " .558 " " Table showing the elongation per inch, at the beginning of permanent set, and at the moment of rupture, in parts of an inch. MARKS. A. 1. C. 1 Q. 331. 1 G. 331. 2 a. 332. 1 G. 332. 2 Elongation at be- ginning of set. .002186 .002328 .002488 .002060 .001983 Elongation at rupture. .007922 .010057") .013300) .010580 [ ■0097 10 J Remabks. Longitudinal specimen from trial cylinder. Kadial specimens contiguous to those for . . pression. Eadial specimens contiguous to those for com pression. com- The elongation at permanent set, and that due to different forces, up to that of rupture, being qualities next in importance to the actual strength of the metal, it is much to be regretted that the testing machine at this place is not adapted to the accurate determination of these qualities ; and the AND OF CANNON POWDEE QUALITIES. 25 construction of a machine adapted to this kind of test, in addition to those to which the present machine is adapted, is earnestly recommended. The values for elongation at set and rupture, recorded in the above table, were obtained by bending a regularly tapered sample around segments of circles, whose radii were known. For the elongation at the beginning of permanent set, the specimen is bent, beginning at the thin end around the arc. It is then removed, and the convex side of the thin end placed in contact with a straight edge, and the thickness of the specimen at the point where it begins to leave the straight edge is measured, and a piece of the same metal, of this or any less thickness, could be bent entirely round a circle of the radius of the arc, round which the specimen was bent, without receiving a permanent set. Then if we suppose the neutral axis of the specimen to be intermediate between the two surfaces, half the thickness of the specimen divided by the radius of the arc, plus the entire thickness of the specimen, will give the elongation at permanent set. The same principle applies in determining the elongation at the moment of rupture. The specimen being bent around an arc of known radius, until it breaks, then, on the same supposition as to the neutral axis, half the thick- ness of the specimen at the point of rupture, divided by the radius of the arc, plus the entire thickness, will give the elongation at the moment of rupture. Owing to uncertainty as to the position of the neutral axis, and to the fact that the very thin lamina on the exterior of the specimen whose elongation is measured, is connected with others less highly strained, this method is not deemed as reliable for the actual elongation as that in which the entire section of the specimen is subjected to the same tensile strain ; it, however, answers very well as a means of comparing one metal with another. Thus an inspection of the foregoing table shows the stiff, strong specimen from (A. 1 C.) to have extended more before taking a permanent set, than the weaker and softer specimen from gun No. 332, while the latter undergoes a greater extension before rupture ; the medium iron in gun No. 331, being superior to both, in both qualities. And a comparison of the results obtained from the two guns in all the tests to which the metal has been subjected, goes to show the superiority of that in gun No. 331, cast hollow, over that in gun No. 332, cast solid. 26 PEOPEETIES OF METALS FOE CANNON, And this is what should have been expected, since the metal was sufficiently low to require, in order to develop its best qualities, a more rapid rate of cooling than was applied to the solid gun No. 332, or than can be, with safety, applied to any solid cast gun of this calibre. Summary of all the tests to which the metal of these guns was subjected. KIND OF TEST. Gun No. 331 (cast hollow) . Gun No. 332 (cast solid) . Specific gravity, Weight of guns, ...... Tensile strength Compression per inch by 73093 lbs. per square inch, Ultimate resistance per square inch to a crushing force, ........ Elongation per inch at beginning of permanent set, Elongation per inch at moment of rupture, . Number of times fired, 7.215 15218 lbs. 31335 lbs. persq. in. .03131 in. 112480 lbs. .002408 in. .011678 in. 315 burst. 7.160 15139 lbs. 29770 lbs. per sq. in. .07871 in. 96219 lbs. .002021 in. .010145 in. 26 burst. The values for specific gravity and tensile strength, are the mean results of 19 specimens each. The values for compression per inch, and for ultimate resistance to a crushing force, are each means of six specimens. Those for elongation at beginning of permanent set, and at rupture, are means of two specimens each. AND OF CANNON POWDEE QUALITIES. 27 ^ s bo "53 S « "ft" ■Is s I! S S co ss S « o „ Ci S" s 3 "K> mj V) ^ s •>^ a -fo Wi 3 ~*-~ t>n s s i-o i~ a •r- *« « s? w. -*J -4^ -f3 C5 c3 c3 C3 c3 _ c3 c4 S *» -» -t» ^= -*3 -*J ■p h+> -4^ -^> cq do 5° «* 00 CQ ta $ a 0Q 0Q u u, h EJ h M hrg h Fh t-> 3 3 3 =5 3 3 3 =1 3 3 fp W fQFQ m Fp Pdcq 01 "* O cq o en co O o o >o ■* US >ri oo "O t- o r-< •* o cq o ■-H cq (M >o cq o o CO o i— i l-H f— 1 © fc tj o .S 3 .2 o ■2§ 2 -° ri ft S.2 is o a |5 l-H F^ 'ft s l-H S3 O ,* (0 ^ - ■+» f— t . fa0 -* ■# •* ■* ic »o cq cq cq cq >o US CO CD CO CO cq cq CO CO CO CO cq cq 1 t-( ^^ i— * i— i ° 03 to *_ M i— I cin C*>3 o o o o CO CO CO O CO O CO CO (S I— 1 I-H I— i l-H l-H 1-1 l-H i-H l— t r-H ^ •* Cq •^ CO CI CO b- O iO o ~- CO CO "-I cq i-i O Oi CO t- "o O cs> 02 CO i—i lO CO lO co t- c3 t- t- t- t- b- CO r^ CO i— < C3 cq cq CO CO CO CO CO CO co cq CH i-H -o CO CO o -* i-H r- »c o fc-> cq cq CO CO Oi Ci CO t- ^H co *«3 Cq Cq O O cq II 00 c3 1Q CO CO CO CO II 00 t— i t-i O II ci ^= ci ^ a 3 w a a 3 _3 o nn o U o o 00 o "o ^3 3 3 •73 ci "« rS -S 3 •ri i-H rJ <-* c« c3 a a a a -s a CJ ca « co a 3 .o -o ^2 r<= ^ to a a a a -+J 3 3 3 3 3 P 3 3 o o ^ -T3 CS rf £3 C3 o o o r-i H H H O C/J 28 PEOPEKTIES OE METALS EOE CANNON, These results appear to leave no doubt as to the superiority of the hollow over the solid cast guns, while new ; what effect time may have upon them can only be ascertained by experiments ; but it is difficult to understand how time could ever so far change their relative endurance, as to cause the solid cast guns to surpass those cast hollow in this quality. The tests show the metal in the solid cast gun to be inferior, in every quality, to that cast hollow ; and it is believed that this inferiority will become more marked as the iron from which the guns are made becomes softer ; and it is now conceded by all who are acquainted with the subject, that solid cast guns cannot, with safety, be made of high iron. They must, therefore, be made of soft, and consequently weak iron, whose qualities would be improved by a more rapid rate of cooling than can be safely applied to the solid cast gun. The only effect of time is supposed to be to relieve the metal from the strain to which it had been subjected in cooling, it not being supposed to effect any change in its actual character. It would, therefore, appear to be reasonable and safe to predict that the utmost effect which time could produce, would be to bring their endurance to an equality; and should this be found to be the case, which is not probable, it would still leave the hollow cast gun superior to the other, as it is believed that it may be mounted for service as soon as finished, and relied on for at least 150 rounds, for 10-inch guns; while the solid cast gun cannot be relied on for a single fire when new ; nor can any limits, below which it shall be safe, be with any degree of certainty assigned to it after any lapse of time. The cause of this difference is believed to be that the hollow cast gun is so far relieved from strain in cooling, as to remove the possibility of its ever being injured in that process ; and the more perfect this relief, the more nearly will the endurance of the new gun approach that of the same gun, after any lapse of time. While in the new solid cast gun, it is certain that the interior is under a very heavy strain, which requires time for its removal ; and there is no certainty that this strain has not been sufficiently great to produce either actual cracks, or such a degree of molecular separation in the interior portions of the gun, as no lapse of time will remedy. AND OF CANNON POWDEE QUALITIES. 29 That good solid cast guns may be made, is not doubted, for they have been made ; but that it is possible to distinguish, before the trial, the safe from the unsafe gun, is not believed. And the number of comparatively recent failures in the proof of guns offered for inspection by our most experienced gun founders, and of those by them rejected for cracks and other imperfections, discovered in boring, added to those which have stood the proof and been received, but burst by a few additional rounds, would seem to show that this incredulity is not without good foundation. To remove this uncertainty is the first great desideratum in the manufacture of cannon ; the next is to place the limit of safety as high as possible. Therefore the introduction of any mode of casting which will enable us to fix a limit, though it should be a low one, below which guns are actually safe, must be regarded as an important step forward in the art of gun founding. From the experiments already made, the mode of hollow casting, and internal cooling, gives fair promise of enabling us to fix a limit of safety which will be above the number of fires to which any one gun would likely be subjected, during any siege, or even during any war. Experiments alone will give us the facts, and settle the question ; and they should be made with as little delay as possible, and on such a scale as to go to the bottom of the subject ; and no additional heavy guns should be cast for service until further experiment shall point out the best mode of manufac- ture, and the best model of gun. Initial Vebcities of 10-inch Shot and Shells. The 10-inch gun being suspended, for convenience in firing, while testing its endurance, was converted into a gun pendulum, by accurately weighing all the material which moved in connection with the gun at each discharge, and attaching a graduated arc, with a movable slide, for measuring the arc of recoil. The formula by which the velocities were computed was that for the gun pendulum used by Major A. Mordecai, in his experiments on gunpowder, viz. : v = 2 sine \ Ap 8 V G o — e N ,in which .D 2 c 30 PEOPEETIES OF METALS FOE CANNON, v = initial velocity in feet per second. A = arc of recoil of the pendulum. p = weight of pendulum = 15823 pounds. g = distance of centre of gravity of pendulum from axis of motion = 289.33 inches. Q= force of gravity = 32.155 feet. o = distance between axis of suspension and oscillation = 294.83 inches. i= distance between axis of suspension and of the gun = 292.1 inches. 3 = weight of shot and sabot = 123.5 +1.5 = 125 pounds. c = weight of charge = 18 pounds. iV=1600 feet. D = diameter of bore = 10 inches. d= diameter of shot = 9.87 inches. The gun made 437 oscillations in 20 minutes. The arcs of recoil, with the same weights of powder and the same shot, were as follows, viz. : — 1st fire, 27° 08' 2d do 26° 56' 3d do 27° 37' 4 th do 27° 12' 5th do 27° 26' 6th do 27° 21' 7th do 27° 44' 8th do 27° 34' >■ Mean = 2 7° 22' These data give the velocity of the solid shot = 1308 feet per second. The arcs of recoil, with 18 pounds of powder and the same shell = 101.75 pounds, and sabot 1.5 pounds, and same pendulum, were as follows, viz. : 1st fire, . . 25°17' > 2d do . . 25° 05' 3d do . . 25° 27' 4th do . . . . 24° 48' 5th do . . 26° 07' , - Mean = 25° 21'. Diameter of shell = 9.87 inches, and iV= 1600 feet. These data give the velocity of the shell = 1427 feet per second. AND OF CANNON POWDEE QUALITIES. 31 OF THE VELOCIMETEE. Its Object. This instrument is intended to measure the successive increments of velo- city impressed, by the expansive force of the ignited charge, upon the shot, from the time it leaves its seat in the gun, till it reaches the muzzle. Principles on which its Action depends. It depends, for its action, upon the fundamental principles of mechanics, that no connected system of bodies has the power of changing the position of its centre of gravity ; and that the same force will impress, upon different bodies, equal quantities of motion in equal times, irrespective of their masses. Description. It consists simply of a cylinder, mounted in a frame, with the means of communicating to it a uniform motion of rotation about its axis ; and of a movable slide, which carries a marking point, and is so placed as to move parallel with the axis of the cylinder. Its position when in Use. When in position for use, the instrument is firmly fixed with the axis of the cylinder parallel to that of the gun, and at such a distance in its rear as to be entirely below the arc described in its recoil ; the movable slide being on a level with the stop by which it is moved, as the gun recoils. This stop is bolted to the under side of the bar which connects the breech and muzzle slings of the gun, is the lowest point of the system, and likewise moves the registering slide on the arc which measures the recoil of the pendulum. The space between this stop and the end of the sliding bar is supplied by a wooden rod, shod at both ends, with iron or brass. The forward end of this rod should be permitted to disengage itself from the stop as soon as the shot has left the gun, otherwise something would be broken. The rear end of the shding bar should also have a wooden cushion to strike against, otherwise the marking point would be broken off at each discharge. Explanation of Action. Let us suppose the pendulum to be at rest, its centre of gravity will be in the vertical plane of the axis of suspension. 32 PKOPEKTIES OE METALS EOE CANNON, Now suppose the cartridge so far inserted into the gun that its centre of gravity shall coincide with that of volume of the bore, and the shot so far inserted that its centre of gravity shall be in the plane of the face of the muzzle, these weights will cause the pendulum to move to the rear ; the centre of gravity of the new system, when at rest, being in the ver- tical plane of the axis of suspension. If now the sliding bar be, by the inter- position of the aforesaid rod, placed in contact with the stop, and the cylinder revolved, the marking point will trace a continuous line around the cylinder. Now suppose the cartridge and shot both rammed home ; they will now occupy a position in rear of the vertical plane of the axis of suspension, and will cause the pendulum to move forward until the centre of gravity of the system again rests in the vertical plane of the axis of suspension. If now the rod and slide be moved forward, in contact, till the end of the rod again comes in contact with the stop, and the cylinder be again revolved, the mark- ing point will trace another continuous line around the cylinder. The distance between the two lines thus traced on the cylinder is equal to that through which the stop moves, while the shot is passing from its seat to the muzzle of the gun. Let the cylinder be now put in motion, and after having attained a uniform velocity, suppose the charge ignited; the shot and charge will move for- ward, and the pendulum backward, while the centre of gravity of the system will remain at rest, so long as the shot remains in the gun, and at the moment when the shot leaves its seat, the marking point will leave the for- ward line around the cylinder, and will cross the rear line at the moment when the shot is leaving the muzzle, all the parts of the system occupying, at this moment, the same positions as those first described with the shot in the muzzle. The cylinder being in motion while the marking point is moving to the rear, will cause it to describe a helix on the surface of the cylinder, whose inclination to the elements of the cylinder will become less as the velocity of the marking point becomes greater ; and that portion of this helix which is traced after the velocity of recoil has become sensibly uniform, and before gravity has sensibly diminished the velocity of recoil, will develop into a right line ; and this portion ought to begin just in the rear of the rear line on the cylinder, since the gas remaining in the gun, at the moment when the shot leaves, ought, in escaping, to impress some additional velocity upon the pendulum ; otherwise it ought to commence on that line. AND OE CANNON POWDEE QUALITIES. 33 And, in practice, the curvature is not perceptible in rear of that line. This instrument is intended to be used in connection with either the gun or ballistic pendulum. Therefore the velocity of the shot, at the moment it leaves the gun, is known. This gives the value, in velocity, of the angle formed by the element of the cylinder, and the tangent to the curve at the point where it crosses the rear line traced by the marking point upon the cylinder ; and that portion of the curve between the lines traced upon the cylinder connects this value with that of every other angle which can be formed by the elements of the cylin- der and tangents to this curve. The cylinders used in these experiments were of birch wood, the marking point being of steel, burnished so as to make a fine mark on the surface of the cylinder. The curves described on the cylinders were retraced by a fine- pointed lead pencil, and the curves thus colored were taken on tissue paper, and transferred to that on which they now are. This mode of determining the velocity of the shot, at different points, along the bore of the gun, is not supposed to give that degree of accuracy to its results which would be obtained by successively diminishing the length of the bore, so as to make its entire length equal to the distance from the bottom of the bore to the point at which the velocity is required, and using the gun thus shortened in connection with the use of the ballistic pendulum. This method is intended rather to give an ocular demonstration, or a view, of the relative velocities of the shot at different points along the bore of the gun. To explain more fully how the velocity of the shot at the muzzle of the gun is connected by the curve with that at any other point, let (a b) and (c d) Fig. 2, Plate 5, be the developed lines traced around the cylinder, and let (e p g) be the curve developed ; traced by the marking point when moving to the rear, as before described, the point (e) corresponds to the moment of time when the shot left its seat, and (p) to that when it left the muzzle ; and (I p) represents the time occupied by the shot in traversing the entire bore of the gun. Let the tangent to the curve at the point (p) be extended till it meets the line {a b), and from the point of intersection let the line (n g) be drawn par- allel to the developed elements of the cylinder, the line (p g) will represent 34 PEOPEETIES OF METALS FOE CANNON, the time required by the shot to traverse the length of the bore when moving with the velocity with which it left the muzzle. Let it now be required to find the distance from the bottom of the bore at which the velocity of the shot was one-half of that with which it left the muzzle. Now since the space passed over = velocity multiplied by the time, if the velocity be one-half, the time will be doubled for the same distance ; if, therefore, ( p m) = (p g) be laid off on the line (c d), and the points (m and n) joined by a right line, and this line be moved parallel to itself till it becomes tangent to the curve, the point of contact will correspond to that in the gun at which the velocity of the shot was one-half of that at the muzzle ; and if a = length of bore, and x = distance required, then n tin gnxia. In like manner the value of x -for any assumed velocity may be found. The dotted curve (parabola) (eT P'), Fig. 1, is that which would he described if the shot were moved by a force of uniform intensity, and such that it would leave the gun with a velocity equal to that produced by the irregular force ; and indicates that the pressure exerted by the expansive force of ignited, grained powder, is very much greater in the rear than in the middle and forward parts of the bore. And these curves afford the data for approximating with considerable accu- racy to the statical pressure per inch exerted on the bore of the gun at differ- ent points. The centre of gravity of the pendulum (whose weight = 15.823 pounds), is found to have been elevated by the force of the powder, through a vertical height of 2.7 feet; and the distance through which it moved while acquiring the velocity due to this height is found to be (.06) six hundredths of a foot; then supposing the pressure per inch to have been of uniform intensity, while the pendulum was moving through this distance, we have for its value (p). jp = 15823 x 2.7 = 9066 pounds. .06 x 78.54, (the area of a section of the bore) == 78.54. The curves described show that the gun and shot had acquired one-half their final velocity in about one-fourth of the time required for the shot to pass from its seat to the muzzle of the gun ; therefore the mean pressure, in the bore of the gun, during the first fourth of that time, must have been double that for the whole time, or =18132 lbs. per square inch; they further show that the shot and pendulum had acquired one-fourth of their final velo- city in about one-sixteenth part of the whole time aforesaid, and that the J'LATE I'. ttcj.2 b J/ .r.sivZ-r* TL.VTEYT- CM.CrosbyZi.th PLATE TH. M ff .J tt ff 2. C.H. Crosby Zith. I'l.ATK I7Z7 -Ftff-1 CJ/.CtvOt Z.i'7- AND OF CANNON POWDEE QUALITIES. 35 mean pressure during the first sixteenth part of that time was =36264 pounds. And the pressure must be still greater during the lower rates of velocity, amounting to, probably, 50000 pounds per inch ; and this estimate is for statical pressure, the strain due to which, as will be shown further forward in this Report, must be considerably less than the actual strain, the rate of applica- tion of the force effecting the strain to which it subjects the resisting body, so far as even to double it, in the extreme case, or when the application of the force becomes instantaneous. And it is believed that velocities equal to, if not greater than those now attained, might be secured, and the gun relieved from a large proportion of the strain to which it is now subjected, by the use of charges which would produce a more nearly uniform pressure upon the shot, during its passage through the bore of the gun. The curves shown in Plates (5 and 6) were described while firing shot, and those on Plates (7 and 8) while firing shells. Weight of shot and sabot, 125 lbs; weight of shell and sabot, 102.25 lbs. Weight of charge in both cases, 18 lbs. I am indebted for the transfer of the curves from the cylinders on which they were traced, to the sheets on which they now are, and for the drawings showing the lines of fracture of the guns, to Lieut. S. Crispin, Ordnance Department. BUESTING GUN HEADS. With a view to subject the metal of a gun to a mechanical test, as nearly similar in its action to that of gunpowder as possible, the head of gun No. 331 (cast hollow) was cut off to two feet in length, and reamed out to 10 in. diameter, with packing recesses of 2 in. wide by 1 in. deep. The distance from out to out of these rings was 14.5 in., there being 4.5 in. from the end next the gun to the edge of one recess, and 5 in. from the sprue end to the outer edge of the other recess. These recesses are shown at (r) in the accompanying drawing (Plate 9). A solid cylinder, with an opening or mortice (m) through it, of the same length as the head, and reduced 1.5 in. in diameter for a space of 10 inches along the middle of its length, was accurately turned, and inserted into the head. 36 PEOPEETIES OF METALS EOE CANNON, This cylinder was also bored through its axis, and furnished with packing glands (g), one at each end, through which two pistons entered the cavity (m) ; one (p), called the penetrating piston, receiving the falling weight, and was thus driven into the cavity (m) ; the other (p'), called the indenting pis- ton, had the indenting tool fitted into its outer end, and was forced out by the pressure produced by the entrance of the penetrating piston. The packing recesses, and the glands around the piston, were supplied with sole leather packing rings. The head, with the solid cylinder and indenting piston inserted, was placed in a cast iron frame, of which the foundation (F\ was a section, 18 inches long, of a 10-inch gun; and the bed piece (B) had a shank of the same length turned and fitted into (F), with an opening (0), to receive the specimen to be indented; also lateral arms, to which the upper portions of the frame were attached. The frame around the head had cast iron upright slides bolted on to it, to guide the drop («) in its descent upon the piston (p). The inner edges, and part of the sides of the slides, were planed, and recesses were planed in ledges, cast on to the sides of the drop, to fit and move freely on the planed parts of the slides, which were accurately adjusted in a vertical position. The spaces (m) and (n) communicated with each other, and were, when the head was to be acted upon by the drop, filled with water. In order to force the packing rings into their proper bearings, so as to pre- vent leakage, it was found to be necessary to attach a force pump to the head; this was done by drilling a small hole through the head, entering the interior at the upper edge of the lower packing recess, and attaching a wrought iron tube by means of a gland ; the tube having a small valve adapted to the end that entered the head, opening inward, and being closed by pressure produced by the fall of the weight, thus relieving the pump attached to the other end of the tube from the aforesaid pressure, and pre- venting a reflux of water. Everything being in position, the penetrating piston was removed, to allow the air a better chance to escape, and the pumps put in motion, and worked till the water reached the upper gland ; the piston was then inserted, and the drop allowed to rest upon it. This weight at first caused the piston to pene- trate, the water leaking past the packing rings. But by working the pump the piston and drop were forced up, and a few falls of the drop were sufficient to force the packing rings to their proper TLATEIX _A. i*:|A _..-&_ A,A ^ f IK /£ ^1 ^ * 1' V K f^t Tn~ ^ AND OF CANNON POWDER QUALITIES. 37 bearings, thus stopping all leaks. The first penetrating piston used, was one inch area of cross section, and the weight of the drop was 272 pounds. Everything being ready, and in perfect working order, the weight was raised to a height of one foot, and let fall upon the piston, causing a penetra- tion of 2.54 in., the weight rebounding as if it had fallen on an elastic spring. Repeated trials gave the same result. The weight was then raised to a height of two feet, and let fall, producing a penetration of 3.85 in., and rebounding higher than before. The weight was then raised to a height of four feet, and let fall, when the rebound threw the piston entirely out of the gland, the drop falling upon and bending it, the displacement being 4.7 in., with an estimated pressure of 3830 lbs. per square inch. These results rendered it evident that this piston was too small to produce displacement sufficient to break the head. Another piston of 2.99 in. arc of cross section was made, and the glands reamed out to fit, the weight of the drop being increased to 1122 pounds. With this arrangement, everything being in complete working order, and a piece of pure hammered copper being placed under the point of the indent- ing tool, and removed to a new position at each variation in the height fallen through by the drop, the results recorded in the following table were ob- tained : — Table showing the amount of indentation produced in pure copper, and the estimated pressure produced, by the fall of 1122 lbs. from different heights on a piston 1.953 in. diameter. Height. Penetration Total fall of Drop. Displacement of wa- ter in cubic inches. Estimated pressure Length of per square inch. Indentation. 4.503 .50 in. 6.364 .64 7.928 .72 9.138 .77 10.254 .895 Behabks. 1ft. 2.47 in. 14.47 in. 2 3.31 27.31 3 3.88 38.38 4 4.43 52.43 5 4.88 64.88 7.399 9.916 11.625 13.270 14.619 The numbers un- der the head of indentation give the length of the indentation. This weight of 1122 lbs. was thrown back by the elasticity of the metal and water, to a height nearly equal to that from which it had fallen, striking a number of blows of decreasing intensity before coming to a state of rest. The rebound from a fall of 5 feet again threw the piston out of the gland ; and although the slides were of sufficient length for a fall of 10 feet, owing to this cause, only 4 feet were available. 38 PEOPEETIES OF METALS FOE CANNON, It was thought that a portion of the elasticity manifested might be due to the air contained in the water ; this was accordingly removed, and other water, that had been well boiled for an hour, substituted, and every precau- tion taken to insure the absence of air from the space intended to be filled with water. The experiments were repeated to a fall of four feet, but with precisely the same results. The penetrations were measured by a graduated cylinder, so arranged that it could be pressed, or clamped as tight as desired, by a screw, but movable by the drop ; the drop and cylinder stopping at the same time, the penetration was shown by the space passed over by the cylinder. Table comparing the actual with the computed pressures producing equal indentations in pure copper. Estimated pressures. Length of indentation. Actual pressure. Length of indentation. 1000 lbs. .21 in. — - 2000 " .28 — - 3000 " .45 4503 lbs. .50 4000 " .53 _ — 5000 " .60 6364 " .64 6000 " .68 7928 " .72 7000 " .72 9138 " .77 8000 " .78 — — 9000 " .84 10254 " .895 10000 " .891 — - 11000 " .952 - - 12000 " 1.000 - - 13000 " 1.040 ~ ~ 14000 " 1.110 The indentations in the second column of the above table were made as before described. The pressures in the first column were obtained by estimating the pressures per square inch due to, or caused by, the fall of the weight, and multiplying by the area of a cross section of the indenting piston. The length of the penetrating piston was such, that the head would strike the gland before the point could come in contact with the inner end of the indenting piston. The pressure produced by the falling weight was thus estimated: — AND OF CANNON POWDEB QUALITIES. 39 Let a = area of cross section of penetrating piston, W— weight of drop, h = total height fallen through, in inches, p = penetration of piston, and p' = pressure per square inch, and n = work done by the falling weight. Then, since the work done in stopping the falling weight is equal to that due to the whole height fallen through, we have, supposing the pressure to increase directly as the penetration, — — - — d x (x) being variable between o and p ; n = a p' (V a;— fl), and when x=p, or= aj? p , or, P a p p' 2 Wh — - — = Wh and p" = ; from which it appears, that for a given weight and fall, the pressure produced is inversely as the volume displaced, and the area of a cross section of the piston inversely as the penetration. It was intended that all the indentations should be made with the same tool; but, on examining the tool before commencing the indentations recorded in the 4th column, it was found that a small piece had been acci- dentally broken out of the indenting edge since making those recorded in the 2d column. The tool was accordingly re-dressed, and made to conform as accurately as possible to the indentations already made. With the tool thus prepared, the indentations recorded in the 4th column were made, the weights producing them being accurately determined by the testing machine. Some deductions should be made from the estimated weights in the 1st column, for the friction of the indenting piston against the packing, through which it moved. The indentations recorded in the 2d column were made by pressure applied for only an instant of time, while those recorded in the 4th column were slowly made ; and it is supposed that the indentation thus made would be greater for equal pressures than those made quickly, as the disturbed metal would have more time to adjust itself to its new condition. That the indentations recorded in the 2d column were produced by pressure, 40 PEOPEETIES OF METALS FOE CANNON, and not by a blow (as might be inferred from the mode of producing them), was clearly proven, by comparing the indentation produced by a single fall of the drop with one caused by repeated falls of the same weight from the same height, the indentations thus produced and compared being precisely equal. Making deductions for friction and time, as above indicated, the corres- pondence between the estimation and actual pressure due to equal indenta- tions is sufficiently accurate to leave no doubt of the truth of the formula by which the estimated pressures were computed. In order, as far as practicable, to prevent the bounding out of the pene- trating piston, the quantity of water was so regulated, that the head of the piston, at its greatest penetration, was only about one inch from the gland. A pressure of 9000 lbs. being the greatest practical pressure attainable with this weight,, and it being supposed that an almost indefinite number of blows of this intensity would be required to break the head, and the advanced state of the season not affording time for a further increase of weight, the experiment was discontinued. The apparatus is all carefully preserved, and the experiment may be renewed at any time.* On the supposition that the increase of capacity, caused by the compression of the solid cy finder, and contained water, is equal to that caused by the elongation of the metal of the gun head, a displacement of 14.619 cubic inches, the greatest produced, would give an elongation of .0032 in. per inch on the interior surface of the head, which is just double that given by Mr. Hodgkinson as the elongation of cast iron at the rupturing strain, viz. : vis of its length, or .0016 of its length. How much of the increased capacity is due to compression of the solid cylinder and contained water, and how much to the elongation of the metal of the head, cannot, from these experiments, be accurately inferred. The elongation per inch, at rupture, of the metal of the gun from which this head was taken, as 'determined by bending specimens, is .011678. It would therefore appear that the metal of the gun head has only been extended to a little over one-fourth of its capacity ; and, supposing equal increments of extension to be produced by equal increments of pressure, a force of about 40000 pounds per inch would be required to break the head ; ♦This apparatus has since been destroyed by the burning of the Fort Pitt Foundry, where the experiments were made. •w AND OF CANNON POWDEK QUALITIES. 41 and I have no doubt it would require a force of that intensity, acting only for an instant, to break it; while one of 30000 pounds per inch, acting for any considerable length of time, would doubtless produce the same effect. Of the Effects of different rates of Application of Straining Forces upon the Bodies to which they are applied. The following remarks are in exemplification of this subject, as treated by Mr. Robert Mallet, in his valuable work "On the Construction of Artillery." In estimating the effect of any force upon an elastic or yielding material to which it may be applied, the rate of application, or the time which elapses from the instant when the force begins to act till it attains its maximum should not be neglected. This effect may be illustrated by supposing the case of an elastic body, as a steel plate, or a slip of elastic wood, firmly secured at one end, in a horizontal position, thus : let the elastic body (a b) be firmly secured in the wall ( W), and let the weight ( p) be placed slowly upon the unsupported end, the resisting body will be depressed to a position (p') when the resistance will equal the weight ( p) • this will be the position of statical equilibrium for this particular weight ( p). But if the weight (p) be removed, and the resisting body (a b) allowed to resume its horizontal position, and the weight be placed in contact with, but not resting upon, the upper surface, and suddenly let go, (a b) will be again depressed, the weight (p) being in excess of the resistance until the position (p') of statical equilibrium be reached; from this point the resistance will be in excess of the weight, and the velocity of (p) will be constantly diminished, and finally become (0), at ( p") ; and if the elasticity of (a b) be perfect, the position (p') will be intermediate between (p) and (p"), and the body (a b) consequently subjected to a strain just double as great as that of statical resistance to the weight (p). On reaching the position (p"), the elastic force of {a b) being in excess of (p), will cause {a b), with its load (p), to rise, and return to its original position ( p) ; and in case of perfect elasticity, and the absence of all other disturbing forces, (a b) would continue to vibrate between the positions (p) and ( jt>") ; but owing to imperfect elasticity, and atmospheric resistance, {a b) 42 PROPERTIES OF METALS FOB CANNON, will vibrate with a constantly decreasing extent of excursion, until it finally stops at the position of statical equilibrium Q»'). Now, if (p) had been restrained between the positions (p), and {p'), and only abandoned to the full force of gravity at some point between these two, then {a b) would have arrived at the position of statical equilibrium with a less velocity than in the previous case, and would have stopped at some point between (p') and (p"). Prom which it appears that the more slowly the straining force is applied, the less will be the velocity with which the resisting body will reach the position of statical equilibrium, and the less distance will it be forced beyond that position, by the momentum of the mass in motion ; and, consequently, the less will be the strain beyond that due to statical equilibrium. This principle is applicable to all intermittent strains ; it being known that a bridge suffers greater strain from the rapid transit of a train of cars, than from one of the same weight, but at a lower speed, and if the whole train were suspended over the bridge, in contact with the rails, and suddenly abandoned to the force of gravity, the strain produced would be double that of a very slowly passing train of the same weight. So it is, to a certain extent, in the discharge of cannon ; for with equal ultimate pressures of gas per square inch of surface, that powder will be most severe upon the gun which attains this pressure in the shortest period of time after ignition.* The fulminating powders, being almost instantaneous, develop strains almost double as great as those due to the statical pressure of the gas evolved, and hence their bursting tendencies ; and, the more rapid the combustion of any charge of powder (eaderis paribus), the greater will be the strain upon the gun in which it is burned. * My views on this subject have been modified since -writing the above, and will be given at the end of these Reports. AND OF CANNON POWDEB QUALITIES. 43 Of the Various hinds of Strain to which a Gun is subjected at each Discharge. (Pig. 1.) There are, first, a tangential strain, tending to split the gun open longitu- dinally, and being similar in its action to the force which tends to burst the hoops or bands upon any expanding substance. 2d. A longitudinal strain, tending to pull the gun apart in the direction of its length ; this tendency is greatest at or near the bottom of the bore (a b), and diminishes with the mass of that portion of the gun in front of any assumed point, being (0) at the muzzle. These strains both tend to increase the volume of the metal to which they are applied. 3d. A strain of compression, exerted from the axis outward, tending to crush the truncated wedges of which a unit of length of the gun may be supposed to consist, to give a cross section of the gun this appearance, and to diminish the thickness of the metal to which it is applied. 4th. A transverse strain, tending to break transversely, by bending out- ward the staves of which the gun may be supposed to consist, and to give to a longitudinal section the appearance shown by the dotted lines. (Fig. 1.) Tangential Strain. On the supposition that the area of a cross section of the gun remains the same before and during the application of the straining force, the resistance which a gun one calibre thick is capable of offering to a central force has been shown ("Keports of Experiments on Metals for Cannon," page A A 210 ) to be = ^-, in which a = tensile strength of r 2 . The general expression, for this resistance, per unit of length, being (^ s r 8 ' = tZZ — l f in which R and r represent the radii of the r R Rr exterior of the gun, and of the bore, and (s) the tensile strength per square inch of metal. 44 PEOPEETIES OF METALS EOE CANNON, The rupturing effort per unit of length, in terms of radius of bore and pressure, per inch, =pr; and the tendency to rupture will = rupturing effort, divided by the resistance which the gun can offer = -= — - 2 — r- = R~r — — J p . = • * ; and when R = Br, or the gun is one calibre thick, the Rrh — rh Ms — r s' ' ° tendency to rupture = ~, or where none but the tangential resistance is offered, rupture will ensue, when the pressure per square inch exceeds two- thirds of the tensile strength per square. inch. Longitudinal Strain. If the bore of the gun terminate in a hemispherical bottom, the maximum tendency to rupture, by a longitudinal strain, will be at, or a little in rear of, the junction of the cylinder and hemisphere of the bore. And it is believed that the resistance which the gun can offer to this force will differ but little from that which a hollow sphere, of the same quality of metal, whose interior and exterior diameters are equal to those of the gun, would offer to a bursting force ; and that the law of diminution of strain, from the interior outward, will be the same for any central section of the sphere, as for a cross section of the gun. To obtain an expression for this resistance, let (r) and (R) represent the interior and exterior radii of the gun, and (s) the tensile strength per square inch of metal, and (x) any variable quantity between (r) and (R). Then s 2 re x d x would express the resistance which an elementary shell, whose radius = x, would offer, if the strain were uniform throughout the entire thickness. But the strain is supposed to be a maximum when x = r, and to diminish as the square of the distance from the centre, or axis, increases ; therefore r 2 . the expression s 2 n x d x must be multiplied by — -, in order to express the elementary resistance, under the law of diminution of strain, due to any value of (x) between (r) and (R). The elementary expression, therefore, becomes s 2 n — x d x = s 2 n r — ; x' X the integral of which, or the total resistance, will = s 2 n r 2 , Nap. log. x; or, integrating between the limits x = r and x = R, we have the AND OF CANNON POWDEK QUALITIES. 45 total resistance = u = s 2 n r 2 (Nap. log. R — Nap. log. r) = 2 s re r* Nap. log. 7? 7? — . Now when _R=: 3 r, or the gun is one calibre thick, the Nap. log. of — = Nap. log. of 3 = 1.0986, and the total resistance = 2 n r 2 5 x 1.0986. But the rupturing effort = n r 2 p, (p) representing the pressure per square inch of gas at the bottom of the bore. The tendency to rupture will therefore = , ,. , „ nn „ ; or, in round J r 2 * r 2 SX 1.0986 ' ' - p numbers, after dividing by n r 2 , it will = ^; or the gun would be three times as strong longitudinally as tangentially, if the bursting effort were resisted by its tangential strength alone. That the tendency to rupture, from the action of this force, is greater at, or a little in rear of the junction of the hemisphere and cylinder of the bore, than forward of that point, is shown by the following considerations, viz : — The effect of the pressure upon that portion in rear of this point, is to crush the truncated pyramids, of which that portion of the gun may be supposed to consist, and to give to a central section of the sphere, after . rupture, the same appearance as a cross section of the cylindrical portion of a gun, of compressible material, after rupture; while no transverse cracks would be found in the cylindrical part of the gun, from this cause. And since the force is all applied at the inner surface of the gun, the strain will be greatest at that surface, and will diminish directly as the distance from the axis increases. The expression for the elementary resistance would therefore = s 2 rt r d x, and its integral, or the total resistance, would =ii=s2ffr«. And inte- grating between the limits x = r, and x = R, we have u = s 2 % r [R — r). Then if the gun be one calibre thick, R will = 3 r, and u = s2nr(J!>r — r) pit T 2 V = s 4 rt r 2 . And the tendency to rupture will =-^—3=^; or the tendency to break at or a little in rear of the junction of the cylinder and hemisphere of the bore, is twice as great as at any point forward of that position, from the action of the longitudinal force alone. The tangential and longitudinal strains are in directions at right angles to each other ; and hence, probably, neither affects the ability of the metal to resist the other, while the compressibility of the metal tends to diminish its capacity to resist either. 12 46 PEOPEETIES OF METALS FOE CANNON, (Fio. I.) Crushing/ Force. Without attempting to give an algebraic expression for the rupturing effect of this force, yet its tendency is known to be to diminish the thickness of metal in the gun, and thus to increase the diameter of the bore, beyond that which would result from the action of the transverse and tangential forces alone. This force diminishes from the bore outward, while the area of resistance increases. The effect of this, upon a compressible truncated wedge, would be to change its form from that of Fig. {a) to that of Kg. (h). (Fio. O.) And the appearance of a cross section of the gun after rupture would be that of Fig. (C). If the metal were incompressible, the appearance of a cross section of the gun after rupture would be that of Fig. (C), and no enlargement of the bore would result from the crushing of the metal ; and any enlargement caused by the action of a central force would be accompanied by an equal enlargement of the exterior diameter of the gun; and hence the strain upon the metal, at the inner and outer surfaces of the gun, would be inversely as the radii of those surfaces, instead of inversely as their squares (as in. the case of compressible metal). The expression for the total resistance per unit of length would be, for a gun one calibre thick, (r s Nap. log. 3,) or, in round numbers (r s), instead of 2 (— r s), as in the case of compressible metal. o Perfect incompressibility would therefore bring into action one-third more tangential resistance than the same metal without it would be capable of offering to a central force. It would therefore appear that, all other qualities being equal, that metal which is least compressible will offer the greatest resistance to a central force. AND OF CANNON POWDEE QUALITIES. 47 Of Resistance to Transverse Strain. In estimating the resistance which a gun can offer to a tendency to transverse rupture, it will be more simple to regard the gun as composed of staves, firmly secured at their ends, the rear ends being supposed to be secured to a central cylinder, or " breech-pin ; " and in this case it will only be necessary to consider a single stave, as all others of equal width and length would be subjected to similar and equal strain. Let us, therefore, consider the action upon a single stave, whose interior breadth is one inch ; and, Fig. 2. (a) X d I C 1 " i' (*)' if the gun be one calibre thick, the exterior breadth of this stave will be 3 inches ; and we shall be something below the actual resistance which the stave can offer, if we consider it as of rectangular section of 2 inches in breadth; this is apparent from inspection of Fig. (a). Let (a) be the stave acted upon by the pressure of gas along its inner surface, and suppose the pressure to be applied between the points {b) and (V). Now this stave is secured at both ends, and the rupturing force equally distributed along its length between the points of support, and suffers a tendency to rupture at three points, as shown in Fig. (2) by the lines (b c), {?" 2 S. Longitudinal = ..... -?— or rupture will ensue when p > 2 S. Resistance to pressure on interior of hemisphere == -*-& or rupture will ensue when p^>2S. Transverse = ...... or rupture will ensue when 2p > 3 /$'. 3# It thus appears that the tendency to rupture is greater with the tangential resistance acting alone, than with any other ; and for lengths above two, or perhaps three calibres, the tangential resistance may be said to act alone ; as the aid derived from the transverse resistance will be but trifling for greater lengths of bore. But for lengths of bore less than two calibres, this resis- tance will be aided by both the transverse and the resistance of the hemispherical breech, which causes this latter to become, probably, the weakest part of this portion of the gun, with the use of our present quick powder. Every gun should, therefore, have sufficient thickness of breech to cause rupture to take place (if at all) along the lines (b c) and {d e) Pig. 2, instead of splitting through the breech ; and after this point has been attained, any additional thickness of breech will add nothing to the strength of the gun. This investigation shows the gun to be strongest at, or near, the bottom of the bore ; and that its strength diminishes rapidly as the length of the bore increases, to a certain point (probably not more than 3 calibres from the bottom of the bore), where, for equal thickness of metal, its strength becomes sensibly uniform. Therefore, in order that the gun may be equally strained in all its parts in firing, or that the maximum velocity of shot for a given maximum pressure may be obtained, it would appear that the charge should consist of a small portion of quick powder, whose expansive force would be rapidly developed, the remainder of the charge being so constituted that the evolution of gas should be proportional to the space passed over by the shot ; and a charge 52 PEOPEETIES OE METALS EOE CANNON, thus constituted would give the maximum velocity of shot attainable, with a given length of bore and thickness of metal. Bursting Effects of Different Weights of Powder and Shot in Guns of Different Calibre. Assuming, with Hutton, that the quantities of action due to the expansive force of different weights of the same quality of powder are proportional to W v 2 W v' 2 those weights, we have the proportion — ^ — : — ■? — : :w:w'; in which ( W) it %/ and ( W) represent the sums of the weights of both powder and shot ; (w) and (w') the weights of powder, (v) and (v') the velocities of the common centres of gravity of the powder and shot, and (g) the force of gravity. Now the mean pressure, or that which, acting with uniform intensity along the distance through which the common centre of gravity of the powder and Wv 2 shot passes, in acquiring the velocity (v), will = ^— (I), being the distance above named, or length of bore of the gun. Now, supposing the guns to be equal in length of bore, the mean pressures Wv 2 W v" will be to each other as K — i • -^ — t, or as Wv 2 : W v'*. And since the 2 g I 2 g I pressure per square inch is equal to the whole pressure upon the shot and charge, divided by the area of the surface pressed, the pressures per square inch Wv 2 W v' 2 Wv 2 Wv' 2 will be to each other as — r : — -^-, or as — j— : — r^— (r) and (r'), repre- senting the radii of the bores. But the bursting effort, per unit of length, equals the pressure per square r Wv 2 inch, multiplied by the radius of the bore, and is therefore = — ^— ; or the r Wv 2 r W v' 2 bursting efforts upon the two guns will be to each other as or as r Wv 2 W v' 2 r r But the tendency to rupture, of any force, equals its rupturing effort, divided by the resistance which the body to which it is applied is capable of offering. This resistance has been shown (page 43) to be ^ . AND OE CANNON POWDEE QUALITIES. 53 W v 2 The tendencies to rupture will therefore be to each other as B r 2 S—r'S' Br W v' 2 / B Wv 2 R W v' 2 B Wi? B W v 2 ■ or as -^r-j-s ro : ™ , * n — r^-Fi, or as Br' 2 S'—r"S ' Br 2 S—r>S ' B f 2 £-r' 3 S' r 2 S(B—r)- PS'iB—r'Y Br 1 And substituting for W v^ its value derived from the proportion W v 2 : W v'*: :w: w', we have the tendencies to rupture as ,„,„ — : : ,„„ TO . rS{B — r) 7 /2 8{B — /) w And supposing the quality of the metal to be the same in both cases, these tendencies will be to each other as -5—^ — - : — . r 2 {B — r) ¥ 2 (R — /) w Now, in comparing the bursting efforts of the charges fired from the ten- inch (10-in.) Army gun with those of the charges fired from the eleven-inch (11-in.) Navy gun, we have for the Army gun if! = 15.5 in., r = 5 in., and w = 18 lbs. ; and for the Navy gun iT = 16 in., r'= 5.5 in., and w>' = 15 lbs. Therefore .R — r = R' — r', and the bursting tendencies will be to each other, supposing the quality of metal the same in both cases, as — : -j, — ; or, in round numbers, as 14 : 10. Again, if we reduce the weight of powder and shot in both cases to a cylindrical shape, the areas of their bases being equal to those of sections of the bores, the heights of the cylinders of powder will be 6.33 in. for the 10-in. Army gun, and 4.41 in. for the 11-in. Navy gun ; and the heights of the cylinders of metal will be 6.10 in. for the Army gun, and 5.5 in. for that of the Navy. Each square inch of the surface of the projectile pressed in the Army gun, will, therefore, have 6.33 cubic inches of powder in its rear, and 6.10 cubic inches of metal in front. While, in the Navy gun, the powder in rear of each square inch of surface pressed is 4.41, and the metal in front 5.5 cubic inches. It would, therefore, seem that the pressure per square inch of surface in the Army gun is greater than that in the Navy gun, in the same proportion as the pressure produced by firing a gun of one inch area of section of bore, with 6.33 cubic in. of powder, and 6.10 in. of metal, to that produced in firing the same gun with 4.41 cubic inches of powder, and 5.5 of metal, a 54 PEOPEKTIES OF METALS FOE CANNON, There can, therefore, be no doubt that the maximum pressure per square inch is greater in the Army than in the Navy gun. The length of bore to which the maximum pressure is applied, and the length of time during which the pressure in the bore exceeds the mean pressure, will also be greater in the Army gun than in that of the Navy. The difference in endurance, due to a given difference in the bursting ten- dency at each fire, is not known ; but it is believed that the difference in bursting tendency at each fire, which the foregoing reasoning indicates as existing between the two guns, will go far to account for the difference in their endurance. The tangential resistance alone has been here considered. To include the transverse resistance, would show a still greater difference in bursting tendency. The foregoing conclusions are based upon the supposition that the max- imum pressures will be proportional to the mean ; which is not known, that I am aware of, to be strictly true. The mean will differ less from the maximum pressure as the length of the bore becomes shorter, for a given charge of powder, down to the point where the maximum pressure is reached. The position of this point will depend upon the weights of the shot and charge, and upon the rapidity of combustion of the charge ; and the more rapid the combustion, and the heavier the shot, the less distance will it have moved from its original position before reaching that of maximum pressure. The maximum pressure itself, other things being equal, will increase with the rapidity of combustion, being an absolute maximum for the kind of charge used, when the combustion is instantaneous; in which case the maximum pressure would be reached before the shot has moved at all. And the max- imum pressure per unit of surface would no longer depend upon the weight of either shot or charge, but upon the difference in volume of the gas from that of the charge from which it was evolved. Not so, however, with the bursting effects, as they would increase with the weight of both shot and charge ; the weight and shape of the charge regu- lating the length of bore subjected to the maximum pressure, and the weight of the shot the time during which pressure, differing but little from the max- imum, is applied. This investigation is offered as a probable explanation of the difference in endurance of guns cast and cooled in the same manner, made from iron of AND OF CANNON POWDEE QUALITIES. 55 the same qualities, treated in nearly the same manner, and differing but slightly in quality, in the guns themselves. Probability, however, is not knowledge, but it is the most that can now be offered, as well upon this as upon many other points of equal importance. We do no^ know, for example, what qualities of iron are necessary to make the best gun ; nor, if we did, do we know how, from any of its ores, constantly to produce iron which shall possess those qualities. We do not know whether guns should be cast hollow or solid, nor the proper rate of cooling for either mode of casting. We do not know the best exterior model for guns, nor whether those of large calibre should be made with or without chambers. We do not know the effects of time upon the endurance of guns, — whether they are better when new, or after they have lain unused for any given length of time. We do not know the maximum statical pressure due to a given weight of powder and shot, nor how much the rate of combustion of the charge, or the rate of application of the force, causes the bursting tendency to exceed that due to the statical pressure. We do not know the difference in endurance due to a given difference in bursting tendency at each discharge, nor what weight of projectile is equiva- lent in bursting tendency to a given weight of powder, nor the difference in endurance due to a given difference in thickness of metal. We do not know the difference in bursting tendency due to a given differ- ence in temperature of the same charge of powder at the moment of ignition. Nor do we know the proper constitution of charge in order to produce a given velocity of projectile with the minimum bursting tendency upon the gun. And it is believed that the true interests of the country would be promoted, in a military point of view, by entering, at as early a period as practicable, upon a series of experiments which would supply positive knowledge in place of probability in some, and positive ignorance in many other points of the utmost importance to the national defence ; for it is better that millions should be expended in time of peace, and from an overflowing treasury, than that a single gun should burst in action. (Signed,) T. J. RODMAN, Copt, of Or'd. To Col. H. K. Craig, Ordnance Office, Washington, D. C. AiiBQHAinr Arsenal, January 30th, 1857. 56 PROPERTIES OF METALS POR CANNON, ETC. I subscribe to those facts and opinions, detailed in the above Report, which have reference to the casting, cooling, proving, and other operations with the trial guns, which were enjoined on me to observe and report upon, in conjunction with Capt. Rodman, by virtue of instructions from the Ordnance Department relative to the matter, dated August 18th, 1856. Also, from the conflicting and unsatisfactory results which have been obtained in the recent trials with Columbiads, I am of the opinion that we do not possess the requisite knowledge to manufacture guns of these calibres which can be relied upon for any given number of fires. I, therefore, fully concur in Capt. Rodman's suggestion, that the casting of service Columbiads be suspended for the present, and that a series of experi- ments be instituted to furnish the requisite data for the fabrication of reliable cannon of these calibres. S. CRISPIN, Lieut, of Ord. To Colonel H. K. Craig, Ordnance Office, Washington City, D. C. Alleghany Arsenal, January 30th, 1857. } REPORT O N THE RELATIVE ENDURANCE OF GUNS MADE FROM THE SAME IEON, BUT MELTED IN FURNACES OF DIFFERENT CONSTRUCTION. ALSO, OF GUNS MADE FROM THE SAME IRON, MELTED IN THE SAME FURNACES, BUT DIFFERENTLY COOLED. 1>LA1 T E I — - . S i .a +- -;9f- J>LAT£ ZZ L_ ~ : i ■ ' ' ■' - 1 J-- REPORT OF EXPEEIMENTS MADE FOE THE PURPOSE OF DETERMINING THE RELA- TIVE ENDURANCE OF GUNS MADE FROM THE SAME IRON, BUT MELTED IN FURNACES OF DIFFERENT CONSTRUCTION ; ALSO, THAT OF THOSE MADE FROM THE SAME IRON, MELTED IN THE SAME FURNACES, BUT DIFFERENTLY COOLED; ONE GUN BEING CAST SOLID, AND COOLED FROM THE EXTERIOR, AND THE OTHER CAST HOLLOW, AND COOLED FROM THE INTERIOR. At the West Point Foundry, on the 2d and 3d of July, 1857, being assisted by Lieut. F. I. Shunk, I selected a sufficient quantity of iron for the fabrica- tion of 6 10-inch Columbiads. The iron selected was in the following proportions, viz. : — Of Greenwood iron, 405 cwt. of No. 1, 500 cwt. of No. 2, and 205 cwt. of No. 3. Of Salisbury iron, 225 cwt. Of Scotch pig, 45 cwt. Of re-melted iron, in the above proportions, 115 cwt. Of this iron, two-thirds, or enough for 4 10-inch guns, was distinctly marked on every pig, and sent to the Fort Pitt Foundry, each kind having a separate mark. The other third was left at the "West Point Foundry, each kind being put in a separate pile. In order to insure identity in the quality of the iron to be used at the different furnaces, it was selected in the following manner, viz. : — It was first classified at the smelting furnace, into the grades Nos. 1, 2, and 3, pig ; after which, the different grades of Greenwood iron were kept separate, loaded into wagons, hauled to the river, unloaded, re-loaded on a boat, landed at the foundry wharf, loaded on cars, hauled to the foundry yard, there unloaded, and piled in separate piles. From the piles thus made were taken, alternately, two pigs to go to Fort Pitt, and one to remain. 60 PEOPEHTIES OF METALS FOE CANNON, The same mode was pursued in selecting the Salisbury, Scotch, and re-melted iron. That sent to the Fort Pitt Foundry was assorted under my own super- vision, for charging the furnaces ; the marks which I had seen put on being perfectly distinct on every pig. Lieut. Shunk supervised the charging of the West Point furnaces, from the iron which he had assisted me to select. The iron was used at both foundries in the same proportions as those in which it had been selected ; and it was intended that on the same day on which the "West Point Foundry would cast one solid 10-inch gun, the Fort Pitt Foundry would cast two, one solid and the other hollow ; but, owing to mismanagement in the Post Office Department, my letter of the 8th of August, advising the proprietors of the "West Point Foundry that we would cast on the 13th of that month, was not received till the morning of the latter date, and the West Point gun was not cast till the following day. The quantities of iron used in charging the West Point furnace, as furnished by Lieut. Shunk, are as follows, viz. : — Greenwood iron, No. 1 pig, tt a u o " ti u a o n Salisbury iron, pig. Scotch iron, pig, Ke-melted iron, Total charge, This iron was charged in three furnaces ; the quantity charged in each furnace is not given. The iron at the Fort Pitt Foundry was as follows, viz. : — . 5936 lbs . 7504 a . 3136 a . 3360 a 672 i( . 2240 a . 22848 lbs. Greenwood iron, No. 1 pig, . 13500 lbs. a it a 2 " . 16500 « a u a o a 6833 " Salisbury iron, pig, 7500 « Scotch iron, pig, 1500 " Re-melted iron, .... 3833 " Total charge, . 49766 lbs. AND OE CANNON POWDEE QUALITIES. 61 This iron was charged in three furnaces, the different kinds of iron being in the same proportion in each. Furnace No. 1 received, . . . .12983 lbs. " " 2 « .... 16848 « " " 3 « .... 19935 " Total, as above, 49766 lbs. Of the Flasks. The flasks in which these guns were cast were all of circular cross section ; and of sufficient size to admit a wall of sand around the gun, of from 4 to 5 inches thick. The flask in which the "West Point gun was cast, was placed in an open pit, and rammed up all round with moulding sand, to a point a little above the position of the trunnions, green or moist sand being used for this purpose. Those in which the Fort Pitt guns were cast, were placed in pits with closely fitting covers, but not rammed up with sand. Of the Pits. The pits in which the Fort Pitt guns were cast, were both heated by fire previous to casting. That in which the solid gun was cast, had fire lighted in it on the 10th of August, which continued to burn till the gun was cast, at which time it was burning moderately, and was allowed to burn out, no fuel being added to it after casting. This pit was in good order for slow and regular cooling. That in which the hollow gun was cast had a brisk fire lighted in it on the evening of the 8th of August, which continued to burn till the 11th ; and on the 12th the ashes were cleaned out, and fresh fuel placed on the grate bars. This fuel ignited from drops of melted iron, which fell upon it while casting, and soon produced a brisk fire, which was kept burning till 1 o'clock, P. M., on the 15th, after which time no more fuel was added. Of the Furnaces. The furnaces in which the iron for all these guns was melted, are what are termed air furnaces. 16 62 PEOPEETIES OF METALS FOE CANNON, In these furnaces the draft is produced by chimneys, instead of a blast, which is used in the cupola furnace. The metal for what is termed a "heat," is all placed in the metal chamber before the furnace is lighted. The fuel is placed on grate bars in the fuel chamber, or fire-place, and the flame from this fuel passes through the metal chamber on its way to the chimney. The iron is melted by this flame, without coming in contact with solid carbon at all, which is not the case in the cupola furnace, where the fuel and iron are mixed together. The main feature of difference between the Fort Pitt and West Point furnaces is, that in the Port Pitt furnaces, the iron, as it melts, runs back towards the bridge wall, the crown of the furnace over the bridge wall being so constructed as to cause the flame to impinge against the surface of the pool of melted metal, while at its greatest temperature ; whereas, in the West Point furnace, the melted iron flows from the bridge wall, or along with the flame, so that the flame does not reach the deepest part of the metal pool till after its temperature has been somewhat reduced, and does not at any time impinge so directly against the surface of the melted iron, as in the Port Pitt furnaces. Figure 1, Plate 1, shows a vertical section ; and Figure 2, same plate, shows the plan of the West Point furnaces. Plate 2, shows a vertical section of the Fort Pitt furnaces. The same letter corresponds to like parts in each (a) ash pit (/), fuel chamber (c), metal chamber (F), chimney flues {g), metal pool, and (t) tap- hole. These drawings are only intended to give a general idea of the construction of these furnaces, not being accurately drawn to any particular scale. Casting. The furnaces from which the Fort Pitt guns were cast, were all lighted at 9 A. M., on the 13th of August, and the metal was melted at 12^ P. M. Furnace No. 3 was tapped at 2h. 5m. P. M., and No. 2 two minutes after ; No. 1 not being tapped till after the metal from No. 3 had ceased to flow ; so that the metal flowed from only two furnaces at the same time. The metal appeared fluid, and in good order for casting, and was all received into one vessel, from which it issued at two orifices, and ran in separate runners to the two gun moulds, which it entered simultaneously by side AND OF CANNON POWDEB QUALITIES. 63 runners, one to each mould, till the metal had risen above the trunnions, when the side runners were closed, and the metal entered the moulds directly at their mouths. The hollow gun mould was filled in 8 minutes, and the solid one in 9| minutes after the first metal entered. Lieut. Shunk writes in relation to casting the West Point gun, as follows, viz. : — " The iron was taken from the piles you specified ; set fire at 8 A. M., down at lj P. M., cast at 4 P. M. — Time in fusion, 1\ hours. The metal appeared as usual, and entered both sides of the mould. The proof bars were failures, from a mixture of slag," Cooling. The West Point gun, No. 983, cooled in the mould, rammed up in green sand, as before described, from which it was removed on the 8th day after casting. Water circulated through the core barrel of the hollow cast gun (No. 334), from the time the metal reached the bottom of the core in casting, till the core barrel was removed, at the rate of three cubic feet per minute. Just after casting, the water entered at 75°, and left at 95°. At 16 hours after casting, the core barrel was removed, and the water circulated through the cavity thus left, at the rate of two cubic feet per minute. Just after the removal of the core barrel, the water entered at 75°, and left at 136°. At 19 \ hours after casting, the flow of water was increased to three cubic feet per minute ; at which time and rate it entered at 75°, and left at 108°. Water continued to circulate through the gun at rates varying but little from three cubic feet per minute, and leaving with a gradually decreasing temperature, till \\\ A. M., on the 21st, at which time it entered and left at the same temperature. The water was then shut off, and the pit uncovered for the removal of the flask from the gun. The gun was hoisted out of the pit at 6 P. M. ; and water which had remained in the gun from ll£ A. M. of same day, till that time, had a temperature of 80°. 64 PKOPEKTIES OF METALS EOE CANNON, The solid cast gun, No. 335, remained in the pit, as before described, till the afternoon of the 25th, when it was removed from flask and pit. Both these guns were cast considerably larger than the finished gun, from a short distance in front of the chase curves to the tops of the sinking heads, which were about 2 J feet long in both guns. There was a marked difference in the appearance of the newly-turned surfaces of these guns ; the solid cast gun presented a clouded or dappled appearance, while the hollow one presented a much finer, and more uniformly mottled appearance. The finished hollow cast gun was marked on one side, just at the neck-ring, with small cavities, some of them to the depth of perhaps three-tenths of an inch. These 1 was at first disposed to attribute to what is called soakage ; but, from the fact that they made their appearance near the exterior surface of the casting, I am rather inclined to regard them as sand holes, caused by small particles of sand or scoria lodging in that part of the casting. The guns were all accurately measured, and found to be within the prescribed limits of variation, except that the West Point gun was 25 inches greater than the prescribed diameter at the swell of the muzzle, and 15 inches less in total length of bore. The West Point gun was received here on the 20th of October; and, together with the Port Pitt guns, had the proof charges fired from it on the 22d of that month. The guns were then suspended for extreme proof, each in its own frame. Proof Charges. First fire, 20 lbs. powder, one strapped shot, and one wad, above the shot. 2d fire, 24 lbs. powder, and one strapped shell. Service Charges. 14 lbs. powder, and one solid shot, strapped ; the charges of powder were all weighed. Of the Powder. All the powder used in the proof of these guns was made by the Messrs. Dupont, in 1857. It was highly glazed, remarkably free from dust, and very uniform in size of grain. AND OF CANNON POWDEE QUALITIES. 65 Mean eprouvette range, from 295 to 300 yards. Eanges of the proof charges were 311 for the shot, and 312 for the shells. In order to insure equality in the proof of the different guns, they were fired alternate rounds, with charges of powder of the same eprouvette range, and taken from the same cask. Six cartridges were taken from each of three bands, and what remained in them all was put into one, and well rolled and shaken up, and from this three other cartridges were made ; this mode gave even rounds for the three guns ; and the same principle was observed in the preparation of the cartridges for the two that remained after the first one broke. The cartridges all had their proof ranges marked on them in the laboratory, and were finished on blocks 7.25 inch diameter, and of sufficient thickness to make the finished cartridge 12 inches long, or to fill the chamber. Of the three shells used in the proof, and afterwards recovered, one was apparently uninjured ; one was found in fragments in the bank into which it had been fired, and the other was considerably distorted, and slightly cracked in the fuze hole. Mode of Discharging the Guns. These guns were all fired with friction tubes ; those used in the first part of the firing were made at Fort Monroe Arsenal, had been on hand for some time, and frequently failed. The latter part of the firing was done with tubes made at the Frankford Arsenal, of which not over two per cent, failed. Endurance of the Guns. These guns all enlarged more than usual under the proof charges ; but no other indications of rupture were observed, till after the 88th fire, when both the solid cast guns were observed to be cracked at the junction of the chamber and taper. Two cracks, on nearly opposite sides, were observable in the Fort Pitt gun, and but one in the "West Point gun ; those in the Fort Pitt gun appeared to be largest. These cracks all gradually increased with subsequent firing ; those in the Fort Pitt gun appeared to open more, while that in the "West Point gun 66 PEOPEETIES OF METALS EOE CANNON, extended longitudinally, and was soon accompanied by another, about 100° distant from it. The West Point gun broke through the above described cracks, at the 169th fire, including the proof charges ; this gun broke into three pieces, splitting through the breech, and forward, just to include the left trunnion in the largest of the two small pieces, leaving one-fourth of breech and cylinder? and the other trunnion attached to the chase. The only peculiarities in the fracture of this gun are, that it had three longitudinal lines of fracture, which I have never before known to occur in guns of this calibre ; and the unusually short distance from the breech to which the fracture extended. The crack in the Fort Pitt solid cast gun continued to open and extend further into the chamber, other cracks making their appearance as the firing proceeded ; till, at the last observation, the two first described cracks extended from the junction of the bore and taper, down into the hemisphere of the chamber, leaving about 2 inches of apparently solid metal between them. This gun broke at the 399th fire, into three pieces, through the first named cracks, the cylinder breaking into two nearly equal pieces through the breech. The fracture extended forward above the left, and below the right trunnions, to a short distance in front of the trunnions, where the two breech pieces broke off from the chase. In the bottom of the bore, and a short distance in front of the seat of shot in this gun, were two or three small cavities, indicative of "soakage." No cracks were observed in the hollow cast gun till after about the 600th fire, when a number of small cracks were observed just at the junction of the chamber and taper ; these cracks had not the tortuous appearance of those in the other guns, but had more the appearance of having been cut and burned out by the gas. Cracks of this character continued to multiply as the firing progressed ; but none of them appeared to enlarge perceptibly from day to day, as those in the other guns, till about the 1000th round, when the whole chamber and taper were marked by an almost infinite number of small, shallow cracks, the bottom of the chamber having a neWike appearance ; and three cracks, larger than the rest, appeared to be extending down into the chamber. These cracks continued to extend very slowly with subsequent firing, and had, at the 1600th fire, extended three or four inches into the chamber, and AND OF CANNON POWDEB QUALITIES. 67 about the same distance into the taper ; the interior appearance of the gun having in no other wise changed. This gun has been fired 1600 rounds, including proof charges ; two new vents were inserted, one after the 529th fire, and the other after the 1040th fire. The vents in all these guns were bored in planes perpendicular to the axis of the guns, at the junction of the hemispheres and cylinders of the chambers ; and 6 inches on the exterior, and 3 inches on the interior, from a plane through the axis of the gun, and perpendicular to that of the trunnions. Plate 3 shows the lines of fracture in the Fort Pitt gun, and Plate 4 shows those of the West Point gun. Plate 5 shows the positions of the different specimens for density and tenacity in the guns from which they were taken ; specimens from corresponding positions being similarly marked and numbered. Head specimens were taken (e) from the axis (1), from near surface of bore (2), from middle of thickness, and (3) from near exterior surface ; axis of all, parallel to those of the guns from which they were taken. 68 PEOPEKTIES OE METALS EOE CANNON, Table showing the interior measurement of the first set of triplicate Columliads, of 1857. Hollow. No. 334. F. P. F. Solid. No. 335. W. P.F. Solid. No. 983. Distance from Muzzle. Before Proof. After Proof. Before Proof. After Proof. Before Proof. After Proof. 10 10.004 10.004 10.002 10.004 10.001 10.002 20 5 5 3 3 1 2 30 6 6 4 5 1 2 40 7 7 5 6 1 2 50 8 8 5 6 2 60 8 8 5 5 2 70 7 7 4 4 1 2 80 6 6 4 13 5 81 5 5 4 10 5 82 5 5 4 7 3 83 6 6 4 7 3 84 5 6 4 7 3 85 5 5 4 7 3 86 5 5 4 8 3 87 6 6 4 8 3 88 6 7 4 9 3 89 6 9 5 24 11 90 6 27 5 57 35 91 6 47 5 80 52 92 6 55 5 83 47 92£ 7 54 4 82 1 22 99£ 8.013 8.015 8.011 8.018 8.022 8.023 103 8.010 .010 9 13 12 22 106i 8.010 .010 9. 10 3 This, and following tables of enlargements, are recorded as read from the Star Gauge, which accounts for apparent discrepancies. TLAT£Hr. Si(7e r7eratiett Top View Scale J ,].5~ a 'Sjze. Gem X° 335 FPF. J } L.4TEJV ,\'ff/c r/fi?fOr// 7b jj I /'err Tl r jP 17 ULATEl Draiting sJtem'ng Jcsitum o f Specimens fyrfJensiiya/uf Tensile Strength, -6 y 3 _> J > '-I AND OF CANNON POWDER QUALITIES. G9 Tablb showing enlargement of bores and chambers above their original diameters, after the undermentioned numbers of fires, in thousandths of an inch. 2d FIRE. (P ■oof.) 14th FIRE. 28th FIRE. 48th FIRE. Distance Fort Pitt. W.P. Fort Pitt. W. P. Fort Pitt. W.P. Fort Pitt. W.P. from Muzzle. Hollow. Solid. Solid. Hollow. Solid. Solid. Hollow. Solid. Solid. Hollow. Solid. Solid. 334 335 983 334 333 983 334 335 983 334 335 983 10 2 2 2 2 20 SO 2 1 2 1 2 2 40 1 1 1 1 1 1 2 50 1 1 2 1 2 1 3 1 60 4 1 5 1 5 1 6 1 70 1 2 3 80 9 3 10 3 11 h 11 3 81 7 4 8 4 9 4 10 4 82 4 1 5 1 5 1 6 1 83 3 1 3 1 3 1 3 1 84 3 3 3 3 3 3 3 3 85 3 3 3 3 3 4 3 4 86 3 3 1 3 3 1 3 4 1 3 4 87 4 3 4 3 4 4 4 4 88 1 5 3 1 6 4 1 7 4 1 8 4 89 3 19 8 14 26 8 14 30 9 14 34 10 90 21 52 27 25 61 28 25 65 28 25 68 29 91 36 75 41 36 90 41 36 93 42 36 95 44 92 49 78 42 50 95 42 50 97 42 50 98 42 Q91 47 78 22 53 87 22 53 92 22 53 94 22 CHAMBERS. 99i 2 7 1 5 13 3 5 10 4 5 16 5 103 4 2 8 2 2 8 3 2 10 4 1061 1 1 6 1 7 1 8 1 70 PBOPEETIES OF METALS EOE CANNON, Enlargement of Bores and Chambers — Continued. 68th FIRE. 88th FIRE. Fort Pitt. W.P. Fort Pitt. W.P. Distance from Hollow. No. 334. Solid. No. 335. Solid. No. 983. Hollow. No. 334. Solid. No. 335. Solid. No. 983. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. 10 2 3 O 20 - - 1 - - - 1 — . 30 - 21 - 3 - - 2 - 3 _ 40 - - 2 - - 1 - 2 _ 50 - - 4 - 1 - - 4 _ 1 60 - - 6 - 1 - — 6 _ 1 70 - - 3 - - - 4 _ 80 - - 12 - 3 - - 12 _ 4 81 - - 10 - 4 - - 11 _ 4 82 - - 6 - 1 - - 6 __ 1 83 - - 4 - 1 - - 4 _ 1 84 - — 3 - 3 - — 4 _ 3 85 - - 3 - 4 - — 4 _ 4 86 - 1 - 3 - 4 - 1 - 3 _ 4 87 - - 4 - 4 — - 4 _ 4 88 1 1 8 9 5 5 1 1 8 10 5 5 89 3 14 9 36 8 12 3 14 9 38 8 13 90 4 25 13 70 9 30 4 25 13 71 9 31 91 5 36 20 96 9 46 5 36 20 96 9 48 92 17 50 25 98 10 43 17 50 26 98 10 44 921 21 53 28 95 1 22 22 30 30 96 1 24 CHAMBEES. 99* 5 5 14 16 6 6 5 5 15 16 7 6 103 1 2 9 10 6 5 1 2 9 11 6 5 106£ 1 8 9 5 2 1 8 9 6 3 AND OF CANNON POWDER QUALITIES. 71 Enlargement of Bores and Chambers — Continued. 108th EIRE. 133d FIRE. Fort*Pitt. W. P. Fort Pitt. W.P. Distance from Hollow. 334. Solid. 335. Solid. 983. Hollow. 334. Solid. 335. Solid. 983. lVTn 7 ?1 p J-'l LLZiZilC- Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. 10 3 3 _ 20 - - 2 — - — 2 - 30 - 2 — 3 - — 2 - 3 - 40 - 1 - 3 - - 1 - 3 - 50 - — 4 - 1 - - 5 - 1 60 - - 6 - 2 - - 7 - 2 70 - - 4 - — - 4 - 1 80 - — 12 - 4 - - 13 - 4 81 — - 11 - 4 - - 12 - 5 82 - - 6 - 1 - - 7 - 2 83 - - 4 - 1 - - 4 - 2 84 - - 4 - 3 - — 4 - 3 85 - - 4 - 4 - - 5 - 5 86 _ 1 - 4 - 4 - 1 - 4 - 5 87 - - 4 - 4 - - 4 - 5 88 1 2 8 12 5 5 1 2 8 14 6 6 89 3 14 9 40 8 13 3 15 9 41 8 14 90 4 25 14 72 9 33 4 25 14 73 9 34 91 5 36 20 97 9 50 5 36 20 97 9 52 92 17 50 27 99 10 46 18 50 28 99 10 49 92i 23 53 32 98 1 25 24 53 33 100 1 27 CHAMBERS. 99J- 5 5 15 17 7 7 5 5 16 17 7 7 103 1 2 9 12 6 6 1 2 9 13 6 7 1061 1 8 9 1 8 9 72 PEOPEETIES OE METALS EOE CANNON, Enlargement of Bores and Chambers — Continued. 158th FIRE. 190th FIRE. 210th FIKE. Distance Fort Pitt. W.P. Fort Pitt. * Fort Pitt. from Hollow. 334. Solid. 335. Solid. 983. Holloa . 334. Solid. 335. Hollon . 334. Solid. 335. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. 10 3 4 4 20 - - 2 — - - 3 - - 3 30 — 2 - • 4 - - 2 - 4 - 2 - 4 40 — 1 - 4 - - 1 - 4 - 1 - 4 50 - - 5 - 1 - - 6 - - 6 60 - - 7 - 2 - - 8 - - 8 70 - - 5 - 1 - - 5 - - 5 80 — — 13 - 5 - - 13 - - 14 81 — - 12 - 6 - - 13 - - 13 82 - - 7 - 2 - - 8 - - 8 83 - - 5 - 2 - - 5 - - 6 84 - - 5 - 4 - - 5 - - 5 85 - - 5 - 5 - - 5 - - 6 86 - 1 - 4 - 5 - 1 - 5 - 1 - 5 87 — - 4 - 5 - - 5 - — 5 88 1 3 8 15 6 6 1 o 9 16 1 3 9 16 89 3 15 9 43 8 15 3 15 10 45 3 15 10 47 90 4 25 15 74 9 35 4 25 15 75 4 25 15 76 91 5 36 20 97 9 54 5 36 21 98 6 36 21 98 92 18 50 28 100 10 50 19 50 29 101 20 51 29 102 92£ 24 53 35 101 1 30 26 53 35 103 28 54 36 107 CHAMBEES. 99i 5 5 17 18 7 8 5 5 18 18 5 6 18 19 103 1 2 9 14 6 8 1 o 10 15 1 2 10 15 106i 1 8 10 1 8 10 1 8 10 AND OF CANNON POWJDEE QUALITIES. 73 Enlargement of Bores and Chambers — Continued. 241st FIRE. 272d FIRE. 304th FIRE. Distance Fort Pitt. Fort Pitt. Fort Pitt. from Hollow 334. Solid. 335. Hollow 334. Solid. 335. Hollow . 334. Solid. 335. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. 10 5 6 1 7 20 - - 4 - - 5 - - 5 30 -. 2 - 5 — 2 — 5 _ 2 — 5 40 - 1 - 5 - 1 - 5 - 1 — 6 50 - - 7 - - 7 - — 8 60 - - 8 — - 9 - — 9 70 - - 6 — - 6 - — 7 80 - - 14 - - It - — 15 81 - - 13 - - 14 - — 14 82 - - 9 - - 9 - - 9 83 - - 6 - - 7 — — 7 84 — - 6 - - 6 — - 7 85 - - 6 - - 7 - 1 — 7 86 - 1 - 6 - 1 - 6 - 1 - 6 87 — - 6 - 1 - 6 - 1 - 7 88 2 4 9 17 3 4 10 19 3 4 10 21 89 3 15 11 49 3 15 12 51 3 15 13 53 90 5 25 16 77 6 25 17 78 7 2-1 17 79 91 6 36 21 99 7 37 22 100 7 37 22 102 92 21 52 30 106 23 52 32 109 24 53 34 114 921 30 55 37 112 33 56 37 116 35 56 38 118 CHAMBEES. 99J- 5 6 18 20 6 7 19 21 6 7 19 22 103 1 o 11 16 1 2 11 17 1 3 11 19 106J- 1 8 12 1 9 13 1 9 14 74 PEOPEETIES OF METALS FOE CANNON, Enlargement of Bores and Chambers — Continued. 336th FIRE. 368th FIKE. 401st FIRE. 457th FIRE. Distance Fort Pitt. Fort Pitt. Fort Pitt. Fort Pitt. from Muzzle. Hollow 334. Solid. 335. Hollow 334. Solid. 335. Hollow. 334. Hollow. 334. Horz. Vert. Horz. Tert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. 10 1 8 1 _ 8 _ 2 2 20 _ - 6 - - 7 - - 1 30 _ 2 — 6 — 2 - 6 - 2 - 2 40 — 1 — 7 - 1 - 8 - 2 - 2 50 — - 9 - - 9 - - 1 60 — — 10 — - 10 - - 70 — 1 — 8 - 1 - 9 - 1 - 2 80 — — 15 — - 15 - - 1 81 — - 14 - - 14 - - 82 — — 10 - - 10 - - 1 83 - - 8 - - 8 - - 84 — — 8 - - 9 - - 1 85 — 1 — 8 - 1 - 9 - 2 — 2 86 - 1 — 6 - 1 - 7 - 2 - 2 87 — 1 - 7 - 1 - 8 - 1 - 2 88 4 5 10 23 4 5 10 27 5 6 5 6 89 4 15 14 56 4 16 15 61 4 16 5 17 90 7 25 18 81 8 26 18 83 8 26 8 26 91 7 38 23 105 7 38 23 112 8 40 8 43 92 25 54 37 118 26 56 41 119 27 59 29 66 921 37 57 41 120 40 60 43 123 42 64 45 69 CHAMBEES. m 6 8 20 24 6 9 22 29 6 10 6 10 103 1 3 11 20 1 3 11 23 2 3 3 4 106i 1 9 15 1 10 17 1 1 AND OF CANNON POWDEE QUALITIES. 75 Enlargement of Bores and Chambers — Continued. 529th EIRE. 590th FIRE. 665th FIRE. 793ed FIRE. 804th FffiE. Distance Fort Pitt. Fort Pitt. Fort Pitt. Fort Pitt. Fort Pitt. from Hollow. 334. Hollow. 334. Hollow. 334. Hollow. 334. Hollow. 334. Muzzle. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. 10 _ 2 2 3 3 4 20 - 1 — 1 _ 2 — 2 _ 3 30 - 2 — 2 — 2 — 2 _ 3 40 - 2 _ 3 _ 4 — 4 _ 4 50 - 1 — 2 - 4 — 4 _ 4 60 - - 1 — 1 — 1 _ 2 70 - 2 - 3 - 3 - 3 _ 3 80 - 1 - 1 - 2 — 2 _ 2 81 - 1 — 1 — 1 — 1 _ 1 82 - 1 • _ 1 — 1 — 1 _ 1 83 - 1 — 1 — 1 — 1 _ 1 84 - 2 — 2 — 2 - 2 _ 2 85 - 3 — 3 - 3 - 3 _ 3 86 - 3 - 3 - 3 — 3 — 4 87 - 3 - 3 - 3 - 3 _ 4 88 6 5 6 6 7 6 7 6 7 6 89 5 17 6 18 6 19 7 20 7 21 90 8 27 8 28 9 30 9 33 10 35 91 8 46 9 50 10 54 11 58 13 63 92 33 74 37 82 43 90 48 96 53 102 921 49 76 54 83 60 89 66 94 71 79 CHAMBEES. 99^ 6 11 6 11 6 13 7 • 15 7 16 103 3 4 4 4 4 4 5 5 5 6 1061 1 1 1 1 1 1 1 1 1 76 PEOPEETIES OF METALS FOE CANNON, Enlargement of Bores and Chambers — Continued. 885th FERE. 961st FIRE. 1040th FIRE. 1120th FIRE. 1200th FIRE. Distance Fort Pitt. Fort Pitt. Fort Pitt. Fort Pitt. Fort Pitt. from Hollow. 334. Hollow 334. Hollow. 334. Hollow. 334. Hollow. 334. Horz. Vert. Horz, Vert. Horz. Vert. Horz. Vert. Horz. Vert. 10 5 5 6 6 . 6 20 - 4 - 5 - 5 - 6 - 6 30 - 4 — 5 - 5 - 6 - 7 40 - 4 - 5 - 5 - 6 - 6 50 - 4 - 5 - 6 - 6 - 7 60 - 2 - 3 - 3 - 4 - 4 70 - 3 - 4 - 4 - 5 - 6 80 - 2 - 3 - 3 - 3 - 4 81 - 2 - 2 - 2 - 3 - 3 82 - 1 - 2 - 3 - 3 - 3 83 - 1 - 2 - 2 - 3 - 3 84 - 3 - 3 - 4 - 4 - 4 85 - 4 - 4 - 5 - 6 - 7 86 - 4 - 4 - 5 - 5 - 6 87 - 4 - 4 - 5 - 5 — 6 88 6 8 7 9 7 10 8 10 8 11 89 8 22 8 23 9 24 10 25 11 26 90 12 36 14 37 14 38 15 39 16 40 91 16 69 20 75 25 77 28 80 32 84 92 57 105 63 108 68 111 74 114 80 117 921 75 105 81 109 85 113 92 116 98 118 CHAMBEES. 991 7 17 8 19 8 20 8 2:1 8 26 103 6 7 6 8 6 8 6 9 6 11 1061 1 2 1 2 1 2 o 3 2 4 AND OF CANNON POWDEE QUALITIES. 77 Enlargement of Bores and Chambers — Continued. 1280th FIRE. 13G0th FIRE. 1440th FIRE. 1520th FIRE. 1600th FIRE. Distance Fort Pitt. Fort Pitt. Fort Pitt. Fort Pitt. Fort Pitt. from Hollow. 334. Hollow. 334. Hollow. 334. Hollow. 334. HoUow. 334. Muzzle. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. Horz. Vert. 10 7 8 • 9 10 11 12 20 - 6 - 7 - 8 — 9 10 9 30 - 8 - 9 - 10 — 11 11 12 40 - 7 - 8 - 9 _ 9 9 10 50 - 8 - 8 - 9 — 10 7 10 60 - 5 - 6 - 7 _ 7 7 8 70 - 7 - 8 - 9 _ 10 10 11 80 - 5 - 5 - 6 —. 6 7 7 81 - 4 - 4 - 5 — 5 8 6 82 - 3 - 4 — 5 _ 5 6 6 83 - 3 - " 4 - 4 _ 5 4 5 84 - 5 - 6 - 6 _ 7 5 8 85 - 7 - 8 - 9 — 10 7 10 86 - 6 - 7 - 7 — 8 9 9 87 - 6 - 7 - 7 — 8 9 9 88 9 12 9 12 10 13 11 14 11 15 89 13 27 15 28 17 29 19 30 20 30 90 17 41 19 42 21 43 22 44 24 44 91 35 88 39 90 42 92 44 94 45 95 92 85 120 90 123 96 127 101 130 105 133 921 101 120 108 121 112 123 114 125 116 126 CHAMBEES. m 8 29 8 33 8 35 9 38 9 42 103 7 13 7 16 7 18 7 20 7 21 106* 2 5 3 6 4 7 4 7 4 7 78 PEOPEETIES OF METALS FOE CANNON, Table showing the diameters of Vents, after the undermentioned numbers of fires, in hundredths of an inch. Distance 133d Fire. 158th Fire. 190th Fire. 210th Fire. 241st Fire. from Exterior. 334 33S 983 334 335 983 334 335 334 335 334 335 1 21 21 22 21 21 22 21 21 22 22 22 22 2 21 21 22 21 21 22 21 21 22 22 23 23 3 22 22 22 22 22 22 22 22 22 22 23 23 4 21 22 22 22 22 23 22 22 23 22 24 24 5 21 22 22 22 22 23 23 22 23 23 24 24 6 22 22 22 22 22 23 23 23 24 23 25 24 7 22 22 22 22 23 24 23 23 24 24 25 24 8 22 22 24 23 24 25 23 24 24 25 26 26 9 22 23 24 24 24 25 24 24 25 25 26 27 10 23 24 26 24 25 28 25 26 26 27 27 28 11 22 24 27 22 24 30 25 25 28 26 28 28 Table of diameters of Vents- —Continued. Distance 272d Fire. 304th Fire. 336th Fire. 368th Fire. 40l3t Fire. 451st Fire. 520th Fire from Exterior. 334 335 334 335 334 335 334 335 334 334 334 l 22 23 24 24 24 24 25 25 26 29 31 2 24 23 24 24 25 25 26 26 27 30 31 3 24 24 25 25 25 26 26 26 28 31 33 4 24 24 25 25 26 26 27 27 30 32 35 5 25 24 26 25 27 20 29 27 30 32 36 6 26 24 26 25 28 26 30 28 31 33 35 7 27 25 27 26 30 28 31 30 32 34 35 8 27 27 27 28 30 30 33 32 33 36 38 9 27 30 28 31 30 33 33 34 34 34 40 10 28 30 29 31 30 32 34 34 84 37 42 11 29 28 30 28 33 31 34 32 38 42 42 Table of diameters of Vents — Continued. Distance 809th Fire. 885th Fire. 961st Fire. 1070ih Fire. 1280th Fire. 1360th Fire. 1440th Fire. 1520th Fire. 1600th Fire. Exterior. 2d vent. 334 2dvent. 334 2d vent. 334 2d vent. 334 2d vent. 334 2d vent. 334 2d vent. ■ 334 2d vent. 334 2d vent. 334 l 23 25 27 30 32 24 25 26 30 2 24 25 26 30 22 24 26 29 30 3 23 25 28 30 22 24 25 29 32 4 24 25 27 31 22 21 27 30 33 5 24 26 30 33 23 26 ■ 28 31 35 6 25 28 31 33 24 26 28 31 35 7 26 30 32 34 21 2S 32 36 40 8 26 30 33 37 25 28 32 34 37 9 27 32 36 40 26 30 33 36 42 10 26 34 38- 40 27 34 35 38 46 11 30 36 39 52 30 34 36 40 65 12 — — 30 — 29 33 37 39 70 AND OF CANNON POWDEE QUALITIES. 79 a s o 3 o « a* CO "3 ^ l e ' -** p p p p o o w | [ | i | | | [ i i | j | | | i | i | | | j i i to CO OS o fa H "C Pi ,— ^^-tCOGOOi-^TOHHCKKlH | | | | | | | I ) | J | ) I | | | 6 fe CO CO ■v. ^ i-i CN 7i 71 oi o i ci co oi co co co co co co o o 3 fa pi -H ^''M-fiOOOOiO^iMOHHIMMiMMOOO'i'-JiOClC'IOOCiOOOOOOO a CO ^rH,-HCNCIOqC17)7l00CMCOCOCOCOCOCO»O cOCOL-t-t^i>-b-L^COCOC/JCCCOaOCO o & CO £i fa 0) p-« P* ■e be 1 11 fa inn f isi&i ii § Mil- iff « -4 o -^ ~ -*€ -w o ,ra - , ^ 12 ^ S <3 St- o .-< o o o H-*H:i05Mt-ClHHOi:iH?lWH^MHHH01rt;iHnHHH'iHKClH CI o fa o of B s S5 £ 1 O^iHHHTKKlHHHHOOClOH-^LiaHHHC'lHClHHHWHOHH 3 o fa -4 t- • .^'iq'aq . .^^ ■fa3^>>:3 .^r^^fa^^fe-^a •^■.- .^^^ Q e eo co coo co co a co -3 co cococo t^roto cococo q CO CO CO CO CO CO to CO CO 'O TO O CO CO CO CO_ CO CO CO CO CO CO CO oi co c5 co -/j -fj 1-4 d -* d ^h -sr3 1- oi co 'O — 1 1-^ co ri co oi t- -n" co co co' t- s COCO^H^^^^COCOlOeOOTCOn-COCO^^COr^^lCO^COCOCOCOCO-^COCOOICOCO a H OOOOcccococot-cocococoior- — H n ra t- h cs ci ci 10 n cq 10 N co W tk ^ ^ ^ ^ : ; w * oo co O m ^ co N ^ co co :i rt ci co ^ co -i< co co co ■* :o :o a co m O oo H O — ' OT Ol OT -H -H -+ CO CO — ' 05 CO -H 71 O CO OT O CS O Ol O O CO CO O CC ~ »_0 71 r-( 71 -+( X (U ^^^^^^^-H-*iOl-CO-H-tHCO-^CO-*HCO^-iCO-t , -^-H-HCT J CO-^-^C7COCO-H-rl Ol ^ O 71 — X) OJ ~ O CO CO 1 -^ CO -Tj CO — ' — < -H t— 71 — CO CO ^-1 -4 CO C7 CO CO -H -^ r*4 CO CO -t< JO CO CO ■* CO CO CO CO -ri 1— 1 -— * ~i -* CO 75 CO CO CO CO CO 71 71 71 CO -*1 33 t- o H CO "0 CO CO CO CO CO -O CO CO CO CO CO O CO CO CO' CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO CO C7 CO CO CO CO CO CO CO CO cr CO CO CO 3 CD 7-1 C7 CI CO '. 1 CO CO 71 71 71 CO t- CO 71 -+ rrJ 3 7i O t^ l-^ CO OH 35 CO CO 'O CO O O 7 1 CO i-O CO ^^^^^^^^^lOCOCO^^COCO^^COi^OICO-^COCOCOCOCO-HCOCO i-l C; CO r-l ■* IO t> CO CO - 'COCOCO p ^^^^^ -^ 07j ^ ^ tj^ Q-> iO b- CO &5 O ^ ^ ^ ^ JO iO O O O l> tH O ^ CO i< i* ^ CO « i< t)< ■* ■* ■* ■* CO ■* O CO CO CO •* -t< H CI a c^MO^OCTomcOHOCT«cocococ^H(OOOOioONcoci«coiOrH^cr;ci -3 CO^^^^^^^CO^WCWCO^COWCO^^nHrM01^^71C0'01COCO«CO71OlcO 5 -s ^ cf t -" co' ~r ~" -J" o "T co' rH o" l-T h of co" -+ co" t-T co" ccT to" t^co&rJ<&cQTitbZ'a$ to determine the value of (c), let u = o, and x = l, (since when x = 1, the sum of the strains = o), and we find c = a, and u = \- a ; and inte- grating between the limits x = 1, and x = 3, we have u = a — - = - a; or, o o the effective resistance of a gun one calibre in thickness will be two-thirds (|) of that which half a calibre in thickness would offer, if the strains were all equal as in the tensile strain. The above results suppose the gun to be entirely free from strain before the application of the straining force, a condition which could only be obtained by allowing it an infinite length of time to cool. For it never could cool without allowing the surface to fall to a tempera- ture below that of the interior of the metal ; and, since iron diminishes in bulk as its temperature decreases, it follows that a gun cast solid, and cooled, AND OF CANNON POWDER QUALITIES. 95 of necessity, from the exterior and in a limited time, will be thrown upon a strain — the exterior being under a force of compression, while the interior is under one of elongation ; and the greater the difference of temperature between the exterior and interior, during the process of cooling, or the more rapidly the cooling is effected, the greater will be this strain. Now, let us suppose a central force applied to a gun thus strained, and bear in mind the law of strain, as above demonstrated : the interior is already under a force of elongation, while the exterior is under one of compression : the action of the central force develops (in a gun, one calibre thick) nine times the strain on the interior that it does on the exterior, independent of previous strain ; so that we have the effort of the exterior to free itself from the force of compression to which it has been subjected in cooling, combined with the central force, to break the interior ; and the wonder is, that guns thus strained, in cooling, endure as long as they have been found to do. It is also known that the contraction of iron, under equal degrees of reduction in temperature, increases as the iron approaches a state of purity ; or, as the percentage of carbon diminishes ; so that what is termed very high iron cannot be cast into shapes — where the times of cooling of the different parts are unequal — with any certainty of enduring the strain to which it is thus subjected, and large rolls for rolling iron have been known to split open, longitudinally, from this cause. Now the iron of which the experimental guns were made was what is termed high iron, all of it having been melted once and run out into pigs, and a portion of these pigs again re-melted, before being melted for- casting the guns ; this degree of purity having been found to produce the greatest tensile strength attainable with this iron, without risk of cavities in the castings. This iron contracted more, in cooling, than that which we have been in the habit of using at this foundry ; as shown by the fact, that the 8-inch guns cast from it, and off the same pattern as heretofore used, were from .10 to .15 of an inch less in diameter than those from other iron. The small endurance of the 8-inch solid gun can only be explained or accounted for by reference to the above stated laws, deductions, and facts. These, to my mind, do satisfactorily account for the early bursting of the solid gun, and for the greater endurance of the hollow one ; for the same 96 PEOPEETIES OF METALS FOE CANNON, causes which contributed to the bursting of one, being reversed in effect, con- tributed to the endurance of the other. The solid gun, cooling from the exterior in an open pit, and being of iron that contracted very much in cooling, was doubtless thrown upon a very- heavy strain, the exterior being compressed, and the interior elongated ; while the hollow gun, being rapidly cooled from the interior, and prevented from cooling from the exterior, was thrown upon a strain just the reverse of the solid one. And I have no doubt that the interior of the solid gun was broken before the exterior was relieved from the pressure to which it had been subjected in cooling ; while in the hollow gun, the great object of my improvement was in part, if not fully, attained, — viz., to throw the gun upon a strain, such, that under the action of the law of strain, as stated above, each one of the infinitely thin cylinders composing the thickness of the gun shall be brought to the breaking strain at the same instant. This condition would give us for the effective resistance to ruptiire, in a gun 1 calibre thick, 2 a, instead of f a — which has been shown to be all that could be obtained from a gun free from strain by cooling — and is, doubtless, much more than can possibly be attained in practice. The higher the metal, and the greater its contraction in cooling, and the more rapidly the gun is cooled, the further will the solid one fall below § a, and the more nearly will the hollow one approach 2 a, provided the cooling be effected from the interior ; also, the greater the diameter of the solid gun, the greater will be the strain from cooling. It is not considered practicable to cool a gun so rapidly from the interior as to cause rupture to commence on the exterior. The less endurance of the 10-inch hollow gun than that of the 8-inch hollow one, is accounted for by the fact, that the 10-inch gun had no fire on the exterior of the flask while cooling, it having been rammed up in the pit, where it was supposed, at the time of casting, that the heat of the gun would be retained by the sand, until the interior would be cooled by the circulation of water through the core barrel. This supposition was found to be erroneous on digging out the sand, as its temperature was found to be much lower than had been expected. The less endurance of the 10-inch solid gun than that of the 8-inch solid gun, is attributable to the increased strain to which it was subjected in cooling, arising from its increased diameter ; and to the greater pressure of AND OF CANNON POWDEE QUALITIES. 97 the gas in firing, which is, directly, as the diameter of the bore; since the weight of the shot increases as the cube of the diameter, while the area of its great circle, or the surface pressed, only increases as the square of the diameter. The 10-inch guns were both rammed up in the same pit for casting. They were thus treated on account of the smallness of the flasks, it having been considered unsafe to cast without this precaution ; as it was feared that the great body of heat in these guns would strike through the small thickness of sand, and weaken the flasks so much as to cause them to give way, and let the metal out of the moulds. The flask being rammed up in the pit, of course rendered it impossible to apply heat to the exterior, during the process of cooling. So that I do not consider my mode of cooling to have been properly applied in this case (although is was intended to be, at the time of casting), nor do I consider the endurance of this gun as any criterion, or measure, of what can be attained by a proper application of the new mode of cooling ; though it certainly does show that, as far as this mode was applied, it was beneficial, as the ratio of endurance of the pair is about 12J to 1 in its favor. I have the utmost confidence that with iron of the same quality as that used in these guns, and a proper application of the new mode, a 10-inch gun may be made to endure from 1000 to 1250 fires. I am, very respectfully, Sir, Your obedient servant, T. J. RODMAN. 28 REPORT OF THE FABRICATION AND PROOF, UP TO 2450 SERYICE CHARGES EACH, OF 2 10-INCH TRIAL GUNS; ONE CAST SOLID, AND COOLED FROM THE EXTERIOR, AND THE OTHER CAST HOLLOW, AND COOLED FKOM THE INTERIOR. EEPOET OF THE FABBICATION AND PROOF, UP TO 2450 SERVICE CHARGES EACH, OF 2 10-INCH TRIAL GUNS; ONE CAST SOLID, AND COOLED FROM THE EXTERIOR, AND THE OTHER CAST HOLLOW, AND COOLED FROM THE INTERIOR. These guns were cast at the Fort Pitt Foundry, from the remainder of the iron selected by Lieut. F. J. Shunk and myself, on the 2d and 3d of July, 1857, at the West Point Foundry, and sent to the Fort Pitt Foundry. They were cast on the 11th of August, 1858, one (No. 362) hollow, and the other (No. 363) solid. Of the Iron. The iron was melted in the same furnaces which were used in casting the first pair of 10-inch guns from the selected iron, in August, 1857, and used in the same proportion, and similarly distributed among the three furnaces, the marks on every pig being still perfectly distinct. Charges and Distribution of Metal. *USED. QUANTITY OF D20N CHARGED. KIND OF IKOj Furnace No. L Furnace No. 2. Furnace No. 3. Total charged. " " No. 2, 3685 lbs. 4532 " 1870 " 2047 " 409 " 1045 " 4819 lbs. 5926 " 2445 " 2677 " 535 " 1368 " 5670 lbs. 6972 " 2877 " 3150 " 630 " 1610 " 14174 lbs. 17430 " " " No. 3, 7192 " 7874 " Sootcb " 1574 " Ee-melted in the above propon ion, . 4023 " Total charge, 13588 lbs. 17770 lbs. 20909 lbs. 52267 lbs. 102 PEOPEETIES OF METALS FOE CANNON, Condition of Pits. The pits in which the guns were cast were in good condition, having had a strong fire burning in them from the 3d to the 7th inst, previous to casting; they were cleaned out on the 9 th, and fresh fuel was placed on the grate bars, preparatory to casting. Preparation of Moulds. The moulds in which these guns were cast were formed in the same flasks in which the guns from the same iron of the previous year were cast ; that for the solid gun was prepared on the 6th, and placed in the pit on the 9th ; that for the hollow gun was prepared on the 9th, and placed in the pit on the 10th. The furnaces were charged on the 10th, each furnace receiving the differ- ent kinds of iron in the same proportion as for casting the 10-inch trial guns, Nos. 334 and 335, of 1857. Casting. The furnaces were all lighted at 9h. 10m. A. M., on the 11th August, 1858. The metal in all the furnaces was melted at lh. 30m. P. M. The fur- naces were tapped, No. 3 at 3h. 2m., No. 2 at 3h. 7m., and No. 1 at 3h. 12m. P. M. The metal from all the furnaces was received in the same vessel, which was furnished with two orifices, from which the flow of the metal could be regulated at will. The metal flowed from both orifices at the same time, and was conducted by runners into the gun moulds, which were filled simul- taneously in 12 minutes after the first metal entered. The metal flowed from only one furnace at the same time. The pits were closely covered with iron covers, and the fuel in that which contained the mould for the solid gun ignited while casting, and that in the hollow gun-pit was fired immediately after casting. The metal entered the moulds by side runners till they were filled above the trunnions, then directly at their mouths, till filled. AND OF CANNON POWDEE QUALITIES. 103 Cooling. Water circulated through the core barrel of the hollow gun at the rate of about three cubic feet per minute, from the time the metal reached the bot- tom of the core in casting, till the casting was cold. At 20 minutes after casting, water entered at 80° and left at 102°. Cooling Table. HOTJBg AFTEB CASTING. Left at 2, 5, 7, 9, 11, 13, 15, 16, 100° 96 94 93 92 92 91 90 At 16 hours after casting the core barrel was removed, and water, at the same rate as before, circulated through the cavity thus left. Just after the removal of the core barrel, water left at 144°. Cooling Table — Continued. Houbs aftbk Casting. Entered at Left at Houbs aftee Casting. Entered at Left at 17 19 21 23, 27 29, 31 33 35 37, 80° 80 80 80 80 80 80 80 80 80 80 144° 134 108 105 103 99 97 96 94 93 92 39 43 45, 53 57 61, 63, 73 75 81, 80° 80 80 80 80 80 80 80 80 80 80 92° 92 91 90 89 87 86 85 84 83 81 Water continued to circulate through the hollow cast gun till 9h. 30m. A. M., on the 19th, at which time it entered and left at the same temperature. 104 PKOPEKTIES OF METALS FOE CANNON, Temperature of the Pits. HOUBS AFTEE CASTINO. Temperature of Solid Pit. Temperature of Hollow Pit. 2 6 19 40 75 125, 155 160° 350 460 472 360 150 125 350° 600 About 900 " 900 315 100 96 No fuel was added to the solid gun-pit after casting ; and none to the hol- low gun-pit after 46 hours after casting ; from which time the fire gradually went down. The guns were both entirely cold before they were removed from the pits. The hollow gun was first removed from the pit, and was sufficiently cold for removal at least two days before it would have been proper to remove the solid cast gun. Both moulds were formed on the same pattern ; yet the hollow cast gun was between. 3 and. 4 inches less in diameter at the breech than the solid cast gun. The difference in the appearance of the newly turned surfaces of the two guns was quite marked ; the hollow gun presenting a rather fine and very regularly mottled surface, while the solid gun presented a clouded or dappled appearance. The original programme, of which the trial of these guns is a part, con- templated that, at the same time this pair of guns was cast, the proprietor of the West Point Foundry would have cast, from the metal selected and left there for that purpose, a solid 10-inch gun ; but at the time the order for casting these guns was under consideration, the proprietor of the West Point Foundry advised the Department and myself that a portion of the iron selected had been used, and that, in his judgment, it would be going against the light of experience to make another gun of the same iron which had been used in that of the previous year. The programme has consequently not been followed. These guns were accurately measured after completion, and found to be within the prescribed limits of variation in all their dimensions, except the AND OF CANNON POWDEB QUALITIES. 105 trunnions, which were not turned, owing to the fact that the trunnion lathe at the Fort Pitt Foundry had been so much injured by the then recent fire that it could not be put in working order in time to secure the proof of the guns during the then coming fall. No defect was discoverable on any part of either gun. The proof charges were fired from these guns on the 2nd of October, when they were suspended for firing with service charges. The proof, with service charges, was continued up to the 24th of December, 1858. The guns were fired alternately with charges of the same eprouvette range, each barrel of powder used being one-half fired from one gun, and the other half from the other gun. Solid shot were used in all cases except one proof charge, and were rejected as soon as found to pass through the small gauge. Proof Charges. 1st fire, 20 lbs. powder, one shot and sabot, and one wad. 2d fire, 24 lbs. powder, one shell and sabot. The shells used in proof were neither of them perceptibly injured, although fired into a bank of moderately hard, shaly rock. Service Charges. The service charges used were 14 lbs. of powder, one solid shot and sabot ; and of these charges 2396 rounds have been fired from each gun. Also, 13 rounds with 18 lbs. powder, 15 " "17 « " 14 " " 16 " " And 12 " " 15 " " making a Total of 2450 rounds, exclusive of proof charges. Of the Powder. There have been used in the proof of these guns, 142 barrels of Dupont's make of 1856, And 95 " " " " " 1857. The other powder used was all of Dupont's make, but of different dates, from 1837 to 1846, and received at this Arsenal in 1852 from the St. Louis Arsenal. 27 106 PKOPEBTIES OE METALS FOE CANNON, The proof charges Were from the 1857 powder, and from the small enlargement of the guns after proof, and from the low eprouvette range of this powder when tested in the old eprouvette mortars at this Arsenal, also, from the fact that some barrels were found to be badly caked, it was appre- hended that the powder had deteriorated since last year. In order to test this point, powder of the same make and same date (1857), but stored, some at Dupont's mills and some at the New York Arsenal was sent here for comparison. These powders have all been compared in both the old eprouvette and in a new one recently received at this Arsenal ; and for pressure, velocity, and recoil in the 42-pdr. gun, with the results recorded in the following table : — Euros os Powdee. No. of Fires. Mean Telocity. Mean pressure of gas, per square inch. ' Mean recoil of gun. Range by old Range by new Eprouvette. Eprouvette. Dupont's, 1857, stored at New York Arsenal. 2 5 4 1370 ft. 40546 lbs. 22° 31' 230 yds. 288 yds. Dupont's, 1857, stored at Dupont's Mills. 4 2 5 1360 ft. 40863 lbs. 22° 37' 238 yds. 289 yds. Dupont's, 1846, received from St. Louis Arsenal in 1852. 2 4 5 1340 ft. 38818 lbs. 22° 08' 256 yds. 304 yds. Dupont's, 1857, stored at Alleghany Arsenal. 3 5 2 4 1322 ft. 40740 lbs. 22° 28' 237 yds. 273 yds. Dupont's, 1856, stored at Alleghany Arsenal. 3 5 2 1321 ft. 36244 lbs. 22° 33' 222 yds. 290 yds. The charges of powder used in the 42-pdr., giving the above results, were 10 lbs. each, and one solid shot without sabot. Pressures were determined by the method of indentations, and recoils were measured on an arc of 26' feet radius. The velocities recorded in the foregoing table were determined by means of M. Navez' electro-ballistic pendulum. One wire passed just in front of the muzzle, and the other at 60 feet from it. The following table compares the ranges of the same barrel, of the different AND OE CANNON POWDER QUALITIES. 107 kinds of powder, obtained from the new eprouvette, with those marked on the barrels, at the time the powder was received. Knros of Powdeb. Ranges marked on bbls. Ranges by new Eprouvette. Dupont's, 1856, .... 1857, .... " " stored at mills, . " " stored at New York, . 1846, stored at St. Louis, . 312 yards. 299 269 " 299 " 266 " 290 yards. 273 " 289 " 288 " 304 " The range marked on the barrel of 1846 powder, stored at St. Louis, is believed to be that obtained in its proof at this Arsenal just after its recep- tion from the St. Louis Arsenal in 1852. It appears from the above table, either that the eprouvettes contradict each other, or that the powder of 1857, stored at Dupont's mill, has improved in strength, while that of the same date and make, stored at the New York and Alleghany Arsenal has deteriorated since the original proof. It also appears that the oldest powder stored, since its reception at this Arsenal, in the same magazine with the powder of 1857, is the strongest powder of all. The following velocities and recoils were obtained from the 10-inch solid gun with the same instrument, and are for measuring recoil, as used in ob- taining the results recorded in the foregoing table. One solid shot and sabot were used in these trials, at each fire. Powder used, — Dupont's, 1846; received from St. Louis Arsenal. Number of Fires. Weight of Charge. Mean Velocity. Mean Recoil. 4 5 14 lbs. 940 feet. 24° 27' 5 4 15 lbs. 980 feet. 25° 09' 5 16 lbs. 971 feet. 26° 11' 4 5 17 lbs. 1022 feet. 26° 48' 5 18 lbs. 1044 feet. 26° 57' 108 PEOPEETIES OF. METALS EOE CANNON, Difference in Velocity of Shot due to Different Amounts of Enlargement in the two Guns. In order to determine the difference, if any, in velocity of shot and recoil of gun, due to the greater enlargement, by firing, of the bore of the solid than of that of the hollow gun around the seat of the charge, a sufficient quantity of St. Louis powder for 20 cartridges of 14 lbs. each, was well mixed together and made into cartridges. These cartridges were fired, 10 from each gun, with one solid shot and sabot. The velocities and recoils of the first five fired from the solid gun, and of the last five fired from the hollow gun, were determined, and are recorded in the following table : — NUMBER OF FIRES. VELOCITIES. EECOILS. Hollow. Solid. Hollow. Solid. Hollow. Solid. 1 2 1 2 3 4 5 1063 1006 958 «1326 24° 14' 24 22 24° 39' 24 30 3 4 5 872 927 1106 998 788 1016 25° 04' 25 02 24 53 24° 05' 24 23 24 40 Mean, 995 968 940 934 24° 43' 25 00 24° 27' 24 24 Mean of last rounds, 15 This result is evidently erroneous, and is neglected. The last three rounds, from the hollow gun, were made with the same shot with which the five rounds, for velocities and recoils, were made from the solid gun. The mean of the last three rounds is therefore regarded as more nearly correct than the mean of the whole five rounds, as regards the informa- tion sought. Either mean, however, leaves no doubt of the fact, that both velocity of shot and recoil of gun are considerably less in the solid than in the hollow gun, and this difference can be attributed to nothing but the greater windage, from wear and enlargement in firing, in the solid than in the hollow gun. All the firing, for velocities and recoils, except the five rounds with the AND OF CANNON POWDEE QUALITIES. 109 hollow gun, whose results are recorded in the foregoing table, was made with the solid gun. In order further to test the regularity of results obtained from the electro- ballistic pendulum, and from the hollow gun, eight rounds more of 14 lbs., one shot and sabot, were fired from that gun, with the results recorded in the following table : — Numbbb of Fires. Weight of Charges. Velocity of Shot. Kecoil. 1 2 3 4 5 6 7 8 14 lbs. COOTHCT(^THlCCOODOSOHC-lCOincoi>QOOSO i^r^i^i^T^r^i^i^C^^C^C^C^G^C^CqCSICO AND OF CANNON POWDER QUALITIES. 123 CO CO J 4 J 3 .s S 3 9S9 aaoae^-S.-g ,£>,£?,£>£>£»,£» Js>J£? i»±>i?^> ^.£_£_b±> Si % ^ ^ ^ **> ^3 T3T3'0 ^13 .T3^3 , . r C r r3'^'i3T3'T3'T3 p z3T3° J -' ^ ,J ^3 r-; r- r; Sg^033h!33!-.tH!33!33303Cl03SSS3S MFHrtrtfHMriMtH^'S ° _C3 -" C Q ° OQOOOOhQO^OOOOOOQOOODOOOO GO o o HNHINNCOMIMHHWIMmTllCqcOMHHmCilHCOHHn o s 3 CO CO QQ JZi 02 02 CO 3 3 GO GO GQ GQ ■< o 01MOO<— iNIMh^ iOONHMOO)«HHMNNr i-hi-h-^ a o © ft} S H H a, en M H B o H O W i 3 COCDCO eo co co co coco CO CO CO CO 00 CO CO CO OCDCO ^ CO CO CO CO COCO COCOCOCO CO coco CO ^WlO^^^^WCOCO^COWCOCOCONC<|COCOCOCOCOCOCO(N B H a a o a PS H a H • COCOCOCO CO CO CO COM COCO CD cococococococoo COCOCOCO CO CO CO cococococo cocococococococo r^O^rHlXl^Nr^HdNCOWHCO'^'QCOOCO^TlIcOCO^d W^IOO'^'^'^^^^'^COCOCOCOCOWNCOCOCOCOCOCOCOCO i s- 2 « HNM^OCOCOaiOHWCOlOCDScOOONCO^lOCOSOJO 124 PEOPEKTIES OE METALS POE CANNON, W a EH -a) H n3 a ^ *H c3 -a ±> fan fan^ -1 JfT» S cs cs'g CD ^ . .02^ . . . . . 1Zh0202 i^i £i £i 02 02 CO 02' £i £, 02 t»!zi|Zi02|2it2i^icO I— o OS CO CO -* CM CO T— < 00 in r^ 10 1^ CO CO _, CM CO CM 10 CO -* CM l-H CI CM ■* TO •* CO CO CM ■* ■* CO CO CM CM CO CO CO CO ^ CO CO CD CO CO CO 00 CO CO CO CO CO CO CO CO en CO 00 CO CO 00 CO CO CO CO CO CO CO CO CO CO 00 CO CO CO CO CO CO CO 00 CO CO Oi OJ CD CI 00 CO -* O CO r-H l-~ CM (^ t- ICO O} 0- 10 1^ en CO 10 1^ ■* CM CO -^ *o CO -f CO 1— 1 CM CO -* lO-^COWCO^^COCOtNCM^CQCOCO^ i-cCMCO^COt-OOCCi i-HrHr-li-lrHrHi-li-iCMCMCM(M H . O CO £cg O +3 Ph cq CO « H r-f eg CO „ SI 'Sol a fl d 3 rid & CD CD O O co co . St co ^ ■43 *-• co co -*3 Ti Hp W H CO 13 S S co ifl *li -*3 co rH rH rOH Sr3 fa "S rH .- " f~, H 3 u hi -*S -r» .3,5 r- P 3 a d O O PhPh WrZi PPffl m m WPQ tern ffl ££ n H « ^3 S3 1 CO I 1 1 IQ^ I <© I °5 CS 1 rQ - 1 n3 O O O OO COCO 000 COO CO CO H4 •* -H ■* -* F4 i-h i-i r-i r-i rH rH 1— 1 rH " rt rH rH rH rH rH 0) R-I rEj* rd rO O - rrj 3 =3 .a 04 c) CO 00 CO CO OO cq cq 00 OO O OO rH 1— 1 CO CO rH rH rH r-t rH rHrH PM • g £ rO 3 rH Cq co^ IO CO O rH rH cq -*iO CO cq co a a CD CD CO CO CO CO CO CD CD rH rH CO CO CO CO OS CO CO g Cast roved. OS OS 1— I rH rHi—t cq cq CD CD t- b- b- CO CO -# HH 10 w lO to tn >o lO 10 lO lO iO 10 10 00 CO COCO COCO CO CO CO CO CO CO CO CO CO i-l r— I 1-1 rH rH rH rH rH rH rH rH rH rH rHrH r=l 13 gt ed 134 PEOPEETIES OF METALS EOE CANNON, Those numbers or pairs of guns included in the same brackets in the above table were cast at the same time, and from the same pool of melted iron ; one being cast solid and cooled from the exterior, and the other cast hollow and cooled from the interior. The results recorded in the above table show the hollow cast gun in every pair (in which one or both guns have been broken) to be superior, in endurance, to the solid cast gun. The results may be grouped as follows, viz. : — Of the two pairs of 8-inch guns, the solid cast guns endured 158 rounds — both burst. Hollow cast guns endured 1751 rounds — one gun unbroken. Of the two pairs of 10-inch guns, fired with 18 lbs. charges, those cast solid were fired 46 rounds — both burst. Those cast hollow were fired 564 rounds — both burst. Of the two pairs of 10 -inch guns, fired with 14 lbs. charges, those cast solid have been fired 2851 rounds — one gun unbroken. Those cast hollow have been fired 4052 rounds — neither gun broken ! Of the pair of 32-pounders, the hollow cast gun was superior, but the comparison was destroyed by double charging the hollow cast gun. Of the three pairs of 10-inch, and two pairs of 8 -inch guns, when one or both of each pair has been broken, those cast solid have been fired 603 rounds — guns all broke. Those cast hollow have been fired 3915 rounds, and two guns unbroken. Out of the five 10-inch guns, made of the same iron (that selected at the West Point foundry), two good hollow cast guns have been made,- one having been fired 1600 rounds, and the other 2452, and neither broken; while of the three solid cast guns two have burst, one at the 169th, and the other at the 399th fire ; the other has been fired 2452 rounds — unbroken — and is the only good solid cast gun out of the seven experimental Columbiads cast in that manner. It should likewise be borne in mind that these results have been obtained under the most unfavorable circumstances for the hollow made ; the only experience in that mode of casting being that furnished in casting the experimental guns ; while the solid mode has had the benefit of long experience, in both Europe and this country, and the iron used in casting the experimental guns has always been selected with a view to making the best solid cast gun possible. AND QUALITIES OF CANNON POWDER. 135 The 32-pounders, and all but one of the 10-inch solid cast guns, have had the benefit, as far as it was possible to apply it, of the new mode of cooling, having been retarded in their rates of cooling by having fires burning in the pits while cooling, which is not the ordinary mode of cooling solid cast guns. EE CAPITULATION. Casting. Guns of 1857 and 1858, for testing merits of solid and hollow modes of cooling, were cast from the same iron, and under similar circumstances. Mechanical Tests. The hollow cast gun, in both pairs, was superior to that cast solid, in both tenacity and density. Model of Guns. The guns of 1857 were made with, and those of 1858 without chambers, with increased thickness of metal in the breech. Proof. The guns of 1857 and 1858 were proved in the same manner, and with the same charges of powder and shot. Powder Used. The powder used in the proof of the guns of 1857 differed but little in quality from that used in the proof of the guns of 1858. Enlargement. The guns of 1857 were enlarged more by the same number of rounds than those of 1858. Deterioration. The solid cast gun of 1858 appears to be greatly more deteriorated from firing than the hollow cast gun, cast at the same time. Comparison of Solid Guns of 1857. The metal in the West Point gun was superior in quality to that of the 136 PEOPEETIES OE METALS EOE CANNON, Fort Pitt gun, yet the latter endured more than double as many rounds as the former. Strain from Contraction in Cooling. The West Point gun of 1857 was subjected, by contraction in cooling, to about three-fourths, and the Port Pitt solid gun, of same year, to about one- third of its ultimate strength. Working of Furnaces. The fort Pitt Furnaces melt quicker, and " work lower," or decarbonize their iron less, than the West Point furnaces. Comparison of Fort Pitt Guns of 1857. The difference in endurance of these guns can be attributed to no other cause than the difference in their mode of cooling. Solid cast gun burst at 399th fire ; hollow cast gun has been fired 1600 times, and is unbroken. Comparison of Fort Pitt Guns of 1857 and 1858. There is no other known cause for difference in endurance of these pairs of guns, than the difference in their model. Accidental or unknown causes have produced as great differences in endurance. Iron for Guns. Solid cast guns must be made of inferior iron to prevent fatal contraction in cooling ; while the hollow method enables us to use the best quality of iron brought to its maximum capacity for work. Endurance. In the six pairs of experimental Columbiads that have been proved, or partly proved, the hollow cast has shown a marked superiority in every instance. 2 Pairs 8-inch Columbiads. Solid guns fired 158 rounds — both burst. Hollow guns fired 1751 rounds — one gun unbroken. AND QUALITIES OF CANNON POWDEE. 137 3 Pairs of 10-inch Columbiads, of which one or both Guns have broken. Solid guns fired 445 times — guns all broken. Hollow guns fired 2164 times — one gun unbroken. One pair of 10-inch guns remains unbroken, each gun having been fired 2452 rounds. Ratio of Good and Bad Guns. Out of seven solid cast experimental Columbiads, one gun only proved to be good. Out of six hollow cast Columbiads, three were good, having been fired 1500, 1600, and 2452 rounds respectively, and neither gun broken. It is not deemed out of place here, in order to show the necessity of further investigations into the properties of cast iron, in its application to the manu- facture of cannon, to notice some facts in the history of gun foundering in this country since 1849. The very low endurance of the first pair (8-inch) of experimental guns which were cast in that year, was attributed to the inferior quality of the iron of which they were made. Two years were spent in searching after a better quality of iron, which was undoubtedly found ; and in 1851 another pair of 8-inch guns was cast. The iron in this pair of guns had a tenacity of near 38000 lbs. ; while that of the iron in the first pair was only between 27000 and 28000 lbs. The solid cast gun of the first pair burst at the 85th fire, and that of the second pair at the 73rd fire ; the superior iron giving the inferior solid cast gun. These results did not, however, destroy confidence in strong iron for solid cast guns, and the first pair of 10-inch guns was made from the same lot of iron; and, with a tenacity of iron of 37000 lbs., the solid cast gun burst at the 20th fire. This result weakened confidence in very strong iron, and the tenacity was reduced. In 1857, after guns of good tenacity had failed at the Fort Pitt, South Boston, and West Point foundries, four out of seven guns offered for inspec- tion at the last named foundry having burst in the proof, Mr. Parrott, proprietor of the West Point foundry, one of our most experienced gun founders, cast his trial contract guns of iron, having a tenacity of 30000 to 138 PEOPEETIES OF METALS FOE CANNON, 32000 lbs. One of these guns has endured 1000 service charges of 14 lbs. powder (800 rounds with shells, and 200 with shot). The iron selected at that foundry, and from which the last five experimental guns have been made, was of the same quality, and in the same proportions, as in the guns last above referred to. In 1858, after the failure, at the 169th fire, of the West Point experimental gun made from this iron, Mr. Parrott condemned it as being too high for heavy guns. From this rejected iron was made the last pair, Nos. 362 and 363, of trial 10-inch guns, at the Fort Pitt foundry, which have been fired 2452 rounds each ; the least charges fired being 14 lbs. powder and one solid shot, and neither gun broke. These guns have since been fired one thousand rounds each, with 18 lbs. of powder and solid shot, and neither gun yet broken. It should also be borne in mind that the proprietors of the West Point foun- dry have control of the smelting furnace at which their gun iron is made ; they ought, consequently, to have a more perfect knowledge of the qualities and properties of their iron, than those founders who are dependent upon the market for their gun iron. These facts, to my mind, are conclusive as to the fact that we are at present far from possessing a practical knowledge of the properties of cast iron in its application to gun foundering ; and it is too much to expect of private enterprise to take up and prosecute so intricate and expensive an inquiry. The subject is one of national importance ; and until the Government shall obtain control of a smelting furnace and a foundry for casting cannon, it is, in my judgment, in vain to look for any marked improvement on our present knowledge of this subject. T. J. RODMAN, Capt. of Ordnance. Having witnessed the operations of casting the trial guns cast at the Fort Pitt foundry, — two on the 13th August, 1857, and two on the 11th August, 1858, respectively, — I concur in statement of facts relative to the mode of conducting these operations, as made by Capt. Rodman, in the foregoing Report. S. CRISPIN, Lieut, of Ordnance. REPORT OF EXPERIMENTS MADE BY CAPT. T. J. RODMAN, U. S. ORDNANCE DEPARTMENT, IN THE YEAES 1857 & 1858, FOR DETERMINING THE PROPERTIES OF GUN METAL, AND THE ACTUAL PRESSURE PER SQUARE INCH DUE TO DIFFERENT WEIGHTS OF POWDER AND SHOT. REPORT EXPERIMENTS MADE AT ALLEGHANY ARSENAL, BY CAPTAIN T. J. RODMAN, U. S. ORDNANCE DEPARTMENT, IN THE YEARS 1857 AND 1858, EOR DETERMINING THE PROPERTIES OF GUN METAL, THE RESISTANCE WHICH GUNS CAN OFFER TO A BURSTING FORCE, THE ACTUAL PRESSURE PER SQUARE INCH DUE TO DIFFERENT •WEIGHTS OF POWDER AND SHOT, ETC., ETC. Transverse Resistance of Guns. For the purpose of determining data which, enter the general formula for the bursting tendencies of different weights of powder and shot on guns of different calibre and thickness, the following experiments were made, viz. : — Nine bars of iron, 22 inches long, were cast on their ends in dry sand moulds, and from the same ladle of metal. Six of these bars were similar, and equivalent in cross section to the staves which would be cut from a 6-pdr. gun, one calibre thick, by planes through its axis, their thin or inner edges being 1.125 inches thick. These bars were dressed on their sides and on their thin edges, their outer edges having the same curvature as that of a 6-pdr. gun, one calibre thick. The other three bars were of rectangular cross section, breadth, b = 2.24 in., and depth, d= 3.45 in., and dressed on all their sides ; they were all broken transversely — three of the stave-shaped bars with their thin edges up, and the other three with these edges down. The distance between the outer bearings was 20 inches, and the breaking force was applied at the middle point of this distance, and acted in a vertical direction, upward. 142 PEOPEETIES OF METALS FOE CANNON, The mean breaking weight of the three broken with their thin edges down was 39686 lbs. ; while that of the three broken with their thin edges up was 35433 lbs. The mean breaking weight of the three rectangular bars was 38456 lbs. The transverse strength of this iron, as determined from the mean breaking L W weight of the three rectangular bars, by the formula S= , 2 > is 7212 lbs. ; and, knowing this element, we can determine the breadth of a bar, of rec- tangular cross section, and depth equal to that of the stave-shaped bars, whose transverse resistance is equal to that of the stave-shaped bars with either edge down. This has been found to = 2.099 inches with the thin edges down. A stave of the same length and same iron as above, one inch thick on their edge and sides in direction of radii, would bear eight-ninths of the weight borne by these bars. And since in the gun the staves are fixed at their extremities, and have the pressure uniformly distributed along their whole length, the total resistance which a single stave, one inch thick on its inner edge, could offer, would be= | (39686 x 2 + 35433) = 102050 lbs. And the breadth of a bar of same length and depth, and of rectangular cross section, which would offer an equal resistance, would result from the formula for the transverse resistance of a bar fixed at its extremities, and having its load uniformly distributed along its whole length, viz., b= 2 • in which b= breadth of bar, d= depth, L = length, TF=load, and S'= transverse strength of the material of which the bar is made. The breadth of a bar which would equal in depth and transverse resistance a stave one inch thick on inner edge, cut from a gun one calibre thick, as determined by the above formula, has been found, by computation, to be 1.8 in. The metal from which these results were obtained was quite soft. Harder metal, it is believed, would have given a greater difference in the resistance of the stave-shaped bars, due to a reversion of position of broad and narrow edges, and this would have increased the breadth of the equivalent rectan- gular bar; so that it is believed that, for gun metal, two inches would be nearer the truth than 1.8, and the following law would result, viz. : The total transverse resistance which a stave of a gun can offer is equal to that which a bar of rectangular cross section, and same metal, length, and depth, and whose breadth = mean of inner and outer edges, would offer AND QUALITIES OF CANNON POWDEK. 143 when fixed at its extremities, and having its load uniformly distributed along its whole length. Deflection of Bars under Loads equally distributed along their Whole Lengths. For the purpose of determining the relation between the length and the deflection or sag of bars of rectangular cross section, and of uniform breadth and depth, when supported at their extremities in a horizontal position, and having their loads uniformly distributed along their whole length, the follow- ing experiments were made, viz. : — A slip of white pine, well seasoned and straight grained, was accurately dressed to a uniform breadth and depth along its whole length, which was, at first, ten feet. This slip was placed in a horizontal position, its length between its bearings being 10 feet, and its sag or deflection at its middle point accurately measured ; it was then turned over and placed on the same bearings, and its sag again measured. A mean of these measurements was taken as the true sag. One foot in length was then cut from each end, and the sag again measured in the same manner as before. This process was continued till the slip was reduced to two feet in length. The deflections thus determined are as follows, viz. : — For a length of 10 feet, sag= 6.595 inches. « " " « 8 " " =2.725 " " " " " 6 " " = .898 " a a a a A u a 299 " u u u ii 9 it k _ 015 " These deflections are very nearly proportional to the fourth powers of the lengths, and accord very closely 'with the theory on this subject, as given in Boucharlat's Mechanics. Transverse Resistance of Hollow Cylinders. For the purpose of determining the value of the transverse resistance offered by hollow cylinders to a bursting force, acting upon different lengths of bore, the following experiments were made, viz. : — 144 PEOPEETIES OF METALS EOE CANNON, Three solid cylinders of iron, 4 inches diameter and 8 feet long, were cast on their ends in dry sand at the same time, being poured from the same ladle of metal. These cylinders were marked A, B, and C. From the lower end of each of these cylinders was taken one specimen 10 inches long, and numbered (1). These specimens were all turned to an exterior diameter of 3.384 in., and bored to an interior diameter of 1.128 in., leaving a thickness of metal below the bottom of the bore, which terminated in a hemisphere, equal to twice its diameter, the bottom of the cylinder being cut square off. At the muzzle ends of the cylinders were left collars, by which they were suspended in another cast iron cylinder while the bursting pressure was being applied, in order that they might be subjected to a longitudinal strain at the same time, as is the case in the discharge of fire-arms. A mean of the results obtained from these three cylinders was taken as the true one, since they were all broken under the precise same circumstances. Other sets of specimens of the same form and dimensions were taken from these cylinders, each set being numbered in regular order from the lower ends of the cylinders upwards, and broken under the same circumstances, except that the length of bore subjected to pressure varied in the different sets. The bursting force was produced by filling the specimens to the height required with bees-wax, and pressing upon it with an accurately fitting steel piston, the pressure being exerted by the testing machine, by which a pressure of 100000 lbs. can be exerted, and accurately weighed. In order to prevent the bees-wax from forcing up around the piston, cups made of sole leather, which accurately fitted the bores of the cylinders, were inserted, with their mouths downward, upon the wax, before inserting the piston. Bees-wax was used instead of water in these experiments, for the reason that water was forced through the pores of the iron at a pressure of 11000 lbs. per square inch. J D -LATEI Qmes expressive ofi2ieFramrer.se and 'TcW^ential ' resistances, offered bv 7ia22ow Cylinders to bnrstm?fbrces y £ij)p2iedto different lenohts of bore. c e - - / ?n ^l c JJ A r?' rf B PLATA' H. A J. / ?. /J 1. £ :! c / c z. ----- gggggp^-"^ ~ LATEXT. AND QUALITIES OF CANNON POWDEE. 145 The following table gives the results obtained from 10 sets of specimens : — Number of Seta of Specimens. BURSTING PRESSURE. Diameters. Length of sur- face pressed. Mean bursting A B C Interior. Exterior. pressure. 1 62000 42600 56450 1.128 3.384 6 inch, 53683 lbs. 2 69000 51150 40000 s 1.128 3.384 5 " 60075 3 40000* 72000 70550 1.128 3.384 4 " 71575 4 80200 80800 72900 1.128 3.384 3 " 77966 5 86000 87075 89800 1.128 3.384 2 " 87625 6 60000 64850 56950 1.128 3.384 7 " 60600 7 8 74000 63800 Broke in 62900 preparation. 71000 1.128 1.128 3.384 3.384 2 " 2 " 74000 65917 9 94400 86600 89600 1.128 3.384 2 " 90200 10 96350 92650 93675 1.128 3.384 2.6" 94225 * These specimens broke by splitting, one (C 2) through the muzzle end, and the other (A 3) through the solid end. Both are regarded as defective, and not included in the mean results. It was believed, before commencing these experiments, that the transverse resistance would dimmish as the length of bore, subjected to pressure, was increased ; and that beyond a certain length, which was not believed to be more than four or five inches, it would be inappreciable ; and it was intended to begin with such a length of surface, pressed, in these experiments. Set No. 2, however, gave a greater resistance than No. 1 ; therefore, after diminishing the length of surface pressed to two inches, with a constantly increasing resistance, set No. 6 was burst with seven inches length of surface pressed, the resistance being greater than that given by set No. 1, with six inches length of surface pressed, and about equal to that of set No. 2, with five inches surface pressed ; from which it is concluded that the transverse resistance is inappreciable beyond five calibres in length, with a thickness of metal of one calibre. The results recorded in the foregoing table show a rapid increase in resistance as the surface pressed diminishes ; and inspection of the re-assem- bled fragments (see Plates 2 to 11 inclusive) gives additional proof that it is due to the increase in transverse resistance. All the specimens of sets Nos. 1, 2, and 6, broke by splitting longitudinally, without any indication of transverse fracture ; those of set No. 5, which had four inches in length of surface pressed, split longitudinally, but the frag- ments had small transverse cracks just opposite the middle of the length of 146 PEOPEETIES OP METALS EOE CANNON, surface pressed; showing that the transverse strain had been greater in these specimens than in those having a greater length of surface pressed, and that the tangential resistance had been first overcome. In set No. 4, with three inches length of surface pressed, specimen A broke, first transversely, and then longitudinally; while B and C both broke, first longitudinally, and then transversely, as shown by the fact that the middle line of transverse fracture is continuous in (A), and the lines of longitudinal fracture on opposite sides of this line terminate abruptly, at different points upon it, while the lines of longitudinal fracture are continuous in (B) and (C), and those of transverse fracture terminate abruptly, and at different points, upon them ; the rule in this set being, that longitudinal rupture first super- vened. The same reasoning, upon the lines of fracture in set No. 5, with two inches length of surface pressed, shows that the rule in this set clearly is, that trans- verse rupture first supervened. It follows from these facts that the length of surface pressed, necessary to produce transverse and longitudinal rupture simultaneously, or to fully develop both the transverse and tangential resistances in these cylinders, is between two and tbree inches, and nearer to two than three. In order further to investigate the value of the transverse resistance, another set of cylinders (No. 7) was prepared, of the same form and lateral dimensions as those used in previous trials, except that they had transverse grooves turned in them at the middle of the length of surface pressed, so as to cut off a portion of the transverse resistance, without in any material degree affecting the tangential resistance, the grooves being .125 inch wide and of such a depth as to leave .125 in. of metal around the bore at that point. Two of these specimens broke in preparation, and the other burst at 74000 lbs. ; while those of set No. 5, with the same length of surface pressed, gave a mean of 87625 lbs. Not being satisfied with a single result, I caused another set (No. 8) to be prepared, similar in all respects to set No. 7, except that there was left at the bottom of the grooves .25 in. instead of .125 in. of metal. The mean bursting pressure of this set was 65917 lbs. ; and, including the tested specimen of set No. 7, the mean bursting pressure of the four speci- mens was 67925 lbs. These grooves did not cut off the whole of the transverse resistance, but AND QUALITIES OF CANNON POWDEE. 147 only that portion which the staves would offer against being bent outwards if merely supported, instead of being fixed at their extremities, leaving the resistance which the fixed ends of the staves could offer unimpaired. The formula for the load which a bar of rectangular cross section can bear, when fixed at its extremities, and having the load uniformly distributed along its whole length, is T7= — — = ; while that for the load which the v same bar would bear if merely supported at its ends, is W= — ; and it is this portion of the transverse resistance which the grooves cut off. If, therefore, the staves were of rectangular cross section, the grooves ought to cut off two-thirds of the whole transverse resistance ; but since they are thicker on their outer than on their inner edges, the grooves ought to cut off more than two-thirds; experiment having shown (page 141) that a bar, similar in cross section to the staves which would be cut from a 6-pdr. gun one calibre thick, by planes through its axis, offers a greater transverse resistance when the thick edge is bent outwards, than when the thin edge is outward, in the proportion of 112 to 100. Therefore the transverse resistance cut off by the grooves should be to that remaining as 112 : 50. If we take a mean of the results obtained from sets of specimens Nos. 1, 2, and 6, for the true tangential resistance, we find it to = 58123 lbs. ; while the mean total resistance of sets Nos. (5) and (9) = 88912. The whole transverse resistance, therefore, (88912 — 58123) = 30789 lbs. The amount of transverse resistance cut off by the grooves= (88912 — 67925) = 20987 lbs. ; while that remaining = (30789 — 20987) = 9802 lbs. The transverse resistance cut off by the grooves is therefore to that remaining, as 20987 to 9802, or in a proportion greater than two to one; and corresponding almost exactly with the previously determined ratio of 112 to 50. These results leave no doubt of the existence of a transverse resistance to the rupture of a hollow cylinder by a central force, when that force acts upon only a portion of the length of the cylinder. They also show this resistance to be equal to about one-half of the tangen- tial resistance for two calibres length of surface pressed, and that it rapidly diminishes as the length of surface acted upon increases, being scarcely appreciable beyond five calibres in cylinders one calibre thick. The value of 148 PEOPEETIES OF METALS FOE CANNON, this resistance depends, also, upon the thickness of metal and diameter of bore of the cylinder. In order to determine the value of this resistance as an auxiliary to the tangential resistance, we regard the staves of which we may suppose the cylinder to consist, as being fixed at their extremities, and having the pres- sure to which they are subjected equally distributed along their whole length. The formula for the transverse resistance of a bar of rectangular cross sec- 1 2 S' b d 2 tion, thus circumstanced, is W= » in which W = breaking weight, 6 ST transverse strength of material, b = breadth, d = depth, and / = length of bar. This formula gives the value of the transverse as an auxiliary to the tan- gential resistance, for that particular length of surface pressed, at which trans- verse and longitudinal rapture supervene simultaneously ; but for all lengths greater than this, the staves offer only so much transverse resistance as equals the force required to produce in them a deflection, or sag, equal to half the increase in diameter of 'the bore, due to the tangential extension of the metal at the moment of rupture. But the amounts of sag produced in the same bar by different weights are, within certain limits, directly proportional to those weights. And both theory and experiment show that the sag produced in bars of the same cross section and material, but of different lengths, will be, when loaded with equal weights per unit of length, directly proportional to the fourth powers of their lengths. (See page 143.) T . x . . US' bd 2 It, therefore, = the resistance which a smgle stave could offer, if bent to its breaking deflection, and x = that which it does offer at the moment of tangential rupture, d' =■ sag which it could undergo, e== extension of metal per inch of length at the moment of rupture, r = radius of bore, / = length of surface pressed at which transverse and longitudinal rupture supervene simultaneously, and L = any other length of stave or surface pressed, we shall have the proportion -^—= : x : : d' : e r : : L* : I* ; or x 12 S' bdU* AND QUALITIES OF CANNON POWDEE. 149 The value of (I) is thus determined. Sup- pose a stave to be bent, with its thick edge out- ward, around a circle of such diameter that the exterior of the stave shall be just at the break- ing strain ; then twice that arc of this circle, whose versed sine is equal to e r, will be the value of (I). Let R = radius of this circle, t = thick- oFortheverys^Uvalneofthearcsubtended neSS °f metal in the Cylinder; then if We by £ i, the sine and arc are regarded M equal. assume the neutral axis of the stave to be at its inner surface, as it doubtless will be very nearly under the simultaneous action of the longitudinal strain, we shall have, as shown by the figure, e : t : : 1 : R", or R' = — , and /= S t er r — 2 sine of arc to versed sine ^ . e -K — r But d=t, and>S"= —; or the transverse unit of strength is one-fourth that of the tensile, and the value of x % as above, is the resistance offered by a single stave ; but experiment has shown (page 142) that each stave offers a resistance equal to that which would be offered by a bar of rectangular cross section, whose depth = that of the stave, and whose breadth = the mean thickness of the outer and inner edges of the stave. And since the resistance offered by the staves is in the direction of radii, or planes through the axis of the cylinder, their total resistance will be esti- mated on the same principle as the bursting effort of a central force ; or, if the staves were of the same thickness on their inner and outer edges, then their total resistance to the rupture of one side of the cylinder would be 12 S" r d 2 1* > the radius of the bore being substituted for the breadth of the stave. But, since each stave offers as much resistance as would be offered by a bar of rectangular cross section, and same depth, whose breadth = mean thickness of inner and outer edges of the stave, this resistance will become, 12 &" l B f 2 ) d 2 1' 2 R4-r by substituting — - — for in which R exterior radius of cylinder; but (d), in this expression, is the depth of the bar or stave, and is equal to the thickness of metal in the cylinder, and therefore 250 PEOPEETIES OF METALS FOE CANNON, __ f _ (i? _ r) ; and substituting this value of (d), and for #' its value -, we have the total transverse resistance to the rupture of one side of the cylin- 12 *? (R + r\ (R — ryi* SS(R + r)(B — r)*l\ der = i \ 2 / = ■ ^-j-t With a view to determine the degree of coincidence between the value of (l), as determined in these experiments, and that given by the equation j_ ?^ 2 /sine of arc whose versed sine = |r£-)> let us assume e = .0029, which is the mean value deduced from four bars, each 35 inches long, cut from 10-inch Columbiads Nos. 335 Port Pitt, and 983 West Point. e 2 r In tbese cylinders (R — r) = 1.125, and r = .5625 ; therefore -^-^ = .0000042 = nat. versed sine of 0° 10'. And by the proportion radius : sine 0° 10' : : : — , we have - = 2' 2 e ! ine °° 10, = 1.128, and consequently I = 2.256. Now it has been shown, radius by reasoning upon the lines of fracture (page 146), that with three inches length of surface pressed, the rule was that the cylinders first broke longitu- dinally ; and that with two inches length of surface pressed, the rule was that they first broke transversely; and consequently that the length of surface pressed, necessary for the full development of both transverse and tangential resistance, was between two and three inches. The foregoing calculation gives it = 2.256 in., while the lines of fracture of sets of cylinders Nos. 4 and 5, indicate that it is nearer to two than three inches. The coincidence is sufficiently close to show that, notwithstanding the very minute quantities which enter the calculation and which have to be mechanically measured, the results are reliable. The value of the transverse as an auxiliary to the tangential resistance is, perhaps, more clearly shown by the diagram (Plate 1) constructed from the tabulated results. This diagram is constructed as follows, viz. : — The axis of abscissa is assumed to be parallel to that of the gun, and the axis of ordinates corresponds, in position, with the bottom of the bore. The diameter of bore and thickness of metal are assumed to be the same as those of the cylinders from which the tabulated results were obtained. The points at which the total resistance was determined will therefore be at AND QUALITIES OF CANNON POWDEE. 151 2, 3, 4, 5, 6 and 7 inches from the origin of co-ordinates. Then, if the gun be of uniform thickness throughout its length, its tangential resistance will likewise be uniform for all lengths of surface pressed, greater than that which fully develops the transverse resistance. Then, since the total resistance does not sensibly diminish beyond about six calibres, the resistance at this point is the measure of the tangential resistance. If, therefore, at the distance of 6 inches, the ordinate B D be erected, corresponding in value with the mean total resistance offered at 5, 6, and 7 inches, and through the point (D) the line D C be drawn parallel to A B, the distance between these lines will represent the tangential resistance for all lengths of surface pressed, greater than that which fully develops the transverse resistance. The value of the transverse resistance will be shown by erecting at the distances 2, 3, 4, and 5 inches, ordinates corresponding in value with the total resistance due to these lengths of surface pressed, and through their extremities drawing the curve D b ; the distance between this curve and the line D C e will correspond in value with the transverse resistance. Both theory and experiment indicate that the transverse resistance is fully developed at about 2 calibres length of surface pressed ; and of course it will be for all lengths less than this. Therefore the transverse resistance for all lengths of surface pressed, less than 2 calibres, will be to that at 2 calibres, inversely as those lengths. The part (c b) of the curve (e b D) is therefore constructed in accordance with this law. But the tangential resistance will not be fully developed for any length of surface pressed less than two calibres, and its value will be found by demitting from the curve (e m m' b) ordinates corresponding in value to the total resistance at the points for which the tangential resistance is required. Thus at 2 inches the ordinate m' n' is drawn, corresponding in value with the total resistance at 2 inches, and at 1 inch the ordinate m n, corresponding with the total resistance at one inch, and through the extremities of these lines the curve I n n' is drawn. The total resistance at one inch is computed from knowing that of a weaker specimen at one and two inches ; the resistance of the cylinders from which the other results were obtained being beyond the capacity of the testing machine at one inch 152 PEOPEETIES OF METALS FOE CANNON, That portion of the curve I n n' d, which lies betwen / and n, is hypo- thetical, as we have no results for such short lengths of surface pressed ; but we know that the tangential resistance which the gun or cylinder offers will diminish as the surface pressed diminishes from this point ; and that if the diameter at the bottom of the bore should not be increased by the pressure, this resistance would be zero at that point. Under this supposition the point (I) would fall at (e,) but the solid metal at the bottom of the bore is doubtless affected, and the diameter at that point increased by a pressure which will produce rupture ; and it will probably not be far from the truth to suppose the tangential resistance to be one-half developed at the bottom of the bore, which would place (/) intermediate between [A and e). The distance between the lines c D and I B corresponds to the total resistance ; that part above the line e D representing the transverse, and that below, the tangential resistance. The ordinates corresponding to resistance to rupture are on a scale of .01 of an inch per 1000 lbs. Effect of Chambers on Endurance of Guns. With a view to determine in what manner the resistance of cannon to the explosive force of gunpowder is affected by the use of the chamber, two other sets of cylinders, Nos. 9 and 10, were prepared, of the same form and dimensions as those of sets Nos. 7 and 8, except that they had no transverse grooves ; and set No. 10 had chambers exactly proportional to that of the 10-inch Columbiad. These two sets of cylinders were broken by pressure upon equal weights of wax in each. This gave different lengths of surface pressed, as shown in the table ; the difference being proportional to what would occur in the 10-inch Columbiad by the use of equal weights of powder in guns with and those without chambers, supposing the shot to be rammed home, and that no space is left around the cartridge. The results obtained from these two sets of cylinders show those with chambers to be the strongest, though the difference is not sufficiently marked to remove the suspicion which lies against the use of the chamber. For although, as in this instance, the chambered specimens might in all cases offer a greater ultimate resistance to a single application of a slowly applied force, yet, for the reasons given at page 48 of my Report of Experiments in 1856, it by no means follows that they would resist a greater number of AND QUALITIES OF CANNON POWDEE. 153 repetitions of a suddenly applied force, of less intensity than that which would produce rupture at a single application. Experiments on guns cast from the same heat, same iron, cooled and proved under precisely similar circumstances throughout, one with and the other without chamber, are required, to give a practical solution to this question, which is believed to be of sufficient importance to justify the experiments. Thickness of Metal in the Breech. For the purpose of determining the proper thickness of metal in the breech of cannon, three other cylindrical bars of iron, marked D, E, and F, were cast in dry sand moulds at the same time, and from the same ladle of melted metal. Prom these bars were prepared nine sets of cylinders, of the same lateral dimensions as those already described, the sets being numbered from the lower ends of the bars upward, and having collars turned around their muzzle ends, on which they hung during the application of the bursting force. All the dimensions of the different sets were the same, except the thickness of metal in the breech. The results of this set of experiments are recorded in the following table : — Number of sets BURSTING PRESSURE. Thickness of Length of Mean bursting of Metal surface pressed. weight. Specimens. D E F in Breech. 1 39150 38250 34500 | calibre, 2 calibres, 37300 2 57700 58850 56800 * " 2 " 57783 3 52200 50730 52750 3 « 5 3 " 51893 4 52430 51400 52820 1 3 " 52217 5 52580 52600 52660 1J " 3 " 52613 6 52050 52800 52600 H " 3 " 52483 7 52700 52450 52800 If " 3 " 52650 8 73600 71450 71150 2 1 " 72066 9 49200 48300 49800 i " 3 " 49100 Set No. 1 all broke, by having the bottom forced out. Set No. 2, with two calibres length of surface pressed, broke transversely at the middle of the length of surface pressed, showing cracks in the exterior of the bottom, but 154 PEOPEETIES OF METALS FOE CANNON, no longitudinal cracks ; thus showing conclusively that by the joint action of the transverse and longitudinal strains, the specimens were broken trans- versely and pulled asunder longitudinally, before the tangential resistance had been fully developed ; while set No. 9, (which was prepared to replace set No. 2,) with 3 calibres length of surface pressed, all split longitudinally. Showing that with the greater length of surface pressed, the tangential resistance was overcome before the transverse had been fully developed; and showing the length of surface pressed, required to fully develop both resistances, to be between two and three calibres, as has been previously shown theoretically. Set No. 8, with one calibre length of surface pressed, broke transversely, and was pulled asunder, showing no sign of longitudinal fracture. These results show that no important increase in resistance is obtained by increasing the thickness of metal in the breech beyond lj calibres, and it is believed that lj calibres would be a safe rule to adopt. Tangential Resistance of Hollow Cylinders. For the purpose of determining the relation between the tangential resist- ance to a bursting force, the tensile strength of the metal, and the thickness of the walls of hollow cylinders, three other iron bars, G, IT, and I, of circular cross section, were cast in dry sand at the same time, and from the same ladle of metal. These bars were cast smaller at their lower ends, and increased by steps, or quick tapers, at intervals of eight inches ; this being the length required to furnish a cylinder five inches long, and a specimen for tensile strength. These bars were cast in this form, in order that the cylinders to be cut from them might have the same thickness of metal to be turned off in preparation, it being believed that this precaution would yield more accurate results than would be obtained by casting the bars of the same diameter throughout their length, and turning away the metal necessary to bring the smaller cylinders to the required exterior diameters. The specimens for tensile strength were taken from the lower portions of the lengths of different diameters, each cylinder having its corresponding specimen for tensile strength contiguous to it. The cylinders were five inches long, one inch in interior diameter, and varied from .2 to 1.2 in. in thickness of walls; the three cylinders cut AND QUALITIES OF CANNON POWDEB. 155 from corresponding parts of the three bars, being always of the same thickness. These cylinders were open at both ends, and were burst by filling them full of bees-wax, with leather packing at both ends, and forcing a steel piston into one end, while the other rested upon a plane surface of iron. The following tables exhibit the results obtained : — Table shouting Bursting Pressure, Mean Bursting Pressure per square inch, and Thickness of Metal NuMBEB OF BURSTING PBESSUBE. Mean bursting Thickness of pressure per square inch. Sets. G H I Mean. metal. 1 7815 7780 7420 7672 9768 .2 inches. 2 11750 12000 11250 11666 14854 .3 ' 3 15830 15970 16000 15933 20286 .4 ' 4 18455 18470 18700 18542 23610 .5 ' 5 21650 21460 21460 21523 27404 .6 ' 6 25000 25160 25190 25116 31979 .7 < 7 29560 28860 28500 28973 36890 .8 ' 8 31070 28000 32555 30542 38887 .9 ' 9 35225 35885 36255 35788 45566 1.0 ' no* 111 30000 f 39315 38950 39082 49760 1.1 ' 36000 43000 38370 39123 49813 1.2 ' * These sets are not regarded as fully reliable, t Defective ; not included in mean pressure. These sets of specimens were numbered from the lower ends of the bar upward. Table showing the Tensile Strength and Density of Specimens contiguous to Cylinders referred to in the above Table. NuMBEB OF TENSILE STRENGTH. Mean density. Sets. G 11 I Mean. 1 20188 21921 19520 20543 7.109 2 20516 17888 20569 19658 7.134 3 21467 21753 20201 21474 7.110 4 20190 20089 20558 20279 7.111 4 19765 19220 19572 19519 7.084 6 18895 19452 19388 19245 7.072 7 18886 19151 18506 18848 7.068 8 18973 18874 18815 18887 7.066 9 18882 18612 18967 18820 7.063 10 18992 19229 19061 19094 6.962 11 12095 14796 15672 14188 6.562 156 PEOPEETIES OF METALS FOE CANNON, Table showing the Mean Bursting Pressure per square inch of Cylinders, the Tensile Strength per square inch, and the Density of Tensile Specimens. Number uf Sets. Bursting pressure per square inch of cylinders. Tensile strength per square inch of specimens. Density of specimens. Thickness of metal. 1 9768 20543 7.109 .2 inches. 2 14854 19658 7.134 .3 " 3 20286 21474 7.110 .4 " 4 23610 20279 7.111 .5 " 5 27404 19519 7.084 .6 " 6 31979 19245 7.072 .7 " 7 36890 18848 7.068 .8 " 8 38887 18887 7.066 .9 " 9 45566 18820 7.063 1.0 " 10 49760 19094 6.962 1.1 " 11 49813 14188 6.562 1.2 " The diameters of the tensile specimens, at their points of rupture, were 1.2 inches. These results are anomalous, and do not conform to theory on this kind of resistance. These results, up to the 9th set, inclusive, show the resistance to increase almost directly as the thickness of metal ; while, on the supposition that the metal is wholly incompressible, the resistance would be proportional to the Napierian logarithm of the thickness ; and since we know that cast iron is compressible, the resistance ought to increase in a less ratio to the thickness. These results also show the tangential resistance per square inch of section in the cylinders to be greater than the tensile strength per square inch. Thus in set No. 1, the bursting pressure per square inch is 9768 = tangen- tial resistance of .4 of one square inch, while four-tenths of 20543= 8216; and this difference between the tangential and tensile resistance per square inch increases with the thickness of metal in the cylinders. These discrepancies are partially explicable from the facts that the bees-wax used was not a perfect fluid, and that the tensile specimens were taken from the axial portions of the bars, and were probably not so strong as the outer portions which resisted the bursting pressure ; but a comparison of the tensile strength of set No. 1, which was but little reduced from its cast diameter, with set No. 10, which underwent the greatest reduction from the cast diameter, shows a difference of only 1449 lbs., which would leave 6305 lbs. AND QUALITIES OF CANNON POWDEE. 157 to be accounted for by the imperfect fluidity of the wax ; which is believed to be much more than should be attributed to that cause. For the reasons above stated, these results are not regarded as entirely reliable ; and as this is a very important element in the modelling of cannon, it is earnestly recommended that these experiments be repeated on a larger scale, and with a more perfect fluid, if one can be found that will not permeate the pores of the iron under the bursting pressure. Extensibility and Compressibility of Gun Metal. For the purpose of determining the extension per inch, in length, of cast iron, due to different weights borne, and its ultimate extension, or that due to the rupturing force, the following experiments were made, viz. : — Four bars, 40 inches long each, were cut from each of the two 10-inch Columbiads, No. 983, cast solid at the West Point foundry, and No. 335, cast solid at the Fort Pitt foundry, both cast in 1857, and from the same quality of metal. Two of these bars, from each gun, were cut from near the surface of the bore, and the other two from near the exterior surface of the gun ; they were all taken from the breech ends of the guns, and the middle of the length of the prepared specimens came from opposite the seat of the charge. These specimens had collars left on them at a distance of 35 inches apart, the space between the collars being accurately turned to the same diameter throughout. The space between the collars was surrounded by a cast iron sheath, eight- tenths of an inch less in length than the distance between the collars ; it was put on in halves and held in position by bands, and was of sufficient interior diameter to move freely on the specimen. When in position, the lower end of the sheath rested on the lower collar of the specimen, the space between its upper end and the upper collar being supplied, and accurately measured by a graduated steel scale tapered to .01 inch to 1 inch. The upper end of the sheath was mounted with a vernier, and the scale was graduated to tenths of inches. This apparatus afforded means of measuring the changes of distance between the collars, to the ten thousandth part of an inch, and these readings, divided by the distance between the collars, give the extension per inch, in length, as recorded in the following tables : — 40 158 PEOPEETIES OF METALS EOE CANNON, Table showing the extension, restoration and permanent set per inch, in length; caused by the undermentioned weights, per square inch of section, acting upon a solid cylinder 35 inches long and 1.382 inches diameter, taken from near the surface of the bore of tripli- cate 10-inch Oolumbiad, No. 983, cast solid, at the West Point Foundry, in 1857. Weight per square inch of section. Extension per inch in length. _ First difference. Restoration per inch ill length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .0000357 .0000357 2000 .0000714 .0000357 .0000714 .0000357 — — 3000 .0001200 .0000486 .0001200 .0000486 — — 4000 .0001742 .0000542 .0001685 .0000485 .0000057 _ 5000 .0002171 .0000429 .0002057 .0000372 .0000114 .0000057 6000 .0002828 .0000657 .0002657 .0000600 .0000171 .0000057 7000 .0003314 .0000486 .0003114 .0000457 .0000200 .0000029 8000 .0003743 .0000429 .0003486 .0000372 .0000257 .0000057 9000 .0004371 .0000628 .0004057 .0000571 .0000314 .0000057 10000 .0004771 .0000400 .0004371 .0000314 .0000400 .0000086 11000 .0005800 .0001029 .0005257 .0000886 .0000543 .0000143 12000 .0006629 .0000829 .0005972 .0000715 .0000657 .0000114 13000 .0007400 .0000771 .0006600 .0000628 .0000800 .0000143 14000 .0008086 .0000686 .0007115 .0000515 .0000971 .0000171 15000 .0003943 .0000857 .0007772 .0000657 .0001171 .0000200 16000 .0009914 .0000971 .0008514 .0000742 .0001400 .0000229 17000 .0011288 .0001374 .0009545 .0001031 .0001743 .0000343 18000 .0012657 .0001369 .0010400 .0000855 .0002257 -.0000514 19000 .0013715 .0001058 .0011058 .0000658 .0002657 .0000400 20000 .0014943 .0001228 .0011686 .0000628 .0003257 .0000600 21000 .0016600 .0001657 .0012600 .0000914 .0004000 .0000743 22000 . .0018685 .0002085 .0013285 .0000685 .0005400 .0001400" 23000 .0020885 .0002200 .0014028 .0000743 .0006857 .0001457 24000 .0023628 .0002743 .0015171 .0001143 .0008457 .0001600 AND QUALITIES OF CANNON POWDEE. 159 Table showing the extension, restoration and permanent set per inch, in length, caused by the undermentioned weights, per square inch of section, acting upon a solid cylinder 35 inches long and 1.366 in. diameter, taken from near the exterior surf ace of triplicate 10-inch Columbiad, No. 983, cast solid, at the West Point Foundry, in 1857. "Weight Extension First difference. Restoration First difference. Permanent set First difference. per square inch of section. per inch in length. per inch in length. per inch in length. 1000 lbs. .0000611 .0000611 _ 2000 .0000794 .0000183 .0000794 .0000183 - — .3000 .0001089 .0000295 .0001089 .0000295 - — 4000 .0001771 .0000682 .0001771 .0000682 - — 5000 .0002129 .0000358 .0002129 .0000358 - - 6000 .0002700 .0000571 .0002686 .0000557 .0000014 - 7000 .0003328 .0000628 .0003299 .0000613 .0000029 .0000015 8000 .0003986 .0000658 .0003943 .0000644 .0000043 .0000014 9000 .0004557 .0000571 .0004486 .0000543 .0000071 .0000028 10000 .0005100 .0000543 .0004991 .0000505 .0000109 .0000038 11000 .0005500 .0000400 .0005343 .0000352 .0000157 .0000048 12000 .0006414 .0000914 .0006157 .0000814 .0000257 .0000100 13000 .0007100 .0000686 .0006800 .0000643 .0000300 .0000043 14000 .0007700 .0000600 .0007343 .0000543 .0000357 .0000057 15000 .0008557 .0000857 .0008080 .0000737 .0000477 .0000120 16000 .0009243 .0000686 .0008714 .0000634 .0000529 .0000052 17000 .0010014 .0000771 .0009371 .0000657 .0000643 .0000114 18000 .0010900 .0000886 .0009886 .0000515 .0001014 .0000371 19000 .0012271 .0001371 .0010300 .0000914 .0001471 .0000457 20000 .0013586 .0001315 .0011572 .0000772 .0002014 .0000543 21000 .0015386 .0001800 .0012486 .0000914 .0002900 .0000886 22000 .0017043 .0001657 .0013057 .0000571 .0003986 .0001086 23000 .0019529 .0002486 .0014000 .0000943 .0005529 .0001543 24000 .0022786 .0003257 .0015257 .0001257 .0007529 , .0002000 25000 .0026037 .0003251 .0015194 —.0000063 .0010843 .0003314 26000 .0032186 .0006149 — ""* "" "" * 160 PEOPEETIES OF METALS FOE CANNON, Table showing the extension, restoration and permanent set per inch, in length, caused by the undermentioned weights, per square inch of section, acting upon a solid cylinder 35 in. long and 1.382 in. diameter, taken from near the surface of the bore of triplicate 10-inch Columbiad, No. 335, cast solid, at the Fort Pitt Foundry, in 1857. "Weight per square inch of section. Extension .per inch in length. First difference. Eestoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .0000600 .0000600 _ 2000 .0001343 .0000743 .0001343 .0000743 - — 3000 .0002057 .0000714 .0002000 .0000657 .0000057 _ 4000 .0002657 .0000600 .0002543 .0000543 .0000114 .0000057 5000 .0003257 .0000600 .0003114 .0000571 .0000143 .0000029 6000 .0003800 .0000543 .0003543 .0000429 .0000257 .0000114 7000 .0004514 .0000714 .0004143 .0000600 .0000371 .0000114 8000 .0005314 .0000800 .0004771 .0000628 .0000543 .0000172 9000 .0006171 .0000857 .0005457 .0000686 .0000714 .0000171 10000 .0007114 .0000943 .0006171 .0000714 .0000943 .0000229 11000 .0008114 .0001000 .0006885 .0000714 .0001229 .0000286 12000 .0009229 .0001115 .0007600 .0000715 .0001629 .0000400 13000 .0010743 .0001514 .0008371 .0000771 .0002372 .0000743 14000 .0012429 .0001686 .0009458 .0001087 .0002971 .0000599 15000 .0014086 .0001657 .0010229 .0000771 .0003857 .0000886 16000 .0016371 .0002285 .0011114 .0000885 .0005257 .0001400 17000 .0019571 .0003200 .0012314 .0001200 .0007257 .0002000 18000 .0023143 .0003572 .0013086 .0000772 .0010057 .0002800 19000 .0028257 .0005114 .0014143 .0001057 .0014114 .0004057 AND QUALITIES OF CANNON POWDEE. 161 Table showing the extension, restoration, and permanent set per inch, in length, caused by the. undermentioned weights, per square inch of section, acting upon a solid cylinder, 35 in. long and 1.382 in. diameter, taken from near the exterior surface of triplicate 10-inch Colwm- liad, No. 335, cast solid, at the Fort Pitt Foundry, in 1857. Weight per square inch of section. Extension per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .0000429 .0000415 .0000014 2000 .0001086 .0000657 .0001057 .0000642 .0000029 .0000015 3000 .0001714 .0000628 .0001657 .0000600 .0000057 .0000028 4000 .0002343 .0000629 .0002257 .0000600 .0000086 .0000029 5000 .0002943 .0000600 .0002814 .0000557 .0000129 .0000043 6000 .0003543 .0000600 .0003372 .0000558 .0000171 .0000042 7000 .0004200 .0000657 .0003971 .0000599 .0000229 .0000058 8000 .0004886 .0000686 .0004600 .0000629 .0000286 .0000057 9000 .0005571 .0000685 .0005200 .0000600 .0000371 .0000085 10000 .0006314 .0000743 .0005828 .0000628 .0000486 .0000115 11000 .0007029 .0000715 .0006400 .0000572 .0000629 .0000143 12000 .0007943 .0000914 .0007143 .0000743 .0000800 .0000171 13000 .0008943 .0001000 .0007857 .0000714 .0001086 .0000286 14000 .0010114 .0001171 .0008571 .0000714 .0001543 .0000457 15000 .0011428 .0001314 .0009299 .0000728 .0002129 .0000586 16000 .0012714 .0001286 .0009600 .0000301 .0003114 .0000985 17000 .0014914 .0002200 .0010000 .0000400 .0004914 .0001800 18000 .0018114 .0003200 .0011514 .0001514 .0006600 .0001686 19000 .0022772 .0004658 .0012144 .0000630 .0010628 .0004028 20000 .0029000 .0006228 .0013000 .0000856 .0016000 .0005372 21000 .0033657 .0004657 Specimen t measured roke before tl le permanent set could be For compressibility, bars contiguous, in the guns, to those for extensibility, were tested ; the measuring apparatus was the same, and the bars were similarly prepared in all respects, except that they were only ten inches long between the collars. The results thus obtained are recorded in the following tables : — 162 PROPEETIES OF METALS FOE CANNON, Table showing the compression, restoration and permanent set per inch, in length, caused by the undermentioned weights, per square inch of section, acting upon a solid cylinder 10 inches long and 1.382 inches diameter, taken from near the surface of the bore of tripli- cate \0-inoh Columbiad, No. 983, cast solid, at the West Point Foundry, in 1857. "Weight per square inch of section. Extension per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .000100 2000 .000240 .000140 — - — _ 3000 .000310 .000070 .000300 - .000010 _ 4000 .000380 .000070 .000340 .000040 .000040 .000030 5000 .000435 .000055 .000375 .000035 .000060 .000020 6000 .000505 .000070 .000440 .000065 .000065 .000005 7000 .000540 .000035 .000475 .000035 .000065 .000000 8000 .000625 .000085 .000555 .000080 .000070 .000005 9000 .000670 .000045 .000540 —.000015 .000130 .000060 10000 .000780 .000110 .000640 +.000100 .000140 .000010 11000 .000850 .000070 .000700 .000060 .000150 .000010 12000 .000890 .000040 .000730 .000030 .000160 .000010 13000 .000965 .000075 .000790 .000060 .000175 .000015 14000 .001010 .000045 .000800 .000010 .000210 .000035 15000 .001055 .000045 .000335 .000035 .000220 .000010 16000 .001095 .000040 .000865 .000030 .000230 .000010 17000 .001240 .000145 .000990 .000125 .000250 .000020 18000 .001295 .000055 .001015 .000025 .000280 .000030 19000 .001345 .000050 .001055 .000040 .000290 .000010 20000 .001430 .000085 .001120 .000065 .000310 .000020 21000 .001490 .000060 .001150 .000030 .000340 .000030 22000 .001570 .000080 .001200 .000050 .000370 .000030 23000 .001620 .000050 .001230 .000030 .000390 .000020 24000 .001720 .000100 .001290 .000060 .000430 .000040 25000 .001790 .000070 .001320 .000030 .000470 .000040 26000 .001880 .000090 .001380 .000060 .000500 .000030 27000 .001955 .000075 .001415 .000035 .000540 .000040 28000 .002040 .000085 .001455 .000040 .000585 .000045 29000 .002105 .000065 .001485 .000030 .000620 .000035 30000 .002250 .000145 .001560 .000075 .000690 .000070 AND QUALITIES OF CANNON POWDEE, 163 Table shotting the compression, restoration and permanent set per inch, in length, caused by the undermentioned weights, per square inch of section, acting upon a solid cylinder 10 inches long and 1.382 ire. diameter, taken from near the exterior surface of triplicate 10-inch ■Columbiad, No. 983, cast solid, at the West Point Foundry, in 1857. Weight per square inch of section. Compression per inch in length. First difference. Restoration jyer inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .000090 •000090 __ _ 2000 .000170 .000080 .000170 .000080 - — 3000 .000255 .000085 •000250 .000080 .000005 - 4000 .000320 .000065 .000305 .000055 .000015 .000010 5000 .000385 .000065 .000360 .000055 .000025 .000010 6000 .000455 .000070 .000425 .000065 .000030 .000005 7000 .000505 .000050 .000170 .000045 .000035 .000005 8000 .000575 .000070 .000530 .000060 .000045 .000010 9000 .000645 .000070 .000590 .000060 .000055 .000010 10000 .000705 .000060 .000635 .000045 .000070 .000015 11000 .000790 .000085 .000680 .000045 .000110 .000040 12000 .000845 .000055 .000725 .000045 .000120 .000010 13000 .000905 .000060 .000775 .000050 .000130 .000010 14000 .000955 .000050 .000810 .000035 .000145 .000015 15000 .001035 .000080 .000865 .000055 .000170 .000025 16000 .001090 .000055 .000905 .000040 .000185 .000015 17000 .001165 .000075 .000955 .000050 .000210 .000025 18000 .001250 .000085 .001015 .000060 .000235 .000025 19000 .001335 .000085 .001065 .000050 .000270 .000035 20000 .001395 .000060 .001095 .000030 .000300 .000030 21000 .001485 .000090 .001150 .000055 .000335 .000035 22000 .001555 .000070 .001190 .000040 .000365 .000030 23000 .001655 .000100 .001250 .000060 .000405 .000040 24000 .001750 .000095 .001295 .000045 .000455 .000050 25000 .001825 .000075 .001330 .000035 .000495 .000040 26000 .001940 .000115 .001385 .000055 .000555 .000060 27000 .002050 .000110 .001440 .000055 .000610 .000055 28000 .002145 .000095 .001475 .000035 .000670 .000060 29000 .002250 .000105 .001515 .000040 .000735 .000065 30000 .002380 .000130 .001560 .000045 .000820 .000085 164 PEOPEETIES OF METALS EOE CANNON, Table showing the compression, restoration, and -permanent set per inch, in length, caused by the undermentioned weights, per square inch of section, acting upon a solid cylinder 10 in. long and 1.382 in. diameter, taken from near the surface of the bore of triplicate 10-inch Columbiad, No. 335, cast solid, at the Fort Pitt Foundry, in 1857. Weight per square inch of section. Compression per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 libs. .000145 .000145 2000 .000225 .000080 .000225 .000080 — _ 3000 .000305 .000080 .000300 .000075 .000005 _ 4000 .000375 .000070 .000360 .000060 .000015 .000010 5000 .000465 .000090 .000440 .000080 .000025 .000010 6000 .000530 .000065 .000485 .000045 .000045 .000020 7000 .000615 .000085 .000560 .000075 .000055 .000010 8000 .000695 .000080 .000610 .000050 .000085 .000030 9000 .000755 .000060 .000660 .000050 .000095 .000010 10000 .000825 .000070 .000695 .000035 .000130 .000035 11000 .000895 .000070 .000730 .000035 .000165 .000035 12000 .000985 .000090 .000800 .000070 .000185 .000020 13000 .001055 .000070 .000850 .000050 .000205 .000020 14000 .001125 .000070 .000875 .000025 .000250 .000045 15000 .001220 .000095 .000955 .000080 .000265 .000015 16000 .001305 .000085 .001000 .000045 .000305 .000040 17000 .001415 .000110 .001065 .000065 .000350 .000045 18000 .001510 .000095 .001115 .000050 .000395 .000045 19000 .001595 .000085 .001165 .000050 .000430 .000035 20000 .001710 .000115 .001215 .000050 .000495 .000065 21000 .001830 .000120 .001275 .000060 .000555 .000060 22000 .001955 .000125 .001315 .000040 .000640 .000085 23000 .002090 .000135 .001370 .000055 .000720 .000080 24000 .002240 .000150 .001430 .000060 .000810 .000090 25000 .002380 .000140 .001475 .000045 .000905 .000095 20000 .002540 .000160 .001475 .000000 .001065 .000160 27000 .002780 .000240 .001565 .000090 .001215 .000150 28000 .003010 .000230 .001655 .000090 .001355 .000140 29000 .003295 .000285 .001770 .000115 .001525 .000170 30000 .003490 .000195 .001720 —.000050 .001770 .000240 AND QUALITIES OF CANNON POWDEE. 165 Table showing the compression, restoration, and permanent set per inch, in length, caused by the undermentioned weights, per square inch of section, acting upon a solid cylinder 10 inches long and 1.382 in. diameter, taken from near the exterior surface of triplicate 10-inch Columbiad, No. 355, cast solid, at the Fort Pitt Foundry, in 1857. Weight Compression Restoration Permanent set per square inch of section. per inch in length. First difference. per inch in length. First difference. per inch in length. First difference. 1000 lbs. .000075 .000075 2000 .000155 .000080 .000150 .000075 .000005 _. 3000 .000205 .000050 .000195 .000045 .000010 .000005 4000 .000265 .000060 .000250 .000055 .000015 .000005 5000 .000355 .000090 .000335 .000085 .000020 .000005 6000 .000465 .000110 .000430 .000095 .000035 .000015 7000 .000545 .000080 .000490 .000060 .000055 .000020 8000 .000625 .000080 .000555 .000065 .000070 .000015 9000 .000685 .000060 .000600 .000045 .000085 .000015 10000 .000770 .000085 .000665 .000065 .000105 .000020 11000 .000835 .000065 .000705 .000040 .000130 .000025 12000 .000895 .000060 .000745 .000040 .000150 .000020 13000 .000970 .000075 .000800 .000055 .000170 .000020 14000 .001055 .000085 .000855 .000055 .000200 .000030 15000 .001140 .000085 .000905 .000050 .000235 .000035 16000 .001230 .000090 .000960 .000055 .000270 .000035 17000 .001310 .000080 .000995 .000035 .000315 .000045 18000 .001385 .000075 .001030 .000035 .000355 .000040 19000 .001485 .000100 .001085 .000055 .000400 .000045 20000 .001575 .000090 .001135 .000050 .000440 .000040 21000 .001690 .000115 ' .001175 .000040 .000515 .000075 22000 .001815 .000125 .001225 .000050 .000590 .000075 23000 .001935 .000120 .001275 .000050 .000660 .000070 ' 24000 .002060 .000125 .001330 .000055 .000730 .000070 25000 .002200 .000140 .001360 .000030 .000840 .000110 26000 .002350 .000150 .001385 .000025 .000965 .000125 27000 .002560 .000210 .001455 .000070 .001105 .000140 28000 .002765 .000205 .001480 .000025 .001235 .000180 29000 .003035 .000270 .001505 .000025 .001530 .000245 30000 .003410 .000375 .001520 .000015 .001890 .000360 42 166 PEOPEETIES OF METALS FOE CANNON, Table showing the extension, restoration, and permanent set per inch, in length, caused ly the repeated applications of 15000 pounds per square inch, — about three-fourths of its ultimate strength, — upon a solid cylinder, 35 in. long and 1.382 in. diameter, taken from near the exterior surface of triplicate 10-inch Columbiad, No. 335, cast solid, at the Fort Pitt Foundry, in 1857. Number of repe- titions. Extension per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1 .001278 .000971 .000307 _ 2 .001291 .000013 .000965 —.000006 .000226 .000019 3 .001304 .000013 .000968 + 000003 .000336 .000010 4 .001320 .000016 .000977 .000009 .000343 .000007 5 .001327 .000007 .000978 .000001 .000349 .000006 6 .001337 .000010 .000978 .000000 .000359 .000010 7 .001346 .000009 .000980 .000002 .000366 .000007 8 .001349 .000003 .000976 —.000004 .000373 .000007 9 .001354 .000005 .000975 —.000001 .000379 .000006 10 .001357 .000003 .000969 —.000006 .000388 .000009 11 .001364 .000007 .000970 +.000001 .000394 .000006 12 .001368 .000004 .000971 .000001 .000397 .000003 13 .001374 .000006 .000974 .000003 .000400 .000003 14 .001377 .000003 .000974 .000000 .000403 .000003 15 .001380 .000003 .000976 .000002 .000404 .000001 16 .001381 .000001 .000975 —.000001 .000406 .000002 17 .001383 .000002 .000976 +.000001 .000407 .000001 27 .001389 .000006 .000976 .000000 .000413 .000006 37 .001403 .000014 .000982 .000006 .000421 .000008 • ''47 .001471 .000068 .000985 .000003 .000486 .000065 57 .001480 .000009 .000986 .000001 .000494 .000008 67 .001487 .000007 .000984 —.000002 .000503 .000009 77 .001491 .000004 .000982 —.000002 .000509 .000006 87 .001503 .000012 .000983 +.000001 .000520 .000011 100 .001510 .000007 .000984 .000001 .000526 .000006 110 .001513 .000003 .000984 .000000 .000529 .000003 120 .001516 :000003 .000985 .000001 .000531 .000002 130 .001517 .000001 .000981 —.000004 .000536 .000005 140 .001527 .000010 .000984 +.000003 .000543 .000007 150 .001530 .000003 .000977 —.000007 .000553 .000010 160 .001533 .000003 .000980 +.000003 .000553 .000000 200 .001533 .000000 .000977 —.000003 .000556 .000003 250 .001537 .000004 .000977 .000000 .000560 .000004 Specimen remained free from strain forty hours, in which time its .00210, and its extension under 15000 lbs. diminished .00080. • For this series, about 500 lbs. too much was once applied. AND QUALITIES OF CANNON POWDEE. 167 Table of repetition of strain — Continued. Number Extension Restoration Permanent set of per inch First difference. per inch First difference. per inch First difference. repetitions. in length. in length. in length. 300 .001528 .000998 _ .000530 _ 350 .001545 .000017 .000999 .000001 .000546 .000016 400 .001556 .000011 .001005 .000006 .000551 .000005 450 .001556 .000000 .001003 —.000002 .000553 .000002 500 .001624 .000068 .001018 +.000015 .000606 .000053 550 .001646 .000022 .001032 .000014 .000614 .000008 600 .001651 .000005 .001031 —.000001 .000620 .000006 625 .001663 .000012 .001017 —.000014 .000646 .000026 After remain ng free from strain 30 days, pennanen t set diminished .00130, and extension under same force as above, .00100. 650 .001647 - .001011 - .000636 - 700 .001666 .000019 .001015 .000004 .000651 .000015 780 .001687 .000021 .001021 .000006 .000666 .000015 After 29 days rest, free fr >m strain, set dimii lished .00340, and extension under sar ne straiD as above , .00295. 800 .001617 — .001037 - .000580 - 900 .001640 .000023 .001046 .000009 .000594 .000014 1000 .001668 .000028 .001037 —.000009 .000631 .000037 1114 .001737 .000069 .001037 .000000 .000700 .000069 1214 .001785 .000048 .001039 +.000002 .000746 .000046 "1325 .002077 .000292 .001049 .000010 .001028 .000282 1436 .002103 .000026 .001060 .000011 .001043 .000015 1536 .002123 .000020 .001063 .000003' .001060 .000017 1736 .002297 .000174 .001139 .000076 .001158 .000098 1800 .002297 .000000 .001106 .000033 .001191 .000033 After 26 days resl , set diminished .0C 290, and extension under same strain as above, .00280. 1892 .002305 Broi | .001128 | .e at the 1956th repetition. .001177 ~~ Mean, — — .001003 " ~ About 1000 lbs. too much was once applied in this series. 168 PEOPEETIES OF METALS FOB CANNON, Deductions and Conclusions from the Foregoing Results. The most interesting features observable in the results recorded in the fore- going tables, and more clearly shown by the curves constructed from them, (see Plates 12 to 23, inclusive,) are, — 1st. The very marked difference in the extension due to equal increments of force, when the total force is small, and when it nearly approaches that of rupture. 2d. The irregularity of increase in these differences ; and 3d. The excess of compression over the extension, due . to the same force, when that force is comparatively small, and the excess of the extension over the compression, due to the same force, when that force nearly approaches the ultimate tensile strength of the specimen. If the specimen were perfectly homogeneous throughout, in both structure and properties, as if it were a single crystal, then the changes which it would undergo, from the application of a regularly increasing force, would be in accordance with some regular law ; and, if the experiments were accurately made, the results would be in strict accordance with that law, and all the points of the curve expressive of the experimental results would fall upon the curve expressive of the law. The measuring apparatus used in these experiments afforded the means of measuring any changes of length in the specimens for extension, amounting to the .000003 part of an inch per inch in length, and any amounting to the .00001 part of an inch per inch in length, in those for compression. The irregularities exhibited by the columns of first differences being so very much greater than either of these quantities, cannot be attributed to errors of measurement, and are believed to be mainly due to the inherent properties of the specimens themselves. These specimens are composed, not of a single crystal, but of groups of crystals, the planes of whose principal faces occupy every conceivable position with reference to the direction of the straining force. Those crystals whose principal faces are perpendicular to the direction of the straining force, are susceptible of the minimum extension or compression in that direction, and must therefore under a force of extension be the first to rupture ; while those whose principal faces are parallel to the direction of that force, will have the maximum extensibility, and will be the last to rupture ; J>LAT£^]I. Czwves expressive ofe^ensiT?i]jty,comj)Tessid'ibcycmclcoiTe.9poi}din^ set, ofmner specimen £-om IVestPoiat GmiK^dS?^. c/cc 66iffS OC9C cose 66 7 S 6676 OC6S 6C GO , COS, OOSO oo^s 004-0 0635 OCS6 .cc: TC2€ 6J0JS OC/C .6166S fO000lbs 2COOO JOOCO 4VOOO SOCCC 66C6C rr'Cfc TLATEXm. Curves exjjressive ofeYtensibiIity,compressiTahtymidcoiTespondinQ set, of cater specimen £-om Westlbint Gm2V?$S3. -X'CSC .(■045 0C4-C CC3S OOoC OOW OCJC /OCGOZ&s zeeoc 3CCOO 4-ecce sococ 6COOO 7COOO J'LATJEXIV Carves expressive ol extensi 6" ib'ty, compressibility aiid 'oi "corresponding set of inner specimen jfrom Fort Kit Gun A^ 33.) CJOf ■ ro9s 00.90 ■ ('(So J— Of. 60 ■ 007S 0070 , 006J -. CC6C' \ .COS .CC45 .004fO 003S C030 oozs 0020 .COS -. 00/0 00C5 ?O0eczmen from Forr Pjrt 6wiJi ^J;>S Cue f-<95 ff'9l e* J ftTl CI 60 I f'j'C c 1 - r ^— - l|i ! ' 1 ! ' ( i ! t h - — -" -— -- - — i — r - - 1 ; ■i ! t I -— ! : ' i i ;. ; — ! - - -- - - i ■ ■ ! 1 1 ; i I i - ' ' ' i ■ j ! — — - -~ - - ' < ' i - — -- - 1 ; 1 -— -- 111: ! I i ', ' i ■ i ; 1 1 t i ' ! i ; 1 ! ^ ■ : [ i r i ! 1 ■ ' • ' ' J I ! : : - - - 1 1 ' 1 1 — - • '- ' ■ j 1 ' ' i : M , ' ' , I i i i ; — i : _^ . : ! i ; I ! i j II ' ; 1 : . • i i ! III; 1 I . . i 1 1 1 1 1 --, — I 1 i > '- ; ! , . . 1 1 . j 1 ' 1 __!_ : j I ■ i -- — ----- -— f~ — r .._. ■ 1 | I ; ' ; , i ! i ! i ■ 1 - ! i i ! ■ ■ i 1 ■ 1 ■ z ; -- - :-~- !' i i i ! 1 1 • ' j i 1 j / 1 , i 1 ! -'- ; ■ i 1 ' , — 1 i i i ' , 1 ! ; i : , I ! i ! : ... . _. < ' ' \ --- - -- - | i -\ v— -- -- ' '■ -— " i . ' i ! ' \ 1 ! 1 i t : ! - 1 \ '■ ' ■ ' ! i i i 1 , t i i : i : i ' ' ' ! ' ' i -' — - - ....: . — 1 1 i 1 / / — jH — ! !■ '■ ■ ' i/l — i * i 77- ; j i i H ' :_ -- - - - - - -- ■i i i ' i 1 | ! ' -- ) __ II i i / y j , 1 | V Mi--. — — - ■ - — _ n ! 1 'J 1 ! ~M r^ 1 1 ' £?-/ i i i : h y' /A ,1 1 i — it, {J> ill/ 1 —■•— t — 1 1 V J /& ' tM j ^/ tM >■<> ; : ' : y j \y^ - - " * \ — - -- __ .__ : ^^ i i I i — j- . i i rjfc 70oclo^As ZOooo 7CO0O JPLATJSYVn CoinpcLrin^, compressibility and set of inner rZncloTrterspecjmensfivm WestFoint 9aa2^°^S3 . c/co HOC cogs CCiO 0C1S CO70 cce,' tree VVS5 ecsc Oc-iS CT^C 0036 0030 0025 VOZO OOfS core OOOjj ■tOOOOIhs zoooo 300CC ■Wooc SOOOC 6C0CS 1CQCC JZATEXTST Comp#rin£ extensibility and set of inner and outer specimen? drom \FortHtt Gmil T ?33J. ,OtO0 C0.9S 0O90 0C15 ootc \O070 0030 oozs ooze OCiO TC~C<. iir 2cOC( ■3<-rot 4COOO soccc 6'oroo roooo T1ATEJIZ Cowj)£j"in<3 compressibih'tr and 'set orinner and outer specimens from Fort Pitt Ganyr , J3S otoo coos ■Of' 90 OOSS COM .CC7S arc CX65 CC6C C'C.'S re 5c ccis CC4-C CC3C n ;c coys -cct<: CCCS 160 OO lis zoooo 30000 40CC0 soooc 6y.ee 7CCC0 PL.XI'Ji XX t'ompaiiij'i exteusitjUitv rind set cfimjor specr/n ens from WrstJ.'cnit GuuXtVSS ami Fort Pitt CrmJ''.AlJ. ■ — ■■* ' ' ■■ i i i ii t ; - : ' : - ,C : ! i J , i i ! 1 ! t t I 1 1 i 1 , 4 i - ■ 1 ■ 1 . , " ■" i i ! i 1 • j i i ;- ]--r- - 1 1" ... i i -■ 1 -■-} \ t ! ,{ ; : : ' , j i r\~\- i ■ , ' ' : - J- - 4 - ! L ' ! : ._ : . ! I j ! - 1 \ ill 1 - | i ! ! - ■ I 1 i ' M '' }-\x. ! ' ■ ' f 1 j ; 1 i i L - f f v i T-L 1 i ■ -i - - - - :. ! j -1- '. j ! !- i ! : ' : i 4 ' , . i ! - 1 --I i 1 ' : * ■' i - : |-| ; j t J.i L ! i j I 11 -J - ~i ! ' 7 1 i : ' - r ! \ ■ 1 i • ■ ? i i i ;-f- "■1 i * I - + ■ I'll ! j' t ' - .- _ - ~~ I j- . ^ • • 1 l ' - : -I • - ~ 7 i : - - - - — — ' — T" 1 ! r -11! i i 1 ' - r;':, — f - 1 i I ; - '- 1 ' ■ ' ; — T ~ ( 1 ! : "'If - i -i 1 - -j M i - -4- l-i \ . , 1 i ; i ; j / j : / — 1 — i — j — ■ ■; i j i ) 1 < \ ' • 7 ^ / . „ ( 1 ' ; 7/ - - / / 7 c/ jt- ^ / /a 9 fifths- if ft ?^ .#v P(> fffl. Iff so ccc bC CCC /'( r PLATE XYZ. Comparing extensibility and. set of outer .specimens front T) est Point Gwi :\ ''.%:> andFortTitt Gun A* 335 H 1 r ~ . •It ' " ' [' 1 1 ] , i— i- — -~| — i— - _ u 1 n | ", I j _Li. idz ~r~ > ; i i j i 1 I ; ~! --]-- r 1,1- iL d~ p . 1 ,1 i 1 ' -1 i 1 ) 1 j t 1 i t j 1 1 ' • — (_ _ , . 1 , j ' ; j _l 1 ' i 1 ' i 1 | | , . ( - ; . - -_ + ! ! II I .. ._. j 1 L_l , 1 j i ■ 1 ' 1 ' : j ii ! j j ! 1 ! i i : j ' i i ! 1 : i ' I | ■ | ! 1 I ; ' i ; i ! ! ; ' . 1 ' : i ' : i i i '■ ■ • ; I 1 1 1 ' 1 1 i i !' ; ' ' '■ 1 ill ! i ' Ml ' 1 1 j ! ! ! 1 ! I 1 1 ' i 1 : ' i i i ■ ■ ' * - i ; , 1 i : ' _+ LUii ^ :■ u t k— -j- i ' ■ '■■ ! ' , , t ! ! -, ^-^l_ i|_L±_._U-|- , j i , ' ■ ' i i \ ' i ! i ! 1 ! ' ' ' , : II ; i 1 ' , : : ' ' ■ ' ! , , i , 1 , 1 i , i i i I ' 1 1 1 1 I I 1 ! | ! i ■ ! i ! ! ' 1 j ! i i j j I mm- i— l. _j 1 /..■'■ 1 ; r "! I !"T" 1 ! ! ' j - - - ! ' i i 1 '';/'/ i i j 1 ! / ^ / 1 : • iv 7 ' *%/ iS 1 / of^ 1 / ^ / j \ /j ■ i ■ i ! 7 ■ v / ' ■ 1 i l 1 i ■ ^t ~^#i t-r ■ "i ! 1 1 ! -tf/ *v SZ •pft ± i l : i ; ' , ^ ■ A~ -H- f ! , [ ■. , / -h/ i : v j ii i -t z ,j + ^ -v^t , / 7 M h^ ^ ^ / r y~r 1 1 ^~y &/' ^t/ :_:_ .. . / ./ / > y^^ \ / \ / ' ^ 5^ ^/ t ^-^ _M— M- = £^Lm-—"'^ (ire C09S -CC9C fC 55 Ct'TJ (< 6.T cr .',. ■cfje C( 15 - Cf-H 0C3J - CC3C a zs -IC2C CCfS -C< /C C(CS /Of'CC Zfc zccec 3C0C0 wove J( rrr Or/ cc FLATEXX11. Comparing compressibility and \sec, of inner specimens from WestToinc Gnny?9S5 and '.Fort T itt 6'zniJ Tr .33S. I (9S U9( ((S3 f.( St fj(TS (cic (res <■( 6( I C'h (035 ( (30 (( ?S (CIC Doeco t7>s- 3C0CC' G(f ( O PLATE XX HI . ('(- -^- - -.. . "iTr " " """ ""'+ 1- ■ — - ■ > T l ■ ■ ■ , 1 l_ i . ._i f / - - i— -H 1— h— i 1 J---I-T-4— !— ~H~ r 4_ / v ^4 , 4 Xl iZ V -*t -1- ~r -A-r-L-l J- 4 — -f u -- i pH- Z^ xcv '■' ' .. . " ' ^ .il.M-..-l- - - - ----- 1 1 . X' " ' "~ ?£$/.— --- - -It "i " , ^ 'V -+- t: jlS/ xx. — - 1*1x4- i h4- -4- .p^ / 1 4 i ~" ^ " \T¥ / !T, j7 - -4- -+-4- -- ■- -*-- -- -+- ~-l 1- 1 — ..f/ / r^ / -H~ '£/-4/ -- / i ; , j -+- - XX " aXX X- f -. - - - Ijl -- 4_r " XX ~~ ^"^;^4 — ^P- ■-+- - -J- _^ — ! — >__ -* y ~ -\ . ^ /—f , , ._ja__ - t — h_ - . _L --I 4--";-^ - " 4 ' "4 £oV~^vfc -1- -^Hr t+- -f -1 ,#--"t -1 rti'">r ^':>m • -T-j i-fi" L -.XJX| -}X 4 ~- ! +--- T>4^44+4^--hr4 u rh" 1 r ~HX 1 [III H I I I M II — I i — j — [ — | — [- u Xj "XX j-*^ 44-*^ 4-+-T - - - ■ -■ ---I- -■- ■■■■'- - ■ ----- - - — -jf ^^i^rff rlTf- : 4 r r ] _.j_. "+ . _. I^r .CC.9:, -cose ■ cat -it rt .((6; . ft 6c< . cess ■ cose .00^5 - oo*c . CC3S CC3C ,cczs CC10 .WIS ,00/0 .CCCS ■/OCCOtt.r ZCCCO 3CC7CC bvooc . rocov CiCCCC wee AND QUALITIES OF CANNON POWDER. 169 those whose principal faces are oblique to the direction of the force, being intermediate, in extensibility, and times of rupture, between these limits. It would therefore appear that rupture, from a force of extension, in a specimen thus constituted, ought to take place in detail, group after group of crystals giving way, and throwing the strain upon those that are susceptible of greater extension, and thus causing irregularity in the increments of exten- sion, due to equal and regular increments of force. And the increments of extension, due to equal increments of force, ought to be less, while that force is small, and while it is rupturing those groups of crystals which have the minimum extensibility, than when it is larger, and when it is rupturing those groups which have greater extensibility ; and finally, those increments should be greater when the force is rupturing those groups which are susceptible of the maximum extension. Compression. Under a force of compression, the foregoing results indicate that motion first takes place among those crystals whose principal faces are parallel to the direction of the force. Equal increments of force ought consequently to give greater increments of compression while the force is so small as to put in motion only crystals thus situated, than when it is greater, and is beginning to bring into action those whose principal faces- are perpendicular to the direction of the force. And the minimum increment of compression, due to a given increment of force, ought to occur at the time the resistance offered by those crystals, whose principal faces are perpendicular to the direction of the force, is being overcome ; and as these crystals begin to yield, by being broken up, as it is believed, the increments of compression begin to increase, and continue so to do until the specimen gives way, and rupture ensues. The curve expressive of these properties in the specimens, and of the experimental results, would be first concave and then convex, towards the axis of abscissa, when the abscissa of the curve represents the force, and the ordinates the corresponding compressions in the specimen. The super-position of the curves of compression upon those of extension shows very clearly that, under a transverse strain, the neutral axis, in a specimen of the same quality of iron as that from which the foregoing results were obtained, would, at first, be nearest the extended or convex face of the 170 PEOPEETIES OP METALS FOE CANNON, specimen, and would, as the force would increase, be further and further removed from that face, and would, when the force would equal about three- fourths of that of rupture, be at the middle of the depth of the specimen; and for all forces greater than this, it would be nearest the concave or com- pressed face of the specimen. The point of intersection of the curve of extension with that of compression, the origin of co-ordinates being the same for both, of any specimen, gives the force which will place the neutral axis in the middle of the depth of the specimen. Intermittent Force of Constant Intensity. The most interesting point in the results obtained from subjecting a spe- cimen of the same quality of iron as that from which the other tabulated results were obtained, to a repetition of a strain equal to about three-fourths of its breaking weight, is the fact that at every interval of rest, of any con- siderable time, the permanent set, and the extension due to the last previous application of the force, diminished. And in some instances it required some fifty repetitions to bring up the extension and set to the same points where they had been at the beginning of the period of rest ; thus indicating clearly that the specimen was partially restored, by the interval of rest, from the injury which it had previously received ; and that it endured a greater number of repetitions, owing to the intervals of rest, than it would have done had the repetitions succeeded each other continuously, and at short intervals of time. These results would therefore lead to the belief that a gun will endure a greater number of rounds if fired at intervals, with periods of rest of con- siderable length intervening, than if fired continuously. The indications of a single result ought not, however, to be regarded as conclusive ; and it is believed to be highly important that this experiment should be repeated, so as to leave no doubt as to the conclusions to be drawn from the results. Another feature of this experiment is, that the ultimate extension is less than it was in the corresponding specimen which was broken by adding 1000 lbs. to the force at each repetition; from which it is concluded that each repetition diminishes the ultimate extensibility of the specimen, and that any molecular disturbance diminishes the quantity of work which the specimen is capable of performing, or the work done, in effecting its rupture. AND QUALITIES OF CANNON POWDEE. 171 Capacity for Work. The term work done does not clearly convey the idea intended here, and has, besides, a technical meaning, signifying the product arising from multiply- ing the intensity of a force by the distance passed over by its point of application. Thus the work done in raising a weight through a given height is, in Mechanics, expressed by the product of the weight by the height to which it is raised. The term capacity for work is believed to be more appropriate, and is intended to signify ability to resist sudden applications of force ; it is analyti- cally expressed by the sum of the products arising from multiplying each successive increment of force, per unit of section, by the total corresponding extension per unit of length ; and is geometrically expressed by the area bounded by the curve expressive of the experimental results of any specimen, the axis of abscissa, and the ordinate of that point of this curve, corresponding to the force of rupture of the specimen ; or when the specimen is broken by a single application of force, and only the ultimate extension is measured, then the capacity for work would be approximately expressed by half the product of this force by the ultimate extension. And although this is not so accurate a measure of the capacity for work as that obtained by the successive application of forces of regularly increasing intensity with the measurement of the corresponding extensions, yet it serves very well as a means of comparing the capacity for work of one specimen with that of another ; is more easily obtained, and would, for a material in which equal increments of force give equal increments of extension, accurately express its capacity for work. The capacity for work is a compound quality, being composed of the tensile strength, and the extensibility of the specimen when it is subjected to a sudden force of extension only; but when the specimen is subjected to a transverse strain, or to the action of a central bursting force, then its capacity for work is composed of the two above named qualities, together with that of incompressibility. This last named quality having the effect, in transverse resistance, to place the neutral axis nearer to the compressed face of the specimen, and thus to increase the area of section which is subjected to a force of extension, and to 172 PKOPEKTIES OF METALS FOE CANNON, increase the distance from the neutral axis at which the resistance offered by this area acts, and consequently to increase the amount of resistance which the specimen is capable of offering. And in the resistance which a gun or other hollow cylinder offers to a central bursting force, the more incompressible the metal, the more perfectly will the force exerted upon the inner surface be transmitted to the surrounding and exterior portions, and the greater will be the tangential resistance of which the gun or cylinder will be capable ; and if only a portion of the interior length of the gun or cylinder be subjected to the action of the central force, then the greater will also be the transverse resistance developed. This compound quality furnishes the best known standard of excellence for a material intended to resist a single sudden application of force ; such as the shock of a falling body, or a single discharge of great bursting tendency, when made into a gun, or in resisting the force of the projecting charge, when made into shells. And in metal intended for shells, its capacity for work should consist rather of tensile strength and incompressibility than of extensibility ; but in metal intended for guns, machinery, or any other purpose requiring it to sustain a great number of repetitions of sudden applications of forces, of nearly constant intensity, then the work due to elasticity is of greater importance. The term work due to elasticity is intended to signify ability to resist repetitions of forces of less intensity than that which would produce rupture by a single application; and is expressed by the sum of the products arising from multiplying the intensity of each repetition of force, per unit of section, by half the corresponding total restoration, per unit of length, which the specimen undergoes on being entirely relieved from strain. The foregoing results also indicate that the application of any force, however small it may be, effects a certain amount of permanent injury to the specimen to which it is applied, or that there is no such property in cast iron as perfect elasticity, no matter how small may be the molecular disturbance. The production of a permanent set of measurable magnitude, by a force of 1000 lbs. per square inch, seems to leave no doubt that a permanent set is actually produced by a much less force, though our present means of measurement are incapable of detecting it. AND QUALITIES OF CANNON POWDEK. 173 All the foregoing results were obtained from specimens which had been subjected to previous strain and vibrations, being taken from fragments of broken guns. It is believed that more reliable results would be obtained by experimenting on specimens which had been subjected to no previous strains, nor molecular disturbances. For it is impossible to reason back to what would have been either the capacity for work, or the work due to elasticity of an unstrained specimen, by knowing to what extent these properties were possessed by that specimen, after it had been subjected to both strains and vibrations of unknown intensity and number. And although it is interesting to know to what extent these properties are possessed by the fragments of worn-out guns, yet it would be of far greater practical utility and importance to know the value of these properties in the new untried gun. And since it is impossible to study the properties and conditions as to strain, &c, of the iron after it has been made into guns, and before they are broken, it is believed to be of the utmost importance that before any lot of iron shall be pronounced fit to be cast into guns, either a sample gun, or a cylinder of equal diameter, and at least half the length of the gun, should be cast, and test specimens cut from it, and tested. The test gun or cylinder should be of the same diameter as the guns to be made, and should be made under the same circumstances which are to attend the preparation of the iron for, and the casting and cooling of, the guns themselves. Specimens thus obtained would afford reliable results ; and if accompanied by the powder proof, with service charges, of guns cast at the same heat, these results would become standards with which to compare other lots of iron, or other guns, and thus to determine beforehand, with sonle degree of certainty, the number of rounds which a gun will stand. And in advance or anticipation of the powder proof for that purpose, it is believed that a fair approximation to the difference in endurance of guns due to different bursting tendencies at each discharge, might be made by breaking a series of specimens cut from the same casting, by a series of repetitions of strains bearing certain definite relations to the tensile strength of the specimens. Say three specimens, by repetitions of a strain equal to 174 PEOPEETIES OF METALS FOE GANNON, half of their tensile strength, three by repetitions of a strain = f , three by repetitions of a strain = f , and three by repetitions of a strain = I of their tensile strength. Then, if we knew the bursting tendency to which a gun of the same quality of iron is subjected at each discharge, it would seem to warrant the prediction that the gun would endure the same number of repetitions as the specimen, of a force bearing the same ratio to its ultimate strength. This would undoubtedly be true if the repetitions were made in the same manner, and the specimens subjected to the same number and extent of vibrations, at each repetition, as the iron in the gun, and the gun were free from previous strain. The difference in endurance due to the difference in mode of application and effects of vibration, at each repetition, could be determined only by accompanying the proof of the test specimens with the powder proof of the guns. Of the Absolute Pressure of Gas in the Bore of a Gun. "With a view to determine the absolute pressure exerted upon the bore of a 42-pdr. gun in firing, under various circumstances, the following experiments were made, viz. : — A 42-pdr. gun was pierced through the cascabel, along the axis, and at intervals of two calibres along its side and perpendicular (except that nearest the muzzle) to its exterior surface, with holes .38 in. in diameter, and extending through to the bore. Concentric with these holes were bored others 1.5 in. in diameter, and 1.5 in. deep. These holes were tapped, and the housing which contained the indenting tool and the copper specimen to be indented, was screwed into one of them, when in use, the others being filled with plugs, tightly screwed in. The diameter of the indenting piston, on the the inner end of which the pressure of the gas was exerted, was 0.368 in. The following Plates show a section, through the axis, and a side elevation of the indenting apparatus. The indenting tool had a snug working fit in the housing. The hole in the housing shown at (c) and the recess around the stem of the indenting tool which it enters, were made for the purpose of letting out any FLATXJXZI^ *ta 3 ' yf7cZe7?./'/npnrsr/tif.r S{R — r)2L 6 + 3RrS(R + r) (R — r) 2 P 2pRrL 6 S(R — r) S2Z 6 r + 3R(R.+ r) (R—r) P Y or if h = height of cylinder of powder, and h' that of cylinder of shot, we have, by taking coefficient of pressure = 1250 lbs., and substituting for (p) its value . . , . 2500 h h'RrL e 1250 h h', m the above equation, e = 8 ^ B _ r) , 2 r £a +3 B{R + r)(R-r)l*\ as the general expression for the bursting tendency, 186 PEOPEETIES OF METALS FOE CANNON, The length of bore in which the maximum pressure is exerted does not generally exceed the value of (/) ; and since the transverse resistance will be fully developed for all values of (Z) less than {I), the foregoing equation will, 2500 hh' RrL for all such values of (L), become z = S (E — r) / 2 r L + 3 R {B + r) (R—r)V The bursting tendencies of our siege and garrison guns, assuming S = 20000 lbs., as determined by this formula, are as follows, viz. : — For the 24:-]}dr. gun, with 8 lbs. powder and one solid shot, supposing L = 13 inches, we have for bursting tendency at each discharge, z = .837 For Z%pdr. gun, with 8 lbs. powder and one solid shot, supposing L = 12 inches, - - - - - z = .656 For 4%-pdr. gun, with 10 lbs. powder and one solid shot, supposing L = 13 inches, ...... = .744 For S-inch gun, with 10 lbs. powder and one solid shot, supposing L = 12 inches, - - - - - z = .528 For 10-inch gun, with 14 lbs. powder and one solid shot, supposing L = 14 inches, - - - - - z = .498 It thus appears that our 24-pdr. is the most heavily strained gun, and that the 10-inch is the least strained gun at each discharge, of the two classes above referred to ; and if their endurance be not found in accordance with the above estimates, it is believed that the only reason why it will not, is, that the larger guns are more injured by contraction in cooling than the smaller. The bursting tendency of the 11-inch Navy gun, with 15 lbs. powder and 135 lbs. shell, supposing Z = 15 inches, is 2 = .477. The same formula gives the bursting tendency of a 15-inch gun of 46 inches exterior diameter, with 35 lbs. powder and 343 lbs. loaded shell, supposing L = 18 inches, g = .577, which is less than for the 32-pdr. gun. Repetition of Strain. A specimen from the Fort Pitt solid cast gun of 1857 broke at the 1956th repetition of a strain equal to three-fourths of its ultimate strength. The 42-pdr. gun is shown (above) to be subjected, at each discharge, to AND QUALITIES OF CANNON POWDER. 187 . 744 thousandths, or about three-quarters of its ultimate strength, and there is no reason, other than the difference in mode of application of the force, why that gun should not endure 2000 rounds, if uninjured in cooling. The 10-inch gun, as shown above, is subjected, at each discharge, to a strain of less than half its ultimate strength ; and two guns of this calibre have endured, without breaking, 2452 rounds each. These results leave no doubt, in my mind, that a 15-inch gun may be safely estimated to endure 1000 rounds of charges proportional to those above named. T. J. RODMAN, Capt. of Ordnance. REPORT EXPEEIMENTS CAPT. T. J. RODMAN, U. S. ORDNANCE DEPARTMENT, AT THE WATERTOWN ARSENAL, IN THE 2d HALF OF 1859, FOR DETERMINING THE PROPER QUALITIES OF IRON, EXTERIOR MODEL, ETC., FOR CANNON, WITH SPECIAL REFERENCE TO THE FABRICATION OF A 15-INCH GUN. REPORT EXPEEIMENTS MADE BY CAPTAIN T. J. EODMAN, AT THE WATEETOWN AESENAL, IN THE SECOND HALF OF 1859, FOE THE PUEPOSE OF DETEEMINING THE PEOPEE QUALITIES OF IEON, EXTEEIOE MODEL, ETC., FOE CANNON, WITH SPECIAL EEFEEENCE TO THE FABEICATION OF A 15-INCH GUN. For the purpose of determining the relation between the thickness of metal and the tangential resistance offered by hollow, open ended cylinders, the following experiments were made : — Three similar solid iron castings, of circular cross section, were cast, on end, in dry sand moulds, out of the same ladle of melted metaL These castings were of sufficient length to furnish six cylinders each, and one specimen each, five inches long, for tenacity. These castings were of different diameters, thus : — So that each cylinder should have the same amount to be turned off in reducing it to its proper exterior dimensions. They were all accurately 192 PKOPEETIES OF METALS EOK CANNON, bored to two inches, interior diameter, were 12 inches long, and varied in thickness from .5 in. to 3 in. by difference of .5 in. Each set in the tables was composed of one cylinder from each casting, and from corresponding parts. The castings were cast with their small ends down, and the specimens for tenacity were taken from the lower ends. The cylinders were burst with powder of .3 to .4 in. in diameter of grain, and their bursting pressures determined by the method of indentations heretofore explained ; they were placed in a strong, heavy housing of cast iron, and their ends firmly closed when fired. The annexed drawing (Plate No. 1) shows the arrangement referred to ; (a) large housing in which the cylinder is confined, (b) plugs for closing ends of cylinders, (c) shoe to receive end of plug, (d) key for forcing shoulders of plugs against end of cylinders, (C) cylinders, (g) gas ,checks to ends of cylinders, (v) vent through which the powder was fired, and (i) indenting apparatus. The plugs entered the ends of the cylinders . 1 inch, with a snug fit ; they were at first made of cast iron, and both broke ; that which rested in the shoe had the shoulders first cut in by the gas, as shown at (e,) and afterwards blown off entirely ; while that which received the indenting apparatus was split entirely through, and broken transversely. The plugs were then replaced with others made of cast steel, and the end of that into which the indenting housing was secured, twice pulled off at the bottom of the cavity which received the indenting housing, or on the section {h %). The castings from which the cylinders were prepared were designated A, B, and C, and the results obtained are recorded in the following tables : — Thickness of metal. Bursting pressure in lbs. per square inch. Mean bursting pressure per square inch. Difference. A B C 1 j 1st fire, 3 4 5 6 .5 inch. 1. 1.5 2. 2.5 3. 37609 37609 37137 62759 79916 94618 °77985 38077 39253 66753 80385 74803 89525 37842 38077 60640 80385 *89922 97879 37842 38313 63384 80229 92270 93702 471 25071 16845 12041 1432 Tenacity, 27950 24500 28150 26866 * Not burst. A (6) and B (5) neglected. I — PLATE J Sen '/{>// r// CD. Xccticn c/i.-LB ^ a W///M \M % ■ ■ §//£ ■40- fc ^ T-f- .5 m Wi- m Wasm, (p V AND QUALITIES OE CANNON POWDEE. 193 The following table compares the above mean result with those computed on the hypothesis that the strain diminishes as the square of the distance from the axis increases, and multiplying the tenacities by such coefficient as makes the actual and computed resistance equal for one inch thickness of metal : — Mean bursting pressure Computed bursting Number of sets. by experiment. Difference. pressure. Difference. 1 37842 25541 2 38313 471 38313 12772 3 63384 25071 46057 7744 4 80229 16845 51085 5028 5 92270 12041 54732 3647 6 93702 1432 57468 2736 The most remarkable features in these results are the very small difference in bursting pressure between the first and second sets, and that the bursting pressures, not only in these two sets, but in the whole series, are greater than required by the tenacity of the iron, even supposing the whole thickness of metal to resist uniformly as in tensile strain. The uniformity in the bursting pressures of the three specimens in each set Nos. 1 and 2, would seem to forbid the conclusion that these results are accidental. These two sets were charged, and burst, with equal charges of powder; and the fact that specimen (A 2) required two charges to burst it, indicates that the resistance of the 2d set was nearly equal to the bursting effort of the charge of powder used; viz., .75 lbs., while the bursting effort of the charge of powder must have been greatly in excess of the resistance offered by set No. 1, and the time during which the bursting force acted on set No. 1 must have been much less than that during which it acted on set No. 2. There is a certain amount of work to be done in rupturing a specimen of any solid substance, and the less the time occupied in accomplishing that work, the greater must be the force. The time during which the explosive force of the powder was occupied in bursting any one of the series of cylinders was exceedingly short, in comparison with that required for rupturing a specimen on the testing machine, and consequently the force was greater. These considerations offer the most probable explanation of these results 40 194 PROPERTIES OF METALS POP CANNON, which now presents itself to my mind, and further experiments alone can tell us whether or not they are sufficient. The column of differences in the foregoing table shows a constant, though irregular, diminution in the resistance added by equal increments of thickness, from the second to the fifth set, and from the fact that (C 5) endured, without bursting, a greater pressure than burst (B 6), we conclude that the last increment added very little to the reliable strength of the cylinder. And the fact that (B 5) burst with so much less pressure than (A 5 and C 5) and even less than (B 4), would seem to indicate that (B 5) ought to be neglected as accidental. Neglecting {B 5) gives a continuous, though still irregular, diminution of resistance offered by the successive equal increments of thickness after the first. And although these results are not sufficiently regular to warrant the deduction from them of the true law of diminution hi value of equal incre- ments of thickness, yet they leave no doubt of the fact that there is a very important diminution in that value as the thickness increases. AND QUALITIES OF CANNON POWDEE. 1Q5 For the purpose of determining the difference in pressure due to equal columns of powder behind equal columns of metal, when fired in guns of different diameters of lore, the following experiment was made : — ■ Three guns, one 42-pdr., (or 7-inch bore,) one 9-inch and one 11-inch bore, were prepared for experiment by reaming out the chambers of the 9-inch and 11-inch guns, (Navy guns — Dahlgreen's Model,) so as to conform in shape with the termination of bore of the Army 42-pdr., the radius of curvature of the surface joining the bottom and sides of the bore, bearing the same proportion to the diameter of bore in all of the tliree guns. These guns were all pierced with holes .4 in. diameter ; one at the bottom of the bore and six others at intervals of 14 inches from the bottom, towards the muzzle. To each of these holes was adapted a small housing for determining the pressure of the gas by the method of indentations. The charges of powder and the weight of projectiles were computed so that there should be the same weight of powder behind, and the same weight of metal in front of each square inch of area of cross section of the bore or surface of projectile passed by the gas in all the guns. The projectiles used were cylindrical, and accurately turned to such diameters that the area of the windage space should bear the same proportion to that of a cross section of the bore in all the guns. The data for pressure of gas at the bottom and six points along the bore of the gun were taken at each discharge, and the housings which held the specimens of copper to be indented were taken out, and the indenting pistons carefully cleaned after each discharge, so that all the circumstances of each discharge should be as nearly identical, one discharge with another, as possible. The charges were accurately weighed, a sufficient quantity of powder for ten rounds with each gun having been well mixed together, so as to insure identity in quality of the powder used in all these guns. The powder used was Hazard's, of 1858, proof range 280 yards. Cartridge blocks were used in all, and the cartridges fit snugly in the bore, so that there was no vacant space either around the cartridge or between the cartridge and the projectile. The velocity of the shot was determined by the use of M. Navez' electro- ballistic pendulum at each fire. The following tables exhibit the results obtained : — 196 PEOPEETIES OF METALS FOE CANNON, Table showing the velocity of shot, in feet, per second, and pressure of gas per square inch, in pounds, due to 5.13 lbs. of powder, and one cylindrical shot of 75.44 lbs. with .07 in. windage, fired in a 42-pdr. gun. Number Peessure at Diffekeht Distances fkom Bottom OF BOEE. Velocity. of Fires. At bottom. At 14 inches. At 28 inches. At 42 inches. At 56 inches. At 70 inches. At 84 inches. 1 922 38000 14000 8750 8000 8000 8750 13000 2 899 37750 16251 8250 6500 6250 7000 5750 3 853 41000 19000 8500 6500 7250 8750 6750 4 815 32500 16250 8250 6000 7750 5750 6000 5 941 34000 14000 8250 6000 6500 9000 5750 6 882 32750 16250 8500 6000 6000 8750 5750 7 856 32750 14750 8250 6500 6000 7000 5750 8 944 37500 16500 9000 6500 7500 8500 6500 9 1031 40000 14000 8000 6000 6000 8500 6000 10 894 38000 17500 8000 6750 7250 8500 6000 Mean, 904 36420 15850 8370 6470 6850 8050 6720 Table showing the velocity of shot, in feet, per second, and pressure of gas per square inch, in pounds, due to 8.48 lbs. of powder, and one cylindrical shot of 124.42 lbs. with .09 in. windage, fired in a 2-inch gun. Number Velocity. Pressure at Different Distances fkom Bottom OF BOEE. of Fires. At bottom. At 14 inches. At 28 inches. At 42 inches. At 56 inches. At 70 inches. At 84 inches. 1 952 38000 24000 » 22500 17000 24000 26000 20000 2 911 76000 20000 15500 15000 35000 19000 27000 3 867 67000 23000 27500 14000 26000 2G000 2L000 4 895 61000 29500 16000 14000 31000 16000 24000 5 Failed 73U00 1*000 14500 14000 30500 16250 20250 6 858 61000 20000 15500 15000 32000 21250 27000 7 870 67000 14500 18400 15000 31250 29000 2S500 8 932 73000 24500 18000 17000 28000 21250 19500 9 861 88000 19500 14500 14000 26000 19000 19500 10 845 67000 18000 15500 14000 31000 16000 21500 Mean, 888 67100 21100 17750 14900 29457 20970 22825 AND QUALITIES OF CANNON POWDEE. 197 Table showing the velocity of shot, in feet per second, and pressure of gas per square inch, in pounds, due to 12.67 lbs. of powder and one cylindrical shot of 186.3 lbs., with .11 in. windage, fired in an 11-inch gun. Pbessuee at Diefebent Distahces feom Bottom of Boee. Velocity. fires. At bottom. At 14 in. At 28 in. At 42 in. At 56 in. At 70 in. At 84 in. 1 Failed. 75000 23000 20000 18000 19000 29500 16000 2 917 86000 32500 25500 21500 27000 38000 20000 3 920 94000 31000 30000 21000 27000 31500 22500 4 934 56500 28000 25000 22000 29000 34000 25500 5 Failed. 95000 29500 31500 29000 27000 29500 22500 6 920 100000 36000 35500 19000 31500 38000 31000 7 1051 95000 26000 25000 18000 31500 28000 29000 8 971 94000 28000 30000 26500 29000 38000 29000 9 767 80000 31000 25500 18250 31500 38000 26000 10 935 94000 28000 30000 21000 31500 34000 29000 Mean, . 927 86750 29300 27800 22420 28400 33850 25050 Table showing the velocity of shot, in feet per second, and pressure of gas per square inch, in pounds, due to equal columns of powder behind equal columns of metal, when fired in guns of different diameter of bore, each result being a mean of ten fires. u tn c3 a Weight of Charge. Weight of Shot. o QJ > Pressure at Different Distances from Bottom of Bore. a ° Q At bottom. At 14 in. At 28 in. At 42 in. At 56 in. At 70 in. At 84 in. 7 in. 9 11 .07 .09 .11 5.13 lbs. 8.48 12.67 74.44 lbs. 121.42 186.03 904 888 927 36420 67100 86750 15850 21100 29300 8370 17750 27800 6470 14900 22420 6850 29475 28400 8050 20970 33850 6720 22825 25050 The points most worthy of note in these results, are the very marked increase in pressure of gas as the diameter of bore increases; and that the indications of pressure are greater at 56 inches, 70 inches, and 84 inches, than at 42 inches, especially in the 9-inch and 11-inch guns. The cause of the difference in pressure developed in these guns of different diameters of bores, is believed to be mainly due to the greater heat developed by the combustion of the larger mass of powder in the large than in the smaller calibre ; and perhaps, also, to the different products of combustion formed under this increased temperature and pressure, and partly to the greater cooling surface in proportion to the weight of charge in the small than in the larger calibre. 198 PEOPEETIES OF METALS FOE CANNON, From the fact that there was no vacant space around the cartridge, all the powder that was burned before the shot moved must have been burned in its own volume, and the maximum pressure would probably be reached before the charge would be nearly consumed, and a greater proportional quantity of gas would escape through the vent of the smaller calibre. The vent in the 42-pdr. was something larger than those in the 9-inch and 11-inch guns, it having been fired some 200 rounds before this series was commenced. The vents in the 9-inch and 11-inch guns were new, and .2 inches in diameter. It is impossible that there could have been any greater pressure beyond 42 inches than at that point ; and it cannot be doubted that the pressure diminishes, as the space occupied by the products of combustion of a given weight of powder increases. The greater indications of pressure beyond than at 42 inches, is believed to be due to the more violent and sudden contraction of the metal in the thin than in the thick part of the gun, and to the greater number of vibrations to which the thin portions are subjected at each discharge. The point 42 inches from the bottom of the bore falls on the rapid taper from the cylinder to the chase, which occurs in these guns, and it is believed that the true pressures are indicated with sufficient accuracy up to this point. For in the thick part of the gun the pressure is much less rapidly developed, and subsides much more gradually, the contained gas forming an elastic cushion, which would, if the bore were long enough, allow this part of the gun to return from its strained to its free condition, without any vibration at all. While in the model used in these experiments the pressure is almost instantaneously developed, and as suddenly subsides in the chase of the gun, so that while the indenting piston is on its way outward, it is suddenly met by the returning specimen, which is drawn in along with the housing by the contraction of the gun, with such violence as to amount, in effect, to a smart blow of the indenting tool against the specimen. Close examination shows a number of marks or cuts of the indenting tool on the specimen in this part of the gun, caused by the tool not striking in the same place at each vibration of the gun. Further evidence of this action is found in the fact that the specimens frequently jarred out of the housing in this part of the gun, but never in the thick part. The same thing occurs, though in a much less degree, in the AND QUALITIES OF CANNON POWDEB. 199 42-pdr. The chase of this gun is much thicker in proportion than in the 9-inch and 11-inch guns, and the taper much more gradual. It is believed that true indications of pressure along the whole length of bore can only be obtained, by this method, from a gun of great and uniform thickness along its entire length ; so that there should be the least possible enlargement of its exterior diameter, and consequently the least vibration at each discharge. For the purpose of determining the difference in pressure of gas, and velocity of shot, due to equal weights of powder of the same quality in all respects, except in diameter of grain, fired from the same gun, the following experiments were made : — Powder of the same quality in every respect, except in diameter of grain, was prepared by the Messrs. Dupont. The powder varied in size of grain as follows : — .1 in., (ordinary cannon powder,) .15 in., .2 in., .25 in., .3 in. and .4 in. diameter of grain. Five rounds of 8 lbs. each were fired, with each size of grain, from the 42-pdr. gun, with one solid shot and sabot, the same shot being used in all the firing. The pressure of gas at the bottom of the bore, and at six other points along the bore, at intervals of 14 inches from the bottom, was determined at each discharge by the method of indentations, and the velocity of the shot at each fire was determined by the use of M. Navez' electro-ballistic pendulum. The eprouvette range of the different sized grains of powder was carefully determined, each result being a mean of three rounds. The results are given in the following tables : — 200 PEOPEETIES OF METALS FOE CANNON, Table showing the velocity of shot, in feet per second, and pressure of gas per square inch, in pounds, due to 8 lbs. of powder, [grain .1 in. diameter,) and one solid shot and sabot, fired in a 42-pdr. gun. Number of Velocity. PRESSURE AT DIFFERENT DISTANCES FROM BOTTOM OF BORE. Fires. At bottom. At 14 in. At 28 in. At 42 in. At 56 in. At 70 in. At 84 in. 1 2 3 4 5 Failed. Failed. 1242 1235 1307 49000 49000 45000 49000 49000 51000 47000 49500 51000 60000 13500 13000 10000 14000 17000 13000 7000 8500 7000 7000 10000 6000 5500 6000 10250 10500 6750 5750 6750 6750 7500 6500 5750 6500 7250 Mean, . 1261 48200 51800 13500 8900 7550 7300 6700 Table showing the velocity of shot, in feet per second, and pressure of gas per square inch, in pounds, due to 8 lbs. of powder, [grain .15 in. diameter,) and one solid shot and sabot, fired in a 4:2-pdr. gun. Number of Velocity. PRESSDEE AT DIFFERENT DISTANCES FROM BOTTOM OF BORI Tires. At bottom. At 14 in. At 28 in. At 42 in. At 56 in. At 70 in. At 84 in. 1 2 3 4 5 Failed. 1259 Failed. 1230 1215 58000 45000 45000 45000 43000 50000 50000 49500 47500 44500 17000 13000 17000 13500 17000 11500 7000 7000 7000 7000 10250 7000 6750 7000 5750 11000 6750 5750 5750 5750 10000 6500 6250 6500 5750 Mean, . 1235 47200 48300 15500 7900 7350 7000 7000 Table showing the velocity of shot, in feet per second, and pressure of gas per square inch, in pounds, due to 8 lbs. of powder, [grain .2 in. diameter,) and one solid shot and sabot, fired in a 42-pdr. gun. Number of Velocity. PRESSURE AT DIFFERENT DISTANCES FROM BOTTOM OF BORE. Fires. At bottom. At 14 in. At 28 in. At 42 in. At 56 in. At 70 in. At 84 in. 1 2 3 4 5 1257 Failed. 1156 1184 Failed. 49000 43500 36000 38000 45000 42500 50000 40500 47500 50000 10000 12750 12750 12750 12750 4750 5750 6000 4750 6250 5000 5750 6000 5000 5000 4750 4850 5250 4850 5250 5500 5500 6500 5500 5500 Mean, 1199 42300 46100 12200 5500 5350 4970 5700 AND QUALITIES OF CANNON POWDEE. 201 Table showing the velocity of shot, in feet per second, and pressure of gas per square inch, in pounds, due to 8 lbs. of powder, {grain .25 in. diameter,) and one solid shot and sabot, fired in a 4:2-pdr. gun. Number of Velocity. FBESSUBB A3 DIFFERENT DISTANCES FBOM BOTTOM OF BOBE. Fires. At bottom. At 14 in. At 28 in. At 42 in. At 56 in. At 70 in. At 84 in. 1 2 3 4 5 1141 1141 Failed. 1119 1202 41500 36000 36000 43000 36000 40500 44500 47500 40000 49500 15000 14000 10000 15000 10000 4750 5750 5750 4750 4000 5000 4000 5750 5750 5750 4250 4250 4750 4750 5250 5000 5750 5000 5500 6250 Mean, . 1151 38500 44400 12800 5000 5250 4850 5500 Table showing the velocity of shot, in feet per second, and pressure of gas per square inch, in pounds, due to 8 lbs. of powder, {grain .3 in. diameter,) and one solid shot and sabot, fired in a 4t2-pdr. gun. Number of Velocity. PRESSURE Al DIFFERENT DISTANCES FEOM BOTTOM OF BOBE. Fires. At bottom. At 14 in. At 28 in. At 42 in. At 56 in. At 70 in. At 84 in. 1 2 3 4 5 1185 Failed. 1152 1088 1157 29000 33500 38000 38000 33500 32250 32750 32000 32000 30000 9500 15000 14000 10000 17000 6000 6250 3750 4300 3750 4500 5000 4000 4250 4000 4250 4250 4750 4150 4500 5000 5500 4500 5000 4500 Mean, . 1146 34400 31800 13100 4810 4350 4380 4900 Table showing the velocity of shot, in feet per second, and pressure of gas per square inch, in pounds, due to 8 lbs. of powder, {grain A in. diameter,) and one solid shot and sabot, fired in a ^2-pdr. gun. Velocity. PBESSDBE AT DIFFERENT DISTANCES FBOM BOTTOM OF BOB! Fires. At bottom. At 14 in. At 28 in. At 42 in. At 56 in. At 70 in. At 84 in. 1 2 3 4 5 Failed. 1157 Failed. 1197 1208 28500 38000 36000 28500 28500 32000 33000 32000 30000 32750 10500 13000 17000 13000 13500 4000 4000 4000 4300 4000 4500 4500 4000 3750 4000 4750 4250 5250 4200 4300 5500 5500 6250 4250 4250 Mean, . 1187 31900 31950 13400 4060 4150 4550 5150 202 PEOPEETIES OF METALS EOE CANNON, Proof Range, ly Eprouvette, of Powder used in the Experiments whose results are recorded in the foregoing tables, as determined August 15, 1859, at Watertown Arsenal. DUPONT'S, OF 1859, OF DIFFERENT SIZED GEAINS. PBOOP BANGES. DlAMETEES OF Gkains. 1st Fire. 2d Fire. 3d Fire. Mean. .1 inch. 314 yards. 295 yards. 305 yards. 305 yards. .15 284 285 282 284 .2 263 252 257 257 .25 154 154 170 159 .3 90 73 88 84 .4 52 51 54 52 Hazard's, of 1858, used in determining difference of pressure of equal columns of powder behind equal columns of metal, in guns of different diam- eter of bore : — 1st fire, 2d " 3d " 296 yards. 284 " 261 " Mean, 280 yards. AND QUALITIES OF CANNON POWDEE. 203 ■S ■5m K c e *» Co ^j ^ 5* e ^ s oo CO -to cs § 'a *§ 03 «g sS s. e s <» a a CO I* CO o l-sg 8 *1 CO EH CO ^ 00 at ^ o o o o o o pj O O © O O lO SOS1QOJH 5 CD t~ »0 IQ "* "O •♦* < m «3 00 10 — moroconio o O t>- t* ^ ^ ^ ^ w b- Pn -»3 o a CO o "9 o o o o o o H o .g io o io r. io io H lO CO CO CI CO >— < O CD t* t* IO o i< ■* M lO s o « ■ fH 0J w ■g o © o © o o 3 a OOOOHto jz; •m C5 C5 IO © 00 © ■< •* C0l>lOlO^'* B ■43 s 5§ § ►S m ■* t~ en ■* cq r-i £ OCOQIOCOO (4-1 o CO CM Ol i-H o o PM CH «M o >» HlflffiHCOS ^S 9 CO CO Oi iQ -^ 00 o o r£3 to (M 01 i-i i—l i-h r-l ffl t> ■♦= CM o ■43 ^3 CQ bo a CO CO CO CO CO CO 2 o i o c3 O a o J3 o JJ $£ a to o f^n rO c ^3 ^ c3 s CD 93 & Cm O O +-» c Ti O fl ■■0 s n cv; •■» CD 6f) O 3 CO s CD VI ^ f3 a & .n 02 > 204 PEOPEETIES OF METALS FOE CANNON, The point of greatest interest in these results, is the fact that the maximum pressure of gas diminishes as the diameter of grain increases, in a much greater ratio than the squares of the corresponding velocities ; thus showing, conclusively, that the velocities due to our present charges of small grained powder may be obtained with a greatly diminished strain upon the gun, by the use of powder properly adapted in size of grain to the calibre and length of bore in which it is to be used ; or that increased velocities may be thus obtained, without any increase of the strain to which our guns are now subjected, in obtaining our present velocities. The results also show the impropriety of taking the eprouvette range as an indication of the projectile force of powder which is to be used in guns of any considerable length and calibre. Powder intended to be used in mortars should be proved in mortars, and that for guns should be proved in guns. It will be observed that the minimum indications of pressure in these experiments are at 70 inches instead of 42 inches, as in the previously described series; this is believed to be due to the greater charges used in this than in the other series. For the purpose of determining the pressure exerted by exploded gun- powder, when the products of combustion occupy a certain number of times the volume occupied by the powder before combustion, the following experiments were made : — A shell was cast of the best gun iron, with a spherical cavity of 5 inches diameter, and 6.25 in. thickness of walls. This shell was pierced with a hole .4 in. diameter, and concentric with that into which the housing or indenting apparatus was screwed, and the hole left by the core stem,. 90° from it, was reamed out to the depth of 2.5 in., and tapped so as to receive a plug 1.5 in. diameter. This plug had a hole .4 in. bored out along its axis, and was tapped at its inner end so as to receive another small plug, which was pierced along its axis with a hole .1 in. diameter, which was the only outlet for the products of combustion when the charge was fired. This plug was replaced by a new one, when the hole in it was burned out to .125 in. The interior volume of the shell, with plug and housing screwed rZAT£JT. /y s. &*/-l>.00 3 CDCDHCOiOOOCO'^SlMS'tCOCOCOcOCDOO £ OHC0C0^WIM>H(MLC:S«DC00ir-ib.ioW»O HHHHHHHHWCqiMlMW^COCDCOCOIMtN H COt^ o5 rQ aJ t-« .OOOiOOOOO S 1 1 ! 1 1 1 looiob-oioool 1 1 | ] £ COtMWCOrHCONCO HlMCvllMCOW-^t, »o CO rQ .a , , . . ,0,00000000 1 1 1 1 loiooioiooioool I I l I IO 00(MCOiOCOCDlMM i— i HCq(M(MC^CO^|> © ooooooooooiooooooooo & 00*0000000»0OSCO(NCO(NSCDrH t3 CO i— i CO OOOOOOOOOOiOiOOOOOOOO § o3 •-H OOOiOiOOi-T iOOOC^b-»OXOOOOOO r Q W ONCOCO^lOsOtDHIMioccDCOCDlNCOiOO^ W 02 ° oi & l„ ft OSCO-^iflCOOSSCOOiOOOCOOOSOO bD o3 OCONWSCOONCDKOWilOOCOOOffiOO | 6 NbtiCOCOOJOOHW^iosOWcOiOCDOO AND QUALITIES OF CANNON POWDER. 207 The above results show that the pressure increases in a higher ratio than that of the volumes of powder ; it being, for the larger charges, almost as the squares of the volumes. And, as indicated by the 42-pdr. 9-inch and 11-inch guns, there is no doubt that, with a constant ratio of powder to the space in which it is burned, the pressure would increase with the charge of powder used ; and it is deemed highly important to know in what ratio.* These results, compared with those obtained from the 42-pdr. gun with 5.13 lbs. of powder, indicate that this charge was entirely consumed in that gun before the shot was 2 calibres from the bottom of the bore, and that the pressure was not truly indicated beyond 42 inches, or 6 calibres, in that gun. They also show a marked difference in the quantities of gas evolved from equal weights of Hazard's and of Dupont's powder, the latter yielding the greater quantity. They also show the effect of prolonging the time during which a given pressure acts ; the 11-inch gun having endured 10 rounds, with a mean pressure of 86750 lbs. per inch, without bursting, while tbe shell li calibre thick, and of a better form to resist pressure, but on which tbe pressure was longer exerted, (the whole of the gas escaping through a 1-inch diameter orifice,) burst under a pressure of 42500 lbs. per inch. The iron in the shell was of good quality, but there were some defects from cavities developed by the bursting which were not before discoverable, but not sufficient to account for the difference in endurance of the shell and gun above referred to. It is believed that the most reliable tests for tbe relative volumes of gas evolved by equal volumes of different kinds of powder, and consequently of their relative merits, would be to determine the pressure due to the combustion of equal volumes, or equal weights of the different kinds of powder to be tested in the same space. Pressure due to Unequal Volumes of Powder when Burned in Spaces bearing a Con- stant Ratio to those Volumes. The foregoing experiments, taken in connection with those of the 42-pdr., 9-inch and 11-inch guns, where equal columns of powder were fired behind equal columns of metal, were thought to indicate that, with a constant ratio * Results at page 208 of this Report modify this conclusion. 208 PEOPEETIES OF METALS EOB CANNON, of powder to the space in which it is burned, the pressure would increase with the mass of powder burned. With a view to determine this point, four shells of different interior capacities were cast and prepared in the same manner as that used in the fore- going series, and shown in Plate 2. The capacities of these shells were all accurately determined by filling them full of powder, while being struck with a heavy mallet, and then emptying them, and weighing their contents separately. Each shell was then charged with one-fourth of the weight of powder which it had been thus found to contain, and fired ; the pressure being determined by the method of indentations, the gas escaping at an orifice one-tenth of an inch in diameter. The shells were all carefully washed out with warm water, and well dried after each fire. The following table exhibits the results obtained : — Number Weight of Charge. PRESSURE OF GAS, IN POUNDS, PER SQUARE INCH. Shells. 1st Fire. 2d Fire. 3d Fire. 4th Fire. 5th Fire. Mean. 1 2 3 4 1772 grs. 3610 5742 11817 9248 10019 10019 9762 9762 10276 8734 9762 10019 9762 8734 10019 8734 9248 8734 10019 9248 8734 7707 10019 9402 9608 8786 9916 These results indicate that where the volume of powder bears a constant ratio to the space in which it is burned, the pressure due to different masses of powder will be sensibly uniform, and consequently that they do not har- monize with the results obtained from the 42-pdr. 9-inch and 11-inch guns above referred to. I am unable to assign any reason at all satisfactory to myself, in explanation of these apparently conflicting results ; and, as the question is one of primary importance, it is believed that further experiments should be made, with a view to its proper and true solution. Of the Absolute Pressure of Powder when burned in its own Volume. The results at page 192, obtained from bursting cylinders of different thick- ness, with equal charges of powder, indicate, from the fact that those of .5 in. thick gave the same indications of pressure as those of 1 in. thick, that time AND QUALITIES OF CANNON POWDEE. 209 is required for the rupture of any mass of iron, though, the rupturing force may be greatly in excess of the resistance of that mass ; and that, by the method of indentations, the intensity of the bursting force may be registered before the rupture of the mass is accomplished. Following these indications, it was thought that we might with certainty establish a higher inferior limit than has heretofore been certainly determined for the pressure due to the combustion of gunpowder in its own volume. Accordingly, six shells, three of 12-inch and three of 1 3-inch exterior diam- eter, were cast and prepared for experiment. Those of 12-inch exterior diameter were intended to hold one pound of powder, and those of 13-inch, two pounds. Fire was communicated to the powder through an orifice one-tenth of an inch in diameter, and this orifice was the only outlet for the gas. The pressure was determined by the method of indentations, as before described. The shells were all filled full, being smartly rapped with a heavy mallet while the powder was being poured in, and their contents accurately deter- mined by weight. The results obtained are recorded in the following table : — DIAMETERS OP SHELLS. NuMBEB. OF Contents of Shells. Pressure of gas per square inch. Shells. Interior. Exterior. Kemabks. 1 3.85 in. 12 in. 5910 grs. 185000 lbs. In copper specimen. 2 3.85 12 6840 113040 In wrought iron specimen. 3 3.85 12 6560 107900 In wrought iron specimen. 4 4.85 13 9752 71932 In wrought iron specimen. 5 4.85 13 11250 100190 In wrought iron specimen. 6 4.85 13 14000 133590 In copper specimen. The edges of the indentations in the wrought iron specimens were much more raised or " burred " than those made in copper, and much more, also, than those made by the testing machine in the same iron. And it is believed that owing to this cause the machine indicates, with the iron specimens, a less pressure than was exerted by the gas. Iron was used instead of copper, for the reason that the pressure exerted at the first fire of this series forced the indenting tool into the copper specimen up to about the limit of its width, or indenting taper, and cut entirely across the specimen. At the first fire the part of the housing that screwed into the shell pulled 63 210 PEOPEETIES OF METALS FOE CANNON, out the metal of the shell into which it was screwed, the metal thus pulled out being broken, and not remaining attached to the housing. This circumstance left it doubtful whether the housing had been detached from the shell before or at the time it burst, and consequently whether or not the full bursting pressure had been exerted upon the indenting piston. In order to remove this doubt, the housing was, for the subsequent fires, screwed into the shells 2.5 in., instead of 1.5, the hold of that which pulled out, and no others pulled out ; but all burst through the hole into which the housing was screwed. This arrangement, however, increased by one inch the distance from the inner end of the indenting piston to the mouth of the housing ; and this aperture, that in which the piston moves, being only .352 in. diameter, it is apprehended that the powder may have been so tightly packed into it before being ignited, as to prevent, by its friction, the full pressure on the interior of the shell from being exerted upon the indenting piston. The friction of the gas check against the sides of the orifice in which it moves, prevents the full force of the gas from being exerted upon the indenting piston; and the more slowly applied pressure exerted by the machine than that exerted by the powder, ought to make an equal indentation with something less pressure ; so that I should feel perfectly safe in fixing the inferior limit of the pressure per square inch, due to the combustion of gunpowder in its own volume, at, in round numbers, 200000 pounds. Further experiments would render the results more certain; but would not, it is believed, reduce this estimate. The powder used in these experiments was Dupont's, of 1859, .1 in. grain, procured for experimental purposes. This series of experiments was made in March, 1860. AND QUALITIES OF CANNON POWDER. 211 Table showing the extension, restoration, and permanent set per inch, in length, caused by the undermentioned weights, per square inch of section, acting upon a solid cylinder 35 in. long and 1.335 in. diameter, taken from near the surface of the bore of 42-pdr. gun, No. 336, cast hollow, at the Fort Pitt Foundry, of Bloomfleld iron {No. 2 pig), and burst at the 4$lst round, with 10 lbs. powder, and one solid shot. Weight per square inch of section. Extension per inch in length. First difference. Restoration _ per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .000017 .000017 2000 .000038 .000021 .000038 .000021 - _ 3000 .000074 .000036 .000074 .000036 - — 4000 .000131 .000057 .000131 .000057 — — 5000 .000188 .000057 .000188 .000057 — _ 6000 .000246 .000058 .000243 .000055 .000003 — 7000 .000291 .000045 .000286 .000043 .000005 .000002 8000 .000366 .000075 .000358 .000072 .000008 .000003 9000 .000420 .000054 .000406 .000048 .000014 .000006 10000 .000502 .000082 .000479 .000073 .000023 .000009 11000 .000582 .000080 .000553 .000074 .000029 .000006 12000 .000666 .000084 .000623 .000070 .000043 .000014 13000 .000762 .000096 .000385 .000062 .000077 .000034 14000 .000882 .000120 .000768 .000083 .000114 .000037 15000 .001011 .000129 .000854 .000086 .000157 .000043 16000 .001140 .000129 .000926 .000072 .000214 .000057 17000 .001331 .000191 .001029 .000103 .000302 .000088 18000 .001528 .000197 .001106 .000077 .000422 .000120 19000 .001758 .000260 .001191 .000085 .000597 .000175 20000 .002160 .000372 .001317 .000126 .000843 .000246 21000 .002600 .000440 .001437 .000120 .001163 .000320 22000 .003217 .000617 .001557 .000120 .001660 .000497 23000 .004030 .000823 .001692 .000135 .002328 .000668 24000 .005128 .001098 .001757 .000065 .003371 .001043 25000 .006600 .001472 .001800 .000043 .004800 .001429 26000 Broke betw een 25000 and 26000 lbs., a nd failed to ge t ultimate ext ension. 212 PKOPEKTIES OE METALS EOE CANNON, Table showing the extension, restoration, and permanent set per inch, in length, caused by the undermentioned weights, per square inch of section, acting upon a solid cylinder 35 in. long and 1.382 in. diameter, taken from near the middle of thickness of 42-pdr. gun, No. 336, cast hollow, at Fort Pitt Foundry, of Bloom- field iron, and burst at the 49 1st round, with 10 lbs. powder and one solid shot. Weight per square inch of section. Extension per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .000020 .000020 _ 2000 .000070 .000050 .000070 .000050 — - 3000 .000126 .000056 .000126 .000056 - - 4000 .000176 .000050 .000172 .000046 .000004 - 5000 .000210 .000064 .000230 .000058 .000010 .000006 6000 .000322 .000082 .000306 .000076 .000016 .000006 7000 .000380 .000058 .000358 .000052 .000022 .000006 8000 .000450 .000070 .000428 .000070 .000022 .000000 9000 .000514 .000064 .000474 .000046 .000040 .000018 10000 .000590 .000076 .000530 .000056 .000060 .000020 11000 .000674 .000084 .000572 .000042 .000102 .000042 12000 .000776 .000102 .000654 .000082 .000122 .000020 13000 .000874 .000098 .000724 ,000070 .000150 .000028 14000 .001000 .000126 .000792 .000068 .000203 .000058 15000 .001136 .000136 .000856 .000064 .000280 .000072 16000 .001304 .000168 .000942 .000086 .000362 .000032 17000 .001542 .000238 .001026 .000084 .000516 -000154 18000 .001800 .000258 .001086 .000060 .000714 .000198 19000 .002170 .000370 .001186 .000100 .000984 .000270 20000 .002714 .000544 .001320 .000134 .001394 .000110 21000 .003514 .000800 .001394 .000074 .002120 .000726 22000 .004542 .001028 .001478 .000034 .003064 .000944 23000 .0054S4 .000942 .001070 —.000408 .004414 .001350 24000 .006948 .001464 .000934 —.000036 .006014 .001600 25000 .011314 .004366 Broke beta ecn 24000 ant 1 25000 lbs. AND QUALITIES OF CANNON POWDEE. 213 Table showing the extension, restoration, and permanent set per inch, in length, caused by the undermentioned iveights, per square inch of section, acting upon a solid cylinder 35 inches long and 1.382 in. diameter, taken from near the exterior surface of 42-pdr. gun, No. 336, cast hollow, at the Fort Pitt Foundry, of Bhomfteld iron, and hurst at the 491s^ round, with 10 lbs. powder and one solid shot. Weight Extension First difference. Restoration First difference. Permanent set First difference. per square inch of section. per inch, in length. per inch in length. per 'inch in length. 1000 lbs. .000026 .000026 _ 2000 .000094 .000068 .000092 .000066 .000002 — 3000 .000148 .000054 .000142 .000050 .000006 .000004 4000 .000204 .000056 .000196 .000054 .000008 .000002 5000 .000260 .000056 .000250 .000054 .000010 .000002 6000 .000316 .000056 .000300 .000050 .000016 .000006 7000 .000380 .000074 .000360 .000060 .000020 .000004 8000 .000448 .000068 .000422 .000062 .000026 .000006 9000 .000522 .000074 .000488 .000066 .000034 .000008 10000 .000596 .000074 .000554 .000066 .000042 .000008 11000 .000674 .000078 .000594 .000040 .000030 .000038 12000 .000768 .000094 .000658 .000064 .000110 .000030 13000 .000886 .000118 .000736 .000078 .000150 .000040 14000 .001028 .000142 .000808 .000072 .000220 .000070 15000 .001176 .000148 .000868 .000060 .000308 .000088 16000 .001380 .000204 .000954 .000086 .000426 .000118 17000 .001670 .000290 .001042 .000088 .000628 .000202 18000 .001928 .000258 .001108 .000066 .000820 .000192 19000 .002556 .000628 .001254 .000146 .001302 .000482 20000 .003702 .00L146 .001362 .000108 .002340 .001038 21000 .004160 .000458 .001410 .000048 .002750 .000410 22000 .005874 .001714 .001584 .000174 .004290 .001540 23000 .009102 .003228 .003142 .001558 .005960 .001670 24000 .009760 .000658 Broke betw een 23000 anc I 24000 lbs. 214 PEOPEETIES OF METALS FOE CANNON, Table showing the compression, restoration, and permanent set per inch, in length, caused by the undermentioned tveights, per square inch of section, acting upon a solid cylinder 10 in. long and 1.382 in. diameter, taken from near the surface of the lore of 4c2-pdr. gun, No. 336, cast hollow, at the Fort Pitt Foundry, of Bloomfield iron, and burst at the 4§\st round, with 10 lbs. powder and one solid shot. Weight per square inch of section. Compression per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First • difference. 1000 lbs. .00006 .00006 _ 2000 .00010 .00004 .00010 .00004 - - 3000 .00015 .00005 .00015 .00005 - — 4000 .00021 .00006 .00021 .00006 - — 5000 .00025 .00004 .00024 .00003 .00001 - 6000 .00031 .00006 .00030 .00006 .00001 .00000 7000 .00038 .00007 .00036 .00006 .00002 .00001 8000 .00046 .00008 .00043 .00007 .00003 .00001 9000 .00054 .00008 .00050 .00007 .00004 .00001 10000 .00062 .00008 .00056 .00006 .00006 .00002 11000 .00068 .00006 .00061 .00005 .00007 .00001 12000 .00076 .00008 .00067 .00006 .00009 .00002 13000 .00086 .00010 .00076 .00009 .00010 .00001 14000 .00092 .00006 .00080 .00004 .00012 .00002 15000 .00101 .00009 .00084 .00004 .00017 .00005 16000 .00108 .00007 .00088 .00004 .00020 .00003 17000 .00119 .00011 .00095 .00007 .00024 .00004 18000 .00129 .00010 .00099 .00004 .00030 .00006 19000 .00139 .00010 .00103 .00004 .00036 .00006 20000 .00150 .00011 .00107 .00004 .00043 .00007 21000 .00161 .00011 .00111 .00004 .00050 .00007 22000 .00172 .00011 .00117 .00006 .00055 .00005 23000 .00185 .00013 .00121 .00004 .00064 .00009 24000 .00200 .00015 .00124 .00003 .00076 .00012 25000 .00217 .00017 .00130 .00006 .00087 .00011 26000 .00237 .00020 .00133 .00003 .00104 .00017 27000 .00257 .00020 .00136 .00003 .00121 .00017 28000 .00283 .00026 .00141 .00005 .00142 .00021 29000 .00309 . .00026 .00142 .00001 .00167 .00025 30000 .00342 .00033 .00140 —.00002 .00202 .00035 31000 .00389 .00047 .00147 +.00007 .00242 .00040 32000 .00426 .00037 .00143 —.00004 .00283 .00011 33000 .00495 .00069 .00156 +.00013 .00339 .00056 34000 .00576 .00081 .00166 .00010 .00410 .00071 35000 .00644 .00068 .00165 —.00001 . .00479 .00069 AND QUALITIES OF CANNON POWDEK. 215 Table showing the compression, restoration, and permanent set per inch, in length, caused hy the undermentioned iveights, per square inch of section, acting upon a solid cylinder 10 inches long and 1.382 inches diameter, taken from near the middle of thickness of AH-pdr. gun, No. 336, cast hollow, at the Fort Pitt Foundry, of Bloomfleld iron, and hurst at the 491s^ round, with 10 lbs. powder and one solid shot. Weight per square inch of section. Compression per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .00006 .00006 _ _^ _ 2000 .00011 .00005 .00011 .00005 - - 3000 .00016 .00005 .00016 .00005 - - 4000 .00020 .00004 .00019 .00003 .00001 - 5000 .00027 .00007 .00025 .00006 .00002 .00001 6000 .00034 .00007 .00032 .00007 .00002 .00000 7000 .00040 .00006 .00037 .00005 .00003 .00001 8000 .00047 .00007 .00043 .00006 .00004 .00001 9000 .00054 .00007 .00049 .00006 .00005 .00001 10000 .00066 .00012 .00060 .00011 .00006 .00001 11000 .00070 .00004 .00062 .00002 .00008 .00002 12000 .00079 .00009 .00070 .00008 .00009 .00001 13000 .00085 .00006 .00074 .00004 .00011 .00002 14000 .00093 .00008 .00079 .00005 .00014 .00003 15000 .00100 .00007 .00083 .00004 .00017 .00003 16000 .00109 .00009 .00089 .00006 .00020 .00003 17000 .00116 .00007 .00093 .00004 .00023 .00003 18000 .00125 .00009 .00098 .00005 .00027 .00004 19000 .00135 .00010 .00103 .00005 .00032 .00005 20000 .00144 .00009 .00107 .00004 .00037 .00005 21000 .00156 .00012 .00110 .00003 .00046 .00009 22000 .00163 .00013 .60116 .00006 .00053 .00007 23000 .00180 .00011 .00118 .00002 .00062 .00009 24000 .00194 .00014 .00122 .00004 .00072 .00010 25000 .00212 .00018 .00126 .00004 .00086 .00014 26000 .00232 .00020 .00130 .00004 .00102 .00016 27000 .00252 .00020 .00132 .00002 .00120 .00018 28000 .00286 .00034 .00139 .00007 .00147 .00027 29000 .00317 .00031 .00142 .00003 .00175 .00028 30000 .00357 .00040 .00147 .00005 .00210 .00035 31000 .00411 .00054 .00149 .00002 .00262 .00052 32000 .00493 .00082 .00154 .00005 .00339 .00077 33000 .00560 .00067 .00157 .00003 .00403 .00064 34000 .00682 .00122 .00209 .00052 .00473 .00070 35000 .00724 .00042 .00063 —.00146 .00661 .00188 216 PEOPEETIES OF METALS FOE CANNON, Table shoiving the compression, restoration, and permanent set per inch, in length, earned by the undermentioned weights, per square inch of section, acting upon a solid cylinder 10 inches long and 1.382 in. diameter, taken from near the exterior surfade of 42-pdr. gun, No. 336, cast hollow, at the Fort Pitt Foundry, of Bloomfield iron, and burst at the 4S)\st round, with 10 lis. powder and one solid shot. Weight per square inch of section. Compression per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .00006 .00006 2000 .00011 .00005 .00011 .00005 — _ 3000 .00016 .00005 .00016 .00005 — — 4000 .00022 .00006 .00022 .00008 _ _ 5000 .00027 .00005 .00026 .00004 .00001 — 6000 .00033 .00006 .00032 .00006 .00001 .00000 7000 .00038 .00005 .00036 .00004 .00002 .00001 8000 .00046 .00008 .00043 .00007 .00003 .00001 9000 .00053 .00007 .00049 .00006 .00004 .00001 10000 .00063 .00010 .00056 .00007 .00007 .00003 11000 .00069 .00006 .00059 .00003 .00010 .00003 12000 .00077 .00008 .00065 .00006 .00012 .00002 13000 .00086 .00009 .00072 .00007 .00014 .00002 14000 .00095 .00009 .00079 .00007 .00016 .00002 15000 .00103 .00008 .00082 .00003 .00021 .00005 16000 .00113 .00010 .00089 .00007 .00024 .00003 17000 .00122 .00009 .00094 .00005 .00028 .00004 18000 .00133 .00011 .00098 .00004 .00035 .00007 19000 .00143 .00010 .00106 .00008 .00037 .00002 20000 .00154 .00011 .00112 .00006 .00042 .00005 21000 .00167 .00013 .00118 .00006 .00049 .00007 22000 .00182 .00015 .00119 .00001 .00063 .00014 23000 .00200 .00018 .00125 .00006 .00075 .00012 24000 .00211 .00011 .00124 —.00001 .00087 .00012 25000 .00233 .00022 .00128 +.00004 .00105 .00018 26000 .00256 .00023 .00127 —.00001 .00129 .00024 27000 .00290 .00034 .00139 +.00012 .00151 .00022 28000 .00325 .00035 .00137 —.00002 .00188 .00037 29000 .00379 .00054 .00147 +.00010 .00232 .00044 30000 .00424 .00045 .00147 .00000 .00277 .00045 31000 .00486 .00062 .00147 .00000 .00339 .00062 32000 .00558 .00072 .00150 .00003 .00408 .00069 33000 .00646 .00088 Speci .00074 men began to —.00076 send. .00572 .00164 PLATE m. Curves expressive ofExtensjdibtjr, Compressibility md the < ■onx-spoudiurr Elasi icily, wirt Terairraeui set, o fTrm pT Middle rind Outer Specimens irorn dfjtdr.CmiiXvlofi Cn-t IwUoyi^altheFortPMFoimdrT^ofBloomReldirOTiy, and Tmrxt at the ^lUC^Mre, w/tA 10 lbs. powder 1 and on e solulshot. o/oo 003S 003O OOSS oeso C'C 7S 007O 0065 0060 OCX? COSO Of-i-S 004C 003S 003C CC2S 002C 0015 0010 coos fOOO 2bs PL ATE TV. . 0036 ■0090 00S5 VC8C C9?S COT 0065 0060 0O6S rOCSO OO'iS -r€/)tO .0035 ( •xtm sj In J U\ r permanent set Ji ehstirrtv ofiiiner specimens Jrmu JOmrJk ChlmnbjsddV 1 'X>'> ' U?j>(h:JS' c JoOlwth ol ''the Fort Tiit -Foundry /oc -0095 f'f-90 C( x5 oo&o 0075 ooro 0065 oceo ocss 0050 0Ct5 ocw -C03S €030 00 25 ccis 10 <0n li J>XA7'E 17 (iurcs C(>ojj),mn m cutset, ^elasticity of miiei' specimens tiromlOmS CoitmilJad A^ 963, cfflhstFoml F(uwihr,S<42j^cbiA r( Sjt),clFartl\ttFomidry. (Cjo CC.9C C'CfiO (COS 206O0 30CCO Soooo TLATU Til. ( '(it vex cmiij)ririxi r j con // irr'ssibil, i typerruanen 't set I- 1 "fas h in ; /v oihii) cj' specimens fromlOmck Colimibind A U 3'M> A' 4 \'ptfr.l\ '' 3s>6,both effort TM Foundry. (t.'>5 CO 10 cr 7S 007c or.os rceo Cr.SS 10'tS cc-4* 0035 60 2 c ;0020 ,tdS CO 10 .ore. AND QUALITIES OF CANNON POWDEE. 217 The foregoing results show remarkable extensibility ; but the increments in extension due to equal increments of strain increase so rapidly towards the breaking weight, as greatly to impair this otherwise highly desirable quality. They also show too great compressibility, and not sufficient elasticity, in this iron, especially in returning from its compressed condition. This renders the effects of firing accumulative, and diminishes the work due to elasticity. This iron would doubtless be improved by further decarbonization. Its ultimate extensibility would be diminished; but so also would its compressibility ; while its tenacity and elasticity would both be increased. I think it highly desirable that this iron should be further tested, both with powder, and mechanically. The curves expressive of the qualities of this iron (see Plates Nos. 3, 4, 5, 6, and 7,) show the inner specimen, which was most rapidly cooled, to be the best iron ; and those comparing the qualities of this iron with those of that in the 10-inch guns, made mainly of Greenwood iron, show it to be intermediate, in quality, between the iron in those guns. Determination of Exterior Model of Guns. The formula for the bursting tendency, Z = S(B-r)/2rL+3L B {B + It)(Ii-r)i:y beCOmeS > ^ SU PP 0sill S (Z) to remain constant, and R, L, and p, to vary, the equation of a portion of the curve of intersection of the exterior of one side of the gun by a plane containing the axis of the bore, since by this hypothesis the gun would be equally strong in all its parts. In this formula (p) will obviously be a function of (L), and if we suppose the maximum pressure to be exerted upon a length {V) of the bore, and that the pressure from the forward extremity of {V) to the muzzle is inversely as the volume occupied by the gas, then the pressure at any distance (L ) from the bottom of the bore, would be expressed by ^-, and the above formula would become, by changing (Z) to (C), and substituting 2prV B_ for {p) its value (^-), C= Sf- X (B _ r w 2ri + ffJ8(J8 + r) {B -r) V \ Now since (^ ) is constant, {B _ r)(2 r z + 3 f (i2 + r)(jB _ r) ^ 55 218 PKOPEETIES OP METALS EOE CANNON, will also be constant ; so that this last expression alone need be regarded in determining the value of (R) corresponding to the assumed values of (L). From the great excess of the transverse over the tangential resistance for the smaller values of (L) and from the rapid diminution of the transverse resistance as (L) increases, the value of this expression, with a constant value of (R) will at first increase to a maximum, and then decrease as (L) increases. In order, therefore, to determine the proper exterior model of a gun, we first determine upon the volume of the charge; and, from the quality of the powder, and form and weight of projectile, the length of bore {V) subjected to the maximum pressure, and the value of that pressure. Then establish the relation between {V) and {L) or the law of variation of pressure, and assume (I) equal to, or a little less, than two calibres, since experiment has shown the transverse resistance to be fully developed for about that length of surface pressed. Then assume (R) equal to, or a little less than the greatest exterior radius of the gun, and determine the value of (L) that renders — R (R — r) {2rL-t-3R(R + r) (R — r) Z 4 \ \ l 1 ) a maximum. Then if (R) have been assumed equal to the greatest exterior radius, the gun will be cylindrical from this point back to the curve of the breech ; and the curve of that portion forward of this point will be determined by assuming values for (X) and determining for (R) such corresponding values as will cause — — - — — — — , 4 , to remain constant, (R — r) / 2 r L + 3 R (R -f- r) (R — r) t \ \ L b f and equal to its maximum. The accidental variations in the pressure produced by the irregularity of combustion of equal charges, are greater in that part of the bore subjected to the maximum pressure than in any other. The quality of the metal is more liable to be injured by the greater pressure to which it is subjected, and by the greater consequent penetration of the gas into its pores, in this, than in any other part of the gun. This portion of the gun is also under pressure for a greater length of time, at each discharge, than any other part. It should, therefore, have an excess of strength over other parts of the gun. And, since the tangential resistance increases very slowly as the thickness of metal increases beyond one calibre, AND QUALITIES OF CANNON POWDEE. 219 while the transverse resistance increases as the square of the thickness, in this part of the gun, this object will be accomplished with the minimum weight of metal, and a more pleasing outline given to the gun, by assuming (R), in deter- mining that value of (X) which gives the maximum bursting tendency, from one-fifteenth to one-tenth less than the maximum exterior radius, and connect- ing the chase curve with that of the breech by another curve, such as to place the maximum diameter of the gun a little forward of the middle of the length of bore subjected to the maximum pressure, when firing the maximum charge. The expression for the tangential resistance which enters the formula for the bursting tendency, was derived from the hypothesis that, in this resist- ance, the strain developed by the action of a central force in the concentric, elementary cylinders, of which we may suppose the gun to consist, diminishes as. the square of the distance from the axis increases. And although experiment has not, thus far, established the rigorous truth of this law, yet the results obtained from bursting cylinders with powder (see page 192) indicate that it would be unsafe to assume the strain to diminish in a less ratio. And in guns made of very soft, and consequently highly compressible metal, I have no doubt that it may vary in even a higher ratio ; and in solid cast guns, where the softest, most porous, and consequently most com- pressible iron in the gun is immediately around the bore, and of necessity subjected to the maximum pressure, the injurious effects of compressibility will reach their maximum. Effects of Compressibility. The effects of compressibility in gun iron, in diminishing both the trans- verse and the tangential resistance, have been already referred to. No definite expression, however, for the value or amount of those effects has yet been determined. In order to determine this expression, let (p) = pressure per square inch of eras on surface of bore, c = compression per inch in length, due to (p), of a square prism one square inch in area of cross section, (R) = exterior radius, (r) = radius of bore, and x = variable between R and r. Then the elementary compression of this prism, due to (p) would — cdx; and if the pressure were uniform throughout the length of this prism, the integral of this expression would give its entire compression. 220 PEOPEETIES OF METALS EOE CANNON, But, in a gun, the pressure per square inch against the interior of each consecutive, elementary cylinder of which we may suppose it to consist, will vary according to some regular law, which must first be determined. For the purpose of determining this law, let us suppose a thin hollow cylinder, — a steam boiler for example, — and let (a) = tangential resistance per unit of length of one side of this boiler, (/) = interior radius, and (p) = pressure per square inch against its interior surface, which would just produce rupture. Then, from the well known formula for the strength of steam boilers, and for the bursting effort of any central force, we have p r =a and consequently p = — . Or the pressure per square inch against the interior of a hollow cylinder, necessary to develop a constant amount of tangential resistance in its sides, is inversely as its interior radius. The expression for the tangential resistance is 8 ~ „ ; hence (Mr — r 2 ) {Br — r 2 ) p r=8 * -g — h a.ndp = 8 R -^ r - ■ The tangential resistance, developed in that portion of the gun whose inte- rior radius is x, will be equal to the total tangential resistance minus that developed in that portion whose exterior radius = x; and consequently ( a (JRr — r*) „ lxr — r*). \ = \ B — / Ji eiice 'the pressure per square inch against the interior surface of the elementary cylinder whose interior radius = x will be expressed by \8 — ^— — — 8 — 2 — -)• And supposing the compression, per inch in length, of the same metal, to be directly propor- 8 - — ^ B x „(xr — r*)\ , Sc/(Br — r 2 ) ( xr — r 2 )\ elementary compression at any distance x from the axis, =d u = — I - — - p~ r 8 c /(Br — r 2 ) r 2 \ = 1 5 ~ Nap. log. x — r Nap. log. x [■ C) But u = P \ Mil XI n Sc I (Rr — r 2 ) >T . „ r »v •whenx = r; hence C== I — = Nap. log. r -f- r Nap. log. r -\ I 8c /{Br — Rr + r°) \ Sc/r' T , \ = J- ( B Na P- l0 S- r + r ) = -J- \R Na P" lo §- ' + ')• AND QUALITIES OF CANNON POWDEB. 221 Hence u = — I log. x — r log. x \- — log. r -f- r I = — \ » l°g- x h ^-l°g- r + r ) = — I -w log. \- r I . And integrating between the limits # = r, and x = R, we have, since u = S c Ir z r t * \ when « = r, w= — I — log. •=■ = + r 1; and if the gun be one calibre thick, R will = 3 r, and we shall have, by substituting this value of R, ^ = —^3 Nap. log. - - - +r )= — ^Nap. log. y + T ). And since the log. of 1 = 0, we have u = — I — — — Nap. log. 3 ). But the Nap. log. 3 = 1.0986. Sc r Assuming this log. = 1, we have u = — X . -»-• Now supposing p = /S, or that the pressure per square inch on the bore of the gun is equal to the y tensile strength of the metal, we have u = c — ; or the increase in diameter o of the bore, due to the compression of the metal, in a gun one calibre thick, is equal to one-third of the total compression which a prism, whose height equals the diameter of the bore, would undergo under a pressure per square inch equal to that against the bore of the gun. Now, if we suppose a given pressure to be exerted upon the surface of the bore of a gun, while its exterior diameter is prevented from undergoing any increment, the total enlargement of the bore, and the consequent extension of the metal, will be wholly due to compression, and all the effects of com- pression will be produced, as if the exterior of the gun were unconstrained. If we now suppose the exterior restraint removed, the interior and exterior diameters would undergo precisely equal increments. Or the gun would expand in the same manner as one of which the metal is perfectly incompressible, the metal having already undergone all the compression which this pressure could produce ; and the extension of the metal at the two surfaces of the gun, which would take place after the removal of the exterior restraint, would therefore be inversely as their radii. It has just been shown that in a gun one calibre thick the total enlarge- 2 r c ment of the bore, due to compression, = — r— ; the total extension at the surface of the bore, due to this enlargement, = 2 n r — ; and the extension 56 222 PEOPEETIES OF METALS FOE CANNON, 2 * r i c per inch at the same surface = —= = -77. Now if a = the total exten- r 2 ir r 3 sion per inch of which the metal is susceptible, then a — — = the extension per inch which the surface of the bore underwent after the removal of the exterior restraint, and the extension per inch of the exterior surface would c 3~ 3 a — e 3 9 To exemplify, let us take cylinder A, 0., (see page 227,) the total exten- sion per inch of which, was .00303, the compression per inch of A, T. was .00441, one-third of which = .00147, and .00303 — .00147 = .00156, one- third of which = .00052 = the extension per inch of the exterior of a gun, one calibre thick, made of this metal, at the moment of interior rupture. By reference to the column of extension of A, 0., it will be seen that 11000 lbs. per inch were required to produce an extension of .00054 ; or the exterior of the gun would be under a strain of between 10000 and 11000 lbs. per inch at the moment of interior rupture ; while, if the metal were perfectly incompressible, it would at the same moment be under a strain of 18000 lbs. The expression r — was derived from the hypothesis that the compression per inch of cast iron is directly proportional to the pressure, — experiment shows the compression of this metal to increase in a higher ratio ; so that the effects of compressibility will be something greater than those just deter- mined. These examples suffice to establish the importance attaching to the prop- erty of compressibility in gun metal, its action being to prevent the full development of both the transverse and the tangential resistance, and to that degree, it is believed, in guns of large calibre, and consequently of great pressure of gas, as to cause interior, longitudinal rupture before the trans- verse resistance is fully developed, even for the shortest practical lengths of surface pressed. For this reason only one-third of the theoretical transverse resistance was used in computing the exterior radii of the 15-inch gun ; the formula used for this computation being, c _ 2prVT ^ B^T (R-r) (2rL + B{R + r)(R-r)J^-)' & riATE ^ T m. JiiaKsi — X-»/, *-' e „ A', --1 AND QUALITIES OP CANNON POWDEE. 223 The value of R, used in determining the value of (Z), which rendered the bursting tendency a maximum, was = 22.5 inches. The outer and extreme inner dotted lines in the figure on Plate 8, give the exterior form and proportions, and diameter of bore, of the gun as cast. The inner curved dotted lines, on the same figure, give the form and propor- tions of a gun of the same bore and maximum exterior diameter, computed on the hypothesis that the pressure of the gas is inversely as the space behind the shot. The middle dotted lines, same figure, give the form and propor- tions of a gun of the same diameter of bore and maximum exterior diameter, on the hypothesis that the pressure is inversely as the square root of the space behind the shot, or as V L. The full lines, same figure, show the form and proportions of this gun, as finished, on a scale of one-twentieth size. It will be observed that this gun is something heavier in the chase, than the hypothesis, that the pressure is inversely as i/L, would give it. It was pur- posely made so ; for the reason that it was intended to use charges of such character as would produce a much more uniform pressure, and consequently, greater pressure, in the chase of the gun, for a given maximum pressure, than is obtained by the use of ordinary powder. It should be here remarked that, even for guns in which a quick powder is to be used, the lines due to the law that the pressure is inversely as (L), should not be strictly adhered to in that part where the most rapid diminution of exterior diameter occurs ; for the reason that, in so doing, the front ends of the staves, for those lengths of bore subjected to the greatest pressure, would be deprived of their proper support, and the transverse resistance would be greatly diminished just where it is most needed, and where its value is greatest in the properly modelled gun. The beginning of the taper should therefore be, say half a calibre further forward, and the taper made less rapid than the law of pressure, in this part of the gun, would give it. Experiment has not yet satisfactorily established the law of variation in pressure due to the ordinary cannon powder. But it is considered that no powder is fit for use in guns of large calibre, that will not so far approximate to uniformity of pressure as to conform to the law that the pressure is inversely as ,) and composed as follows, viz. : Green- wood and Salisbury pigs were melted together in the following proportions, viz. : — No. 1 Greenwood, a 2 « " 3 " Salisbury pigs, Total, 1093 lbs. 1914 1093 900 5000 lbs. Being 82 per cent, of Greenwood and 18 per cent, of Salisbury iron, which was cast into 2d fusion pigs. From these 2d fusion pigs cylinder (D,) was cast. From each of these cylinders a slab 4.5 in. thick was cut, by planes parallel to and equi-distant from the plane containing the axis of the cylinder and the conjugate axis of its base. Specimens were taken from like parts of each of these slabs, and subjected to similar tests. The accompanying drawing (Plate No. 9,) shows the position whence the specimens for the various tests were taken, arid the following tables exhibit the results obtained : — TZATE2Z. AND QUALITIES OF CANNON POWDEE. 227 EXTENSION OP A, 0. Table showing the extension, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 30 in. long and 1.382 in. diameter, cut from exterior of trial cylinder (A), made of 2d fusion Bloomfield iron. Weight Extension Restoration Permanent set per square inch per inch First difference. per inch First difference. per inch. First difference. of section. in length. in length. in length. 1000 lbs. .00002 .00002 _ _ 2000 .00004 .00002 .00004 .00002 - - 3000 .00013 .00009 .00013 .00009 - - 4000 .00019 .00006 .00019 .00006 - - 5000 .00023 .00004 .00023 .00004 - - 6000 .00027 .00004 .00027 .00004 - - 7000 .00033 .00006 .00033 .00006 - - 8000 .00037 .00004 , .00037 .00004 - - 9000 .00043 .00006 .00042 .00005 .00001 - 10000 .00049 .00006 .00047 .00005 .00002 .00001 11000 .00054 .00005 .00052 .00005 .00002 .00000 12000 .00060 .00006 .00057 .00005 .00003 .00001 13000 .00066 .00006 .00062 .00005 .00004 .00001 14000 .00073 .00007 .00067 .00005 .00006 .00002 15000 .00080 .00007 .00073 .00006 .00007 .00001 16000 .00087 .00007 .00078 .00005 .00009 .00002 17000 .00095 .00008 .00084 .00006 .00011 .00002 18000 .00103 .00008 .00089 .00005 .00014 .00003 19000 .00112 .00009 .00096 .00007 .00016 .00002 20000 .00122 .00010 .00103 .00007 .00019 .00003 21000 .00133 .00011 .00110 .00007 .00023 .00004 22000 .00144 .00011 .00116 .00006 .00028 .00005 23000 .00159 .00015 .00123 .00007 .00036 .00008 24000 .00173 .00014 .00130 .00007 .00043 .00007 25000 .00191 .00018 .00138 .00008 .00053 .00010 26000 .00210 .00019 .00145 .00007 .00065 .00012 27000 .00237 .00027 .00153 .00008 .00084 .00019 28000 .00267 .00030 .00163 .00010 .00104 .00020 29000 .00303 .00036 .00179 .00009 .00131 .00027 30000 Broke betwee n 29000 and 3 0000 lbs. 228 PEOPEETIES OF METALS FOE CANNON, EXTENSION OE A, I. Table showing the extension, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square . inch of section, acting on a solid cylinder 30 in. long and 1.382 in. diameter, cut from near the axis of trial cylinder (A), made of 2d fusion Bbomfield iron. Weight per square inch of section. Extension per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .00003 _ .00003 2000 .00007 .00004 .00007 .00004 — — 3000 .00011 .00004 .00011 .00004 — — 4000 .00016 .00005 .00016 .00005 — — 5000 .00021 .00005 .00021 .00005 — — 6000 .00026 .00005 .00026 .00005 — — 7000 .00029 .00003 .00029 .00003 — _ 8000 .00035 .00006 .00034 .00005 .00001 _ 9000 .00039 .00004 .00038 .00004 .00001 .00000 10000 .00050 .00011 .00047 .00009 .00003 .00002 11000 .00057 .00007 .00053 .00006 .00004 .00001 12000 .00064 .00007 .000-->9 .00006 .00005 .00001 13000 .00071 .00007 .00065 .00006 .00006 .00001 14000 .00079 .00008 .00072 .00007 .00007 .00001 15000 .00087 .00008 .00079 .00007 .00008 .00001 16000 .00096 .00009 .00085 .00006 .00011 .00003 17000 .00105 .00009 .00091 .00006 .00014 .00003 18000 .00115 .00010 .00097 .00006 .00018 .00004 19000 .00129 .00014 .00108 .00011 .00021 .00003 20000 .00142 .00013 .00115 .00007 .00027 .00006 21000 .00159 .00017 .00124 .00009 .00035 .00008 22000 .00175 .00016 .00131 .00007 .00044 .00009 23000 .00193 .00018 .00138 .00007 .00055 .00011 24000 .00217 .00024 .00147 .00009 .00070 .000 L 5 25000 .00245 .00028 .00156 .00009 .00089 .00019 26000 .00277 .00033 .00163 .00007 .00114 .00025 27000 .00334 .00057 ~~ ~~ — — AND QUALITIES OF CANNON POWDEE. 229 REPETITIONS OF A, 0. Table showing the extension, restoration and permanent set, per inch in length, caused by the repeated application of 22000 lbs. (or three-quarters of breaking weight) per square inch, on a solid cylinder 30 in. long and 1.382 in. diameter, cut from exterior of trial cylinder (A), made of 2d fusion Bhomfield iron. Number Extension. Restoration Permanent set of per inch First difference. per inch First difference. per inch First difference. repetitions. in length. in length. in length. i .00150 .00118 .00032 10 .00153 .00003 .00116 — 00002 .00037 .00005 60 .00183 .00030 .00122 +.00006 .00061 .00024 100 .00183 .00000 .00119 —.00003 .00064 .00003 150 .00183 .00000 .00118 —.00001 .00065 .OOuOl 200 .00185 .00002 .00117 —.00001 .00068 .00003 300 .00187 .00002 .00119 +.00002 .0U068 .00000 400 .00188 .00001 .00119 .00000 .00069 .00001 500 .00189 .00001 .00120 +.00001 .000 J9 .00000 600 .00189 .00000 .00119 —.00001 .00070 .00001 700 .00189 .00000 .00119 .00000 .00070 .00000 800 .00190 .00001 .00119 .00000 .00071 .00001 After 14 hra. rest. 801 .00189 — .00115 . - .00069 - 900 .00191 .00002 .00120 .00004 .00071 .00002 1000 .00191 .00000 .00120 .00000 .00071 .00000 1200 .00192 .00001 .00119 —.00001 .00073 .00002 1400 .00194 .00002 .00120 +.00001 .00074 .00001 After 36 brs. rest. 1401 .00193 — .00122 - .00071 — 1600. .00195 .00002 .00121 —.00001 .00074 .00003 1735 .00196 .00001 .00122 +.00001 .00074 .00000 1800 .00197 .00001 .00123 .00001 .00074 .00000 2000 .00197 .00000 .00123 .00000 .00074 .00000 2100 .00198 .00001 .00123 .00000 .00075 .00001 2300 .00198 .00000 .00123 .00000 .00075 .00000 Broke at 2301s t application — 1 ead pulled off. 00120 230 PEOPEKTIES OF METALS EOK CANNON, REPETITIONS OF A, 0. — (Duplicate. ) Table showing the extension, restoration and permanent set, per inch in length, caused by the repeated application of 26000 lbs. (or .9 of breaking weight) per square inch, on a solid cylinder 30 in. long and 1.382 in. diameter, ad from exterior of trial cylinder (A), made of 2d fusion Bloomfield iron. Number Extension Eestoration Permanent set of per inch First difference. per inch First difference. per inch. First difference. repetitions. in length. in length. in length. 1 .00221 .00157 .00064 10 .00223 .00007 .00155 —.00002 .00073 .00009 30 .00242 .00014 .00156 +.00001 .00086 .00013 50 .00249 .00007 .00156 .00000 .00093 .00007 70 .00253 .00004 .00156 .00000 .00097 .00004 100 .00203 .00010 .00157 .00001 .00106 .00009 150 .00274 .00011 .00159 .00002 .00115 .00009 200 .00277 .00003 .00156 —.00003 .00121 .00006 250 .00292 .00015 .00164 +.00008 .00128 .00007 281 Broke at the 2 82d repetition. .00157 AND QUALITIES OP CANNON POWDEE. 231 EXTENSION OF B, 0. Table showing the extension, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 30 in. long and 1.382 in. diameter, cut from exterior of trial cylinder (3), made of Greenwood and Salisbury iron. ■Weight per square inch of section. Extension per inch in length. First difference. Hestoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 Its. .00003 _ .00003 m m _ _ 2000 .00008 .00005 .00008 .00005 — - sooo .00013 .00005 .00013 .00005 - — 4000 .00018 .00005 .00018 .00005 - — 5000 .00021 .00006 .00024 .00006 - - 6000 .00030 .00006 .00030 .00006 — - 7000 .00035 .00005 .00034 .00004 .00001 - 8000 .00042 .00007 .00041 .00007 .00001 .00000 9000 .00048 .00006 .00046 .00005 .00002 .00001 10000 .00056 .00008 .00052 .00006 .00004 .00002 11000 .00065 .00009 .00060 .00008 .00005 .00001 12000 .00074 .00009 .00067 .00007 .00007 .00002 13000 .00083 .00009 .00074 .00007 .00009 .00002 14000 .00093 .00010 .00080 .00006 .00013 .00004 15000 .00108 .00015 .00090 .00010 .00018 .00005 16000 .00122 .00014 .00098 00008 .00024 .00006 17000 .00140 .00018 .00106 .00008 .00034 .00010 18000 .00160 .00020 .00115 .00009 .00045 .00011 19000 .00191 .00031 .00131 .00016 .00060 .00015 20000 .00235 .00044 .00140 .00009 .00095 .00035 21000 .00291 .00056 .00154 .00014 .00137 .00042 22000 Broke betwe en 21000 and 22000 lbs. 232 PKOPEETIES OF METALS EOR CANNON, EXTENSION OF B, I Table showing the extension, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 30 in. long and 1.382 in. diameter, cut from near the axis of trial cylinder (B), made of Greenwood and Salisbury iron. "Weight Extension First difference. Restoration First difference. Permanent set per square inch of section. per inch in length. per inch in length. per inch in length. First difference. 1000 lbs. .00005 _ .00005 _ 2000 .00010 .00005 .00010. .00005 - — 3000 .00015 .00005 .00015 .00005 — — 4000 .00020 .00005 .00020 .00005 - — 5000 .00024 .00004 .00023 .00003 .00001 — 6000 .00031 .00007 .00029 .00003 .00002 .00001 7000 .00041 .00010 .00039 .00010 .00002 .00000 8000 .00052 .00011 .00049 .00010 .00003 .00001 9000 .00059 .00007 .00055 .00006 .00004 .00001 10000 .00070 .00011 .00065 .00010 .00005 .00001 11000 .00031 .00011 .00072 .00007 .00009 .00004 12000 .00091 .00010 .00078 .00006 .00013 .00004 13000 .00105 .00014 .00086 .00008 .00019 .00006 14000 .00119 .00014 .00094 .00008 .00025 .00006 15000 .00139 .00020 .00104 .00010 .00035 .00010 16000 .00161 .00022 .00112 .00008 .00049 .00014 17000 Broke betwee n 16000 and 17000 lbs. AND QUALITIES OF CANNON POWDER. 233 REPETITIONS OF B, 0. Table showing the extension, restoration and permanent set, per inch in length, earned by the repeated application of 20000 lbs. (or .9 of breaking weight) per square inch, on a solid cylinder 30 in. long and 1.382 in. diameter, cict from the exterior of trial cylinder (B), made of Greenwood and Salisbury iron. Number Extension Eestoration Permanent set of per inch. First difference. per inch. First difference. per inch First difference. repetitions. in length. in length. in length. 1 .00278 .00135 .00143 10 .00295 .00017 .00134 —.00001 .00159 .00016 30 .00320 .00025 .00131 —.00003 .00189 .00030 After 12 hrs. rest. 31 .00324 — .00135 - .00189 — 40 .00333 .00009 .00137 +.00002 .00196 .00007 50 .00338 .00005 .00133 —.00004 .00205 .00009 70 .00353 .00015 .00141 +.00008 .00212 .00007 100 .00362 .00009 .00140 —.00001 .00222 .00010 150 .00373 .00011 .00140 .00000 .00233 .00011 200 .00382 .00009 .00139 —.00001 .00243 .00010 251 .00392 .00010 .00137 —.00002 .00255 .00012 .00137 REPETITIONS OP B, 0. — (Duplicate. ) With 20000 lbs. per inch. Extension Restoration Permanent set of per inch First difference. per inch First difference. per inch First difference. repetitions. in length. in length. in length. 1 .00218 .001 28 _ .00090 10 .00246 .00028 .00134 .00006 .00112 .00022 30 .00270 .00024 .00136 .00002 .00134 .00022 50 .00292 .00022 .00135 —.00001 .00157 .00023 70 .00309 .00017 .00143 +.00008 .00166 .00009 100 .00323 .00014 .00149 .00006 .00174 .00008 150 Broke at the 11 0th repetition. .00138 59 234 PKOPEETIES OF METALS FOE CANNON, EXTENSION OF O, 0. Table showing the extension, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 30 in. long and 1.382 in. diameter, cut from exterior of trial cylinder (0), made from equal parts of 2d fusion Greenwood and Bloomfield iron. Weight Extension First difference. Restoration ' First difference. Permanent set First difference. per square inch of section. per inch in length. per inch in length. per inch in length. 1000 lbs. .00004 _ .00004 2000 .00008 .00004 .00008 .00004 — — 3000 .00012 .00004 .00012 .00004 — — 4000 .00018 .00006 .00018 .00006 — — 5000 .00022 .00004 .00021 .00003 .00001 — 6000 .00027 .00005 .00026 .00005 .00001 .00000 7000 .00032 .00005 .00030 .00004 .00002 .00001 8000 .00041 .00009 .00039 .00009 .00002 .00000 9000 .00046 .00005 .00043 .00004 .00003 .00001 10000 .00052 .00006 .00048 .00005 .00004 .00001 11000 .00059 .00007 .00054 .00006 .00005 .00001 12000 .00066 .00007 .00060 .00006 .00006 .00001 13000 .00075 .00009 .00067 .00007 .00008 .00002 14000 .00083 .00008 .00074 .00007 .00009 .00001 15000 .00093 .00010 .00081 .00007 .00012 .00003 16000 .00102 .00009 .00086 .00005 .00016 .00004 17000 .00114 .00012 .00094 .00008 .00020 .00004 18000 .00127 .00013 .00102 .00012 .00021 .00001 19000 .00141 .00014 .00110 .00004 .00031 .00010 20000 .00156 .00015 .00118 .00008 .00038 .00007 21000 .00173 .00017 .00125 .00007 .00048 .00010 22000 .00196 .00023 .00135 .00010 .00061 .00013 23000 .00221 .00025 .00144 .00009 .00077 .00016 24000 .00252 .00031 .00153 .00009 .00099 .00022 25000 .00287 .00035 .00160 .00007 .00127 .00023 26000 Broke betwet n 25000 and 26000 lbs. AND QUALITIES OF CANNON POWDEK. 235 EXTENSION OF C, I. Table showing the extension, restoration and permanent set, per inch in length, caused hj the undermentioned weights, per square inch of section, acting on a solid cylinder 30 in. long and 1.382 in. diameter, cut from near the axis of trial cylinder (O), made from equal parts of 2d fusion Greenwood and Bloomfield iron. Weight Extension First difference. Restoration Permanent set per square inch of section. per inch in length. per inch in length. First difference. per inch in length. First difference. 1000 lbs. .00004 .00004 2000 .00008 .00004 .00008 .00004 — — 3000 .00012 .00004 .00012 .00004 — _ 4000 .00018 .00006 .00018 .00006 — _ 5000 .00022 .00004 .00022 .00004 — — 6000 .00027 .00005 .00026 .00004 .00001 _ 7000 .00033 .00006 .00032 .00006 .00001 .00000 8000 .00038 .00005 .00036 .00004 .00002 .00001 9000 .00045 .00007 .00043 .00007 .00002 .00000 10000 .00051 .00006 .00048 .00005 .00003 .00001 11000 .00058 .00007 .00054 .00006 .00004 .00001 12000 .00065 .00007 .00060 .00006 .00005 .00001 13000 .00073 .00008 .00066 .00006 .00007 .00002 14000 .00081 .00003 .00072 .00006 .00009 .00002 15000 .00091 .00010 .00080 .00008 .00011 .00002 16000 .00099 .00008 .00085 .00005 .00014 .00003 17000 .00109 .00010 .00092 .00007 .00017 .00003 18000 .00120 .00011 .00098 .00006 .00022 .00005 19000 .00133 .00013 .00106 .00008 .00027 .00005 20000 .00147 .00014 .00114 .00008 .00033 .00006 21000 .00163 .00016 .00122 .00008 .00041 .00008 22000 .00183 .00020 .00131 .00009 .00052 .00011 23000 .00207 .00024 .00137 .00006 .00070 .00018 24000 .00232 .00025 .00146 .00019 .00086 .00016 25000 .00269 .00037 .00158 .00012 .00111 .00025 26000 .00336 .00067 .00188 .00030 .00148 .00037 27000 .00382 J .00046 .00180 —.00008 .00202 .00054 28000 Broke betw( sen 27000 and 28000 lbs. 236 PEOPEETIES OE METALS FOE CANNON, REPETITIONS OF C, 0. Table showing the extension, restoration and permanent set, per inch in length, caused by the repeated applications of 22500 lbs. (or .9 of breaking weight) per square inch, on a solid cylinder 30 in. long and 1.382 in. diameter, cut from the exterior of trial cylinder (0), made from equal parts of 2d fusion Greenwood and Bbomfield iron. Number of repetitions. Extension per inch in length. _ First difference. Kest oration per inch in length. First difference. Permanent set per inch in length. First difference. i .00208 .00138 .00070 10 .00223 .00015 .00137 —.00001 .00086 .00016 30 .00233 .00010 .00135 —.00002 .00098 .00012 50 .00243 .00010 .00139 +.00004 .00104 .00006 70 .00250 .00007 .00139 .00000 .00111 .00007 After 15 hrs. rest. 71 .00247 — .00139 * — .00108 — 100 .00251 .00004 .00140 .00001 .00111 .00003 150 .00260 .00009 .00140 .00000 .00120 .00009 250 .00271 .00011 .00142 .00002 .00129 .00009 300 .00278 .00007 .00143 .00001 .00135 .00006 350 .00285 .00007 .00143 .00000 .00142 .00007 450 .00311 .00026 .00149 .00006 .00162 .00020 After 15 hrs. rest. 451 .00309 — .00148 - .00161 _ 550 .00323 .00014 .00151 .00003 .00172 .00011 650 .00332 .00009 .00149 —.00002 .00183 .00011 Broke at 72] st repetition. .00142 AND QUALITIES OF CANNON POWDEE. 237 REPETITIONS OF 0, 0. — (Duplicate. ) Table showing the extension, restoration and permanent set, per inch in length, earned by the repeated application of 23500 lbs. (or .9 4 of breaking weight) per square inch, on a solid cylinder 30 in. long and 1.382 in. diameter, cut from the exterior of trial cylinder (O), made from equal parts of 2d fusion Greenwood and Bhomfleld iron. Number of repetitions- Extension per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch. in length* First difference. i .00128 .00133 .00085 ii .00245 .00027 .00139 .00006 .00106 .00021 30 .00262 .00017 .00140 .00001 .00122 .00016 50 .00272 .00010 .00140 .00000 .00132 .00010 70 .00281 .00009 .00143 .00003 .00138 .00006 100 .00300 .00019 .00150 .00007 .00150 .00012 150 .00327 .00027 .00157 .00007 .00170 .00020 200 .00334 .00007 .00149 —.00008 .00185 .00015 After 20 hrs. rest 201 .00334 — .00150 — .00184 — 250 .00337 .00003 .00148 —.00002 .00189 .00005 300 .00348 .00011 .00150 +.00002 .00198 .00009 350 .00356 .00008 .00149 —.00001 .00207 .00009 After 15 hrs. reBt. 351 .00355 — .00151 — .00204 - 400 .00357 .00002 .00148 —.00003 .00209 .00005 450 .00374 .00017 .00154 +.00006 .00220 .00011 457 Broke at 45"3 th repetition. .00146 238 PEOPEETIES OF METALS FOE CANNON, EXTENSION OP D, 0. Table showing the extension, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 30 in. long and 1.382 in. diameter, cut from exterior of trial cylinder (D), made from Greenwood and Salisbury iron, re-melted. ■Weight Extension Restoration Permanent set per square inch per inch First difference. per inch First difference. per inch First difference. of section. in length. in length. in length. 1000 lbs. .00003 .00003 _ _ 2000 .00009 .00006 .00009 .00006 - - 3000 .00016 .00007 .00016 .00007 - - 4000 .00021 .00005 .00021 .00005 - - 5000 .00026 .00005 .00025 .00004 .00001 - 6000 .00033 .00007 .00032 .00007 .00001 .00000 7000 .00040 .00007 .00038 .00006 .00002 .00001 8000 .00047 .00007 .00045 .00007 .00002 .00000 9000 .00054 .00007 .00051 .00006 .00003 .00001 10000 .00061 .00007 .00056 .00005 .00005 .00002 11000 .00069 .00008 .00063 .00007 .00006 .00001 12000 .00078 .00009 .00070 .00007 .00008 .00002 13000 .00085 .00007 .00074 .00004 .00011 .00003 14000 .00094 .00009 .00079 .00005 .00015 .00004 15000 .00106 .00012 .00086 .00007 .00020 .00005 16000 .00119 .00013 .00094 .00008 .00025 .00005 17000 .00135 .00016 .00101 .00007 .00034 .00009 18000 .00150 .00015 .00107 .00006 .00043 .00009 19000 .00171 .00021 .00115 .00008 .00056 .00013 20000 .00194 .00023 .00121 .00006 .00073 .00017 21000 .00231 .00037 .00130 .00009 .00101 .00028 22000 .00278 .00047 .00142 .00012 .Q0136 .00035 23000 .00327 .00049 .00146 .00004 .00181 .00045 24000 .00424 .00097 .00158 .00012 .00266 .00085 25000 Broke bet-wee q 24000 and 2 5000 lbs. AND QUALITIES OF CANNON POWDEK. 239 EXTENSION OP D, I. Table showing the extension, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 30 in. long and 1.382 in. diameter, cut from near t/ie axis of trial cylinder (D), made from Greenwood and Salisbury iron, re-melted. Weight per square inch of section. Extension per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .00002 _ .00002 _ 2000 .00008 .00006 .00008 .00006 — - 3000 .00012 .00004 .00012 .00004 — — 4000 .00020 .00008 .00020 .00008 — - 5000 .00029 .00009 .00029 .00009 — - 6000, .00037 .00008 .00036 .00007 .00001 — 7000 .00047 .00010 .00045 .00009 .00002 .00001 8000 .00057 .00010 .00053 .00008 .00004 .00002 9000 .00069 .00012 .00062 .00009 .00007 .00003 10000 .00081 .00012 .00070 .00008 .00011 .00004 11000 .00096 .00015 .00080 .00010 .00016 .00005 12000 .00114 .00018 .00091 .00011 .00023 .00007 13000 .00132 .£0018 .00100 .00009 .00032 .00009 14000 .00156 .00024 .00115 .00015 .00045 .00013 15000 .00184 .00028 .00122 .00007 .00062 .00017 16000 .00223 .00039 .00136 .00014 .00087 .00025 17000 Broke between 16000 and 17000 lbs. 240 PBOPERTIES OF METALS EOE GANNON, REPETITIONS OP D, Table showing the extension, restoration and permanent set, per inch in length, caused by the repeated application of 21600 lbs. {or .9 of breaking weight) per square inch, on a solid cylinder 30 in. long and 1.382 in. diameter, cut from the exterior of trial cylinder (D), made from Greenwood and Salisbury iron, re-melted. Number Extension Restoration Permanent set of per inch First difference. per inch First difference. per inch First difference. repetitions. in length.. in length. in length. 1 .00271 .00143 .00128 10 .00307 .00036 .00148 .00005 .00159 .00031 . 30 .00337 .00030 .00148 .00000 .00189 .00030 50 .00354 .00017 .00149 .00001 .00205 .00016 70 .00371 .00017 .00150 .00001 .00221 .00016 100 .00402 .00031 .00149 —.00001 .00253 .00032 150 .00427 .00025 .00148 —.00001 .00279 .00026 171 Broke at the 1 72d repetition. .00148 AND QUALITIES OF CANNON POWDEE. 241 COMPRESSION OF A, 0. Table showing the compression, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 10 in. long and 1.382 in. diameter, cut from exterior of trial cylinder (A), made of 2d fusion Bhomfleld iron. Weight per square inch of section. Compression per inch in length. First difference. Restoration jper inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .00003 .00003 2000 .00008 .00005 .00008 .00005 — _ 3000 .00013 .00005 .00013 .00005 — _ 4000 .00019 .00006 .00019 .00006 — _ 5000 .00023 .00004 .00023 .00004 — — 6000 .00029 .00006 .00029 .00006 — — 7000 .00034 .00005 .00034 .00005 — — 8000 .00038 .00004 .00038 .00004 - — 9000 .00044 .00006 .00044 .00006 — — 10000 .00049 .00005 .00049 .00005 — — 11000 • .00054 .00005 .00054 .00005 — — 12000 .00059 .00005 .00058 .00004 .00001 — 13000 .00063 .00004 .00061 .00003 .00002 .00001 14000 .00068 .00005. .00066 .00005 .00002 .00000 15000 .00073 .00005 .00070 .00004 .00003 .00001 16000 .00080 .00007 .00077 .00007 .00003 .00000 17000 .00085 .00005 .00081 .00004 .00004 .00001 18000 .00091 .00006 .00087 .00006 .00004 .00000 19000 .00096 .00005 .00091 .00004 .00005 .00001 20000 .00102 .00006 .00097 .00006 .00005 .00000 21000 .00109 .00007 .00103 .00006 .00006 .00001 22000 .00116 .00007 .00107 .00004 .00009 .00003 23000 .00120 .00004 .00111 .00004 .00009 .00000 24000 .00126 .00006 .00114 .00003 .00012 .00003 25000 .00136 .00010 .00122 .00008 .00014 .00002 26000 .00142 .OJ3006 .00125 .00003 .00017 .00003 27000 .00149 .00007 .00127 .00002 .00022 .00005 28000 .00161 .00012 .00135 .00008 .00026 .00004 29000 .00169 .00008 .00139 .00004 .00030 .00004 30000 .00181 .00012 .00145 .00006 .00036 .00006 31000 .00191 .00010 .00150 .00005 .00041 .00005 32000 .00202 .00011 .00154 .00004 .00048 .00007 33000 .00214 .00012 .00157 .00003 .00057 .00009 34000 .00227 .00013 .00161 .00004 .00066 .00009 35000 .00249 .00022 .00167 .00006 .00082 .00016 36000 .00267 .00018 .00171 .00004 .00096 .00014 37000 .00288 .00021 .00172 .00001 .00116 .00020 38000 .00314 .00026 .00178 .00006 .00136 .00020 39000 .00346 .00032 .00181 .00003 .00165 .00029 40000 .00393 .00047 .00192 .00011 .00201 .00036 41000 .00457 .00064 .00146 —.00056 .00311 .00110 242 PEOPEETIES OF METALS EOE CANNON, COMPRESSION OF A, T. Table showing the compression, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 10 in. long and 1.382 in. diameter, cut transversely 13 inches from lower end of trial cylinder (A), made of 2d fusion Bloomfleld iron. Weight Compression First difference. Restoration Permanent set per square inch of section. per inch in length. per inch in length. First difference. per inch in length. First difference. 1000 lbs. .00012 .00012 _ 2000 .00022 .00010 .00022 .00010 — _ 3000 .00032 .00010 .00032 .00010 — — 4000 .00042 .00010 .00042 .00010 — — 5000 .00051 .00009 .00051 .00009 — — 6000 .00055 .00004 .00055 00004 — — 7000 .00060 .00005 .00060 .00005 — _. 8000 .00068 .00008 .00068 .00008 — — 9000 .00074 .00006 .00073 .00005 .00001 _ 10000 .00080 .00006 .00078 .00005 .00002 .00001 11000 .00086 .00006 .00083 .00005 .00003 .00001 12000 .00093 .00007 .00089 .00006 .00004 .00001 13000 .00102 .00009 .00096 .00007 .00006 .00002 14000 .00108 .00006 .00101 .00005 .00007 .00001 15000 .00114 .00006 .00106 .00005 .00008 .00001 16000 .00121 .00007 .00112 .00006 .00009 .00001 17000 .00127 .00006 .00114 .00002 .00013 .00004 18000 .00135 .00008 .00120 .00006 .00015 .00002 19000 .00142 .00007 .00125 .00005 .00017 .00002 20000 .00151 .00009 .00131 .00006 .00020 .00003 21000 .00159 .00008 .00137 .00006 .00022 .00002 22000 .00163 .00004 .00138 .00001 .00025 .00003 23000 .00172 .00009 .00143 .00005 .00029 .00004 24000 .00181 .00009 .00148 .00005 .00033 .00004 25000 .00191 .00010 .00155 .00007 .00036 .00003 26000 .00200 .00009 .00161 .00006 .00039 .00003 27000 .00208 .00008 .00165 .00004 .00043 .00004 28000 .00217 .00009 .00169 .00004 .00048 .00005 29000 .00227 .00010 .00173 .00004 .00054 .00006 30000 .00237 .00010 .00177 .00004 .00060 .00006 31000 .00249 .00012 .00181 .00004 .00068 .00008 32000 .00263 .00014 .00186 .00005 .00077 .00009 33000 .00276 .00013 .00191 .00005 .00085 .00008 34000 .00294 .00018 .00197 .00006 .00097 .00012 35000 .00309 .00015 .00199 .00002 .00110 .00013 36000 .00330 .00021 .00204 .00005 .00126 .00016 37000 .00351 .00021 .00206 .00002 .00145 .00019 38000 .00378 .00027 .00209 .00003 .00169 .00024 39000 .00408 .00030 .00215 .00006 .00193 .00024 40000 .00441 .00033 .00224 .00009 1 .00217 .00024 AND QUALITIES OF CANNON POWDER. 243 COMPRESSION OF B, 0. Table showing the compression, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 10 in. long and 1.382 in. diameter, cut from exterior of trial cylinder (B), made of Greenwood and Salisbury iron. Weight Compression First difference. Kestoration First difference. Permanent set First difference. per square inch of section. per inch in length. per inch in length. per inch in length. 1000 lbs. .00016 .00016 2000 .00025 .00009 .00025 .00009 - - 3000 .00032 .00007 .00032 .00007 - - 4000 .00038 .00006 .00038 .00006 - - 5000 .00045 .00007 .00044 .00006 .00001 - 6000 .00050 .00005 .00048 .00004 .00002 .00001 7000 .00056 .00006 .00053 .00005 .00003 .00001 8000 .00061 .00005 .00057 .00004 .00004 .00001 9000 .00066 .00005 .00061 .00004 .00005 .00001 10000 .00072 .00006 .00066 .00005 .00006 .00001 11000 .00078 .00006 .00072 .00006 .00006 .00000 12000 .00083 .00005 .00076 .00004 .00007 .00001 13000 .00089 .00006 .00081 .00005 .00008 .00001 14000 .00094 .00005 .00085 .00004 .00009 .00001 15000 .00101 .00007 .00090 .0000.5 .00011 .00002 16000 .00108 .00007 .00096 .00006 .00012 .00001 17000 .00114 .00006 .00100 .00004 .00014 .00002 18000 .00119 .00005 .00104 .00004 .00015 .00001 19000 .00127 .00008 .00110 .00006 .00017 .00002 20000 .00136 .00009 .00117 .00007 .00019 .00002 21000 .00145 .00009 .00122 .00005 .00023 .00004 22000 .00155 .00010 ' .00127 .00005 .00028 .00005 23000 .00166 .00011 .00134 .00007 .00032 .00004 24000 .00178 .00012 .00141 .00007 .00037 .00005 25000 .00192 .00014 .00147 .00006 .00045 .00008 26000 .00211 .00019 .00155 .00008 .00056 .00011 27000 .00227 .00016 .00159 .00004 .00068 .00012 28000 .00256 .00029 .00168 .00009 .00088 .00020 29000 .00275 .00019 .00172 .00004 .00103 .00015 30000 .00297 .00022 .00173 .00001 .00124 .00021 31000 .00334 .00037 .00180 .00007 .00154 .00030 32000 .00390 .00056 .00183 .00003 .00207 .00053 33000 .00473 .00083 .00186 .00003 .00287 .00080 34000 .00542 .00069 .00195 .00009 .00347 .00060 35000 .00621 .00079 .00201 .00006 .00420 .00073 244 PEOPEETIES OF METALS FOE CANNON, COMPRESSION OF B, T. Table showing the compression, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 10 in. long and 1.382 in. diameter, cut transversely 13 in. from the lower end of trial cylinder (B), made of Greenwood and Salisbury iron. Weight per square inch of section. Compression per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .00010 .00010 2000 .00021 .00011 .00021 .00011 — _ 3000 .00029 .00008 .00029 .00008 — — 4000 .00034 .00005 .00034 .00005 _ _ 5000 .00041 . .00007 .00041 .00007 — ._ 6000 .00047 .00006 .00046 .00005 .00001 _ 7000 .00052 .00005 .00050 .00004 .00002 .00001 8000 .00059 .00007 .00056 .00006 .00003 .00001 9000 .00066 .00007 .00062 .00006 .00004 .00001 10000 .00073 .00007 .00068 .00006 .00005 .00001 11000 .00079 .00006 .00074 .00006 .00005 .00000 12000 .00086 .00007 .00080 .00006 .00006 .00001 13000 .00093 .00007 .00085 .00005 .00008 .00002 14000 .00101 .00008 .00092 .00007 .00009 .00001 15000 .00108 .00007 .00097 .00005 .00011 .00002 16000 .00117 .00009 .00104 .00007 .00013 .00002 17000 .00123 .00006 .00108 .00004 .00015 .00002 18000 .00130 .00007 .00112 .00004 .00018 .00003 19000 .00137 .00007 .00117 .00005 .00020 .00002 20000 .00146 .00009 .00124 .00007 .00022 .00002 21000 .00154 .00008 .00130 .00006 .00024 .00002 22000 .00164 .00010 .00136 .00006 .00028 .00004 23000 .00173 .00009 .00141 .00005 .00032 .00004 24000 .00184 .00011 .00147 .00006 .00037 .00005 25000 .00194 .00010 .00152 .00005 .00042 .00005 26000 .00207 .00013 .00158 .00006 .00049 .00007 27000 .00219 .00012 .00162 * .00004 .00057 .00008 28000 .00232 .00013 .00167 .00005 .00065 .00008 29000 .00247 .00015 .00171 .00004 .00076 .00011 30000 .00267 .00020 .00177 .00006 .00090 .00014 31000 .00294 .00027 .00194 .00017 .00100 .00010 32000 .00307 .00013 .00187 —.00007 .00120 .00020 33000 .00330 .00023 .00192 +.00005 .00138 .00018 34000 .00358 .00028 .00195 .00003 .00163 .00025 35000 .00396 .00038 .00204 .00009 .00192 .00029 36000 • .00434 .00038 .00214 .00010 .00224 .00032 37000 .00475 .00041 .00214 .00000 .00261 .00037 38000 .00525 .00050 .00217 .00003 .00308 .00047 39000 .00583 .00058 .00225 .00008 .00358 .00050 40000 .00666 .00083 .00227 .00002 .00439 .00081 41000 .00745 .00079 .00228 .00001 .00517 .00078 42000 .00845 .00100 .00228 .00000 .00617 .00100 AND QUALITIES OF CANNON POWDEE. 245 COMPRESSION OF O, Table showing the compression, restoration and permanent set, per inch in length, earned by the undermentioned weights, per square inch of section, acting on a solid cylinder 10 in. long and 1.382 in. diameter, cut from exterior of trial cylinder (0), made from equal parts of Id fusion Greenwood and Bloom f eld iron. Weight per square inch of section. Compression per inch in length. First difference. Restoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .00011 .00011 — . _ 2000 .00021 .00010 .00021 .00010 - - 3000 .00031 .00010 .00031 .00010 - - 4000 .00037 .00006 .00036 .00005 .00001 — 5000 .00041 .00004 .00038 .00002 .00003 .00002 6000 .00045 .00004 .00041 .00003 .00004 .00001 7000 .00045 .00000 .00038 —.00003 .00007 .00003 8000 .00049 .00004 .00042 +.00004 .00007 .00000 9000 .00056 .00007 .00049 .00007 .00007 .00000 10000 .00064 .00008 .00056 .00007 .00008 .00001 11000 .00067 .00003 .00059 .00003 .00008 .00000 12000 .00073 .00006 .00065 .00006 .00008 .00000 13000 .00085 .00012 .00076 .00011 .00009 .00001 14000 . .00089 .00004 .00078 .00002 .00011 .00002 ' 15000 .00098 .00009 .00086 .00008 .00012 .00001 16000 .00108 .00010 .00094 .00008 .00014 .00002 17000 .00113 .00005 .00099 .00005 .00014 .00000 18000 .00116 .00003 .00101 .00002 .00015 .00001 19000 .00125 .00009 .00109 .00008 .00016 .00001 20000 .00131 .00006 .00114 .00005 .00017 .00001 21000 .00135 .00004 .00115 .00001 .00020 .00003 22000 .00145 .00010 .00120 .00005 .00025 .00005 23000 .00151 .00006 .00124 .00004 .00027 .00002 24000 .00161 .00010 .00134 .00010 .00027 .00000 25000 .00171 .00010 .00138 .00004 .00033 .00006 ' ' 26000 .00179 .00008 .00141 .00003 .00038 .00005 27000 .00191 .00012 .00150 .00009 .00041 .00003 28000 .00200 .00009 .00152 .00002 .00048 .00007 29000 .00217 .00017 .00156 .00004 .00061 .00013 30000 .002? 1 .00014 .00157 .00001 .00074 .00013 31000 .00257 .00026 .00168 .00011 .00085 .00015 32000 .00278 .00021 .00172 .00004 .00106 .00017 33000 .00310 .00032 .00177 .00005 .00133 .00027 34000 .00336 .00026 .00181 .00004 .00155 .00022 350U0 .00369 .00033 -.00186 .00005 .00183 .00028 36000 .00403 .00034 .00187 .00001 .00216 .00033 37000 .00526 .00123 .00199 .00012 .00327 • .00111 38000 .00574 .00048 .00205 .00006 .00369 .00042 39000 .00622 .00048 .00213 .00008 .00409 .00040 40000 .00683 .00061 .00213 .00000 .00470 .00061 246 PKOPEKTIES OF METALS FOE CANNON, COMPKESSION OF 0, T. Table showing the compression, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 10 in. long and 1.382 in. diameter, cut transversely 13 in. from lower end of trial cylinder (C), made from equal parts of Id fusion Greenwood and Bbomfleld iron. Weight per Compression First Restoration First difference. Permanent set First square inch of section. per inch in length. difference. per inch in length. per inch in length. difference. 1000 lbs. .00011 .00011 2000 .00017 .00006 .00017 .00006 - — 3000 .00025 .00008 .00025 .00008 — — 4000 .00030 .00005 .00030 .00005 - — 5000 .00039 .00009 .00038 .00008 .00001 — 6000 .00039 .00000 .00037 —.00001 .00002 .00001 7000 .00048 .00009 .00044 +.00007 .00004 .00002 8000 .00054 .00006 .00050 .00006 .00004 .00000 9000 .00061 .00007 .00057 .00007 .00004 .00000 10000 .00067 .00006 .00062 .00005 .00005 .00001 11000 .00074 .00007 .00069 .00007 .00005 .00000 12000 .00079 .00005 .00073 .00004 .00006 .00001 13000 .00086 .00007 .00079 .00006 .00007 .00001 14000 .00091 .00005 .00082 .00003 .00009 .00002 15000 .00098 .00007 .00088 .00006 .00010 .00001 16000 .00106 .00008 .00095 .00007 .00011 .00001 17000 .00111 .00005 .00098 .00003 .00013 .00002 18000 .00118 .00007 .00103 .00005 .00015 .00002 19000 .00124 .00006 .00103 .00000 .00021 .00006 20000 .00132 .00008 .00111 .00008 .00021 .00000 21000 .00139 .00007 .00117 .00006 .00022 .00001 22000 .00147 .00008 .00120 .00003 .00027 .00005 23000 .00155 .00008 .00126 .00006 .00029 .00002 24000 .00163 .00008 .00129 .00003 .00034 .00005 25000 .00172 .00009 .00133 .00004 .00039 .00005 26000 .00181 .00009 .00141 .00008 .00040 .00001 27000 .00191 .00010 .00146 .00005 .00045 .00005 28000 .00201 .00010 .00150 .00004 .00051 .00006 29000 .00213 .00012 .00154 .00004 .00059 .00008 30000 .00227 .00014 .00160 .00006 .00067 .00008 31000 .00240 .00013 .00165 .00005 .00075 .00008 32000 .00257 .00017 .00169 .00004 .00088 .00013 33000 .00272 .00015 .00168 —.00001 .00104 .00016 34000 .00294 .00022 .00169 +.00001 .00125 .00021 35000 .00316 .00022 .00176 .00007 .00140 .00015 36000 .00341 .00025 .00190 .00014 .00151 .00011 37000 .00370 .00029 .00193 .00003 .00177 .00026 38000 .00409 .00039 .00202 .00009 .00207 .00030 39000 .00444 .00035 .00201 —.00001 .00243 .00036 40000 .00496 .00052 .00210 +.00009 .00286 .00043 41000 .00554 .00058 .00224 .00014 .00330 .00044 42000 .00602 .00048 .00218 —.00006 .00384 .00054 43000 .00650 .00048 .00231 +.00013 .00419 .00035 AND QUALITIES OF CANNON POWDEE, 247 COMPKESSION OF D, 0. Table showing the compression, restoration and permanent set, per inch in length, caused ly the undermentioned weights, per square inch of section, acting on a specimen 10 in. long and 1.382 in. diameter, cut from exterior of trial cylinder (D), made from Greenwood and Salisbury iron, all re-melted. Weight per square inch of section. Compression per inch in length. First difference. Bestoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 lbs. .00008 .00008 2000 .00013 .00005 .00013 .00005 - — 3000 .00017 .00004 .00017 .00004 - — 4000 .00022 .00005 .00022 .00005 — — 5000 .00028 .00006 .00028 .00006 - — 6000 .00034 .00006 .00033 .00005 .00001 — 7000 .00040 .00006 .00039 .00006 .00001 .00000 8000 .00046 .00006 .00044 .00005 .00002 .00001 9000 .00052 .00006 .00049 .00005 .00003 .00001 10000 .00059 .00007 .00056 .00007 .00003 .00000 11000 .00064 .00005 .00060 .00004 .00004 .00001 12000 .00070 .00006 .00065 .00005 .00005 .00001 13000 .00076 .00006 .00070 .00005 .00006 .00001 14000 .00082 .00006 .00075 .00005 .00007 .00001 15000 .00089 .00007 .000S2 .00007 .00007 .00000 16000 .00095 .00006 .00087 .00005 .00008 .00001 17000 .00101 .00006 .00092 .00005 .00009 .00001 18000 .00108 .00007 .00097 .00005 .00011 .00002 19000 .00116 .00008 .00105 .00008 .00011 .00000 20000 .00123 .00007 .00111 .00006 .00012 .00001 21000 .00130 .00007 .00116 .00005 .00014 .00002 22000 .00139 .00009 .00122 .00006 .00017 .00003 23000 .00149 .00010 .00128 .00006 .00021 .00004 24000 .00160 .00011 .00132 .00004 .00028 .00007 25000 .00170 .00010 .00135 .00003 .00035 .00007 26000 .00182 .00012 .00140 .00005 .00042 .00007 27000 .00196 .00014 .00143 .00003 .00053 .00011 28000 .00213 .00017 .00147 .00007 .00066 .00013 29000 .00232 .00019 .00150 .00003 .00082 .00016 30000 .00258 .00026 .00152 .00002 .00106 .00024 31000 .00287 .00029 .00156 .00004 .00131 .00025 32000 .00326 .00039 .00159 .00003 .00167 .00036 33000 .00367 .00041 .00162 .00003 .00205 .00038 34000 .00420 .00053 .00165 .00003 .00255 .00050 35000 .00496 .00076 .00172 .00007 .00324 .00069 36000 .00571 .00075 .00176 .00004 .00395 .00071 37000 .00639 .00068 .00176 .00000 .00463 .00068 38000 .00739 .00100 .00179 .00003 .00560 .00097 39000 .00843 .00104 .00098 —.00081 .00745 .00185 40000 .00918 .00075 Uncertaii .00159 l beyond 3800 +.00061 Olbs. .00759 .00014 248 PEOPEETIES OF METALS EOE CANNON, COMPRESSION OF D, T. Table showing the compression, restoration and permanent set, per inch in length, caused by the undermentioned weights, per square inch of section, acting on a solid cylinder 10 in. long and 1.382 in. diameter, cut transversely 13 in. from lower end of trial cylinder (D), made from Greenwood and Salisbury iron, all re-melted. Weight per square inch of section. Compression per inch in length. First difference. Eestoration per inch in length. First difference. Permanent set per inch in length. First difference. 1000 Its. .00000 .00000 2000 .00006 — .00006 - — — 3000 .00016 .00010 .00016 .00010 — — 4000 .00029 .00013 .00029 .00013 — — 5000 .00036 .00007 .00036 .00007 — — 6000 .00047 .00011 .00046 .00010 .00001 — 7000 .00051 .00004 .00050 .00004 .00001 .00000 8000 .00059 .00008 .00057 .00007 • .00002 .00001 9000 .00067 .00008 .00064 .00007 .00003 .00001 10000 .00075 .00008 .00069 .00005 .00006 .00003 11000 .00084 .00009 .00076 .00007 ■ .00008 .00002 12000 .00091 .00007 .00081 .00005 .00010 .00002 13000 .00099 .00008 .00087 .00006 .00012 .00002 14000 .00106 .00007 .00092 .00005 .00014 .00002 15000 .00115 .00009 .00099 .00007 .00016 .00002 16000 .00124 .00009 .00105 .00006 .00019 .00003 17000 .00128 .00004 .00106 .00001 .00022 .00003 18000 .00140 .00012 .00114 .00008 .00026 .00004 19000 .00149 .00009 .00119 .00005 .00030 .00004 20000 .00159 .00010 .00124 .00005 .00035 .00005 21000 .00170 .00011 .00130 .00006 .00040 .00005 22000 .00179 .00009 .00136 .00006 .00043 .00003 23000 .00191 .00012 .00141 .00005 .00050 .00007 24000 .00202 .00011 .00146 .00005 .00056 .00006 25000 .00213 .00011 .00151 .00005 .00062 .00006 26000 .00224 .00011 .00152 .00001 .00072 .00010 27000 .00240 .00016 .00159 .00007 .00081 .00009 28000 .00253 .00013 .00160 .00001 .00093 .00012 29000 .00275 .00022 .00167 .00007 .00108 .00015 30000 .00297 .00022 .00172 .00005 .00125 .00017 31000 .00319 .00022 .00174 .00002 .00145 .00020 32000 .00355 .00036 .00184 .00010 .00171 .00026 33000 .00379 .00024 .00181 —.00003 .00198 .00027 34000 .00412 .00033 .00182 +.00001 .00230 .00032 35000 .00457 .00045 .00187 .00007 .00268 .00038 36000 .00500 .00043 .00196 .00007 .00304 .00036 37000 .00550 .00050 .00200 .00004 .00350 .00046 38000 .00615 .00065 .00199 —.00001 .00416 .00086 39000 .00675 .00060 .00207 -(-.00008 .00468 .00052 40000 .00750 .00075 .00218 .00011 .00532 .00064 AND QUALITIES OF CANNON POWDEE. 249 Table comparing extensions of outer specimens from trial cylinders A, B, C and D. Weight pee Sqxjabb EXTENSION, PEB INCH IN LENGTH. Ihoh of Sectiok. A B C » 1000 lbs. .00002 .00003 .00004 .00003 2000 .00004 .00008 .00008 .00009 3000 .00013 .00013 .00012 .00016 4000 .00019 .00018 .00018 .00021 5000 .00023 .00024 .00022 .00026 6000 .00027 .00030 .00027 .00033 7000 .00033 .00035 .00032 .00040 8000 .00037 .00042 .00041 .00047 9000 .00043 .00048 .00046 .00054 10000 .00049 .00056 .00052 .00061 11000 .00054 .00065 .00059 .00069 12000 .00060 .00074 .00066 .00078 13000 .00066 .00083 .00075 .00085 14000 .00073 .00093 .00083 .00094 15000 .00080 .00108 .00093 .00106 16000 .00087 .00122 .00102 .00119 17000 .00095 .00140 .00114 .00135 18000 .00103 .00160 .00127 .00150 19000 .00112 .00191 .00141 .00171 20000 .00122 .00235 .00156 .00194 21000 .00133 .00291 .00173 .00231 22000 .00144 — .00196 .00278 23000 .00159 — .00221 .00327 24000 .00173 - .00252 .00424 25000 .00191 - .00287 - 26000 .00210 - - - 27000 .00237 - - - 28000 .00267 - - - 29000 .00303 ~ """ — 250 PEOPEETIES OF METALS EOE CANNON, Table comparing restoration from extension of outer specimens from trial cylinders A, B, O and D. Weight pee Square RESTORATION, PER INCH IN LENGTH. Ikoh of Section. A B c D 1000 Its. .00002 .00003 .00004 .00003 2000 .00004 .00008 .00008 .00009 3000 .00013 .00013 .00012 .00016 4000 .00019 .00018 .00018 .00021 5000 .00023 .00024 .00021 .00025 6000 .00027 .00630 .00026 .00032 7000 .00033 .00034 .00030 .00038 8000 .00037 .00041 .00039 .00045 9000 .00042 .00046 .00043 .00051 10000 .00047 .00052 .00048 .00056 11000 .00052 .00060 .00054 .00063 12000 .00057 .00067 .00060 .00070 13000 .00062 .00074 .00067 .00074 14000 .00067 .00080 .00074 .00079 15000 .00073 .00090 .00081 .00086 16000 .00078 .00098 .00086 .00094 17000 .00084 .00106 .00094 .00101 18000 .00089 .00115 .00106 .00107 19000 .00096 .00131 .00110 .00115 20000 .00103 .00140 .00118 .00121 21000 .00110 .00154 .00125 .00130 22000 .00116 — .00135 .00142 23000 .00123 — .00144 .00146 24000 .00130 - .00153 .00158 25000 .00138 _ .00160 _ 26000 .00145 — _ _ 27000 .00153 — _ _ 28000 .00163 — _ _ 29000 .00172 — — AND QUALITIES OE CANNON POWDEE. 251 Table comparing permanent set from extension of outer specimens from trial cylinders A, B, O and D. Weight pee Squabe PERMANENT SET, PEE INCH IN LENGTH. Inch op Section. A B C D 1000 lbs. 2000 — _ _ _ 3000 — _ _ _ 4000 - — — — 5000 - — .00001 .00001 6000 - - .00001 .00001 7000 - .00001 .00002 .00002 8000 - .00001 .00002 .00002 9000 .00001 .00002 .00003 .00003 10000 .00002 .00004 .00004 .00005 11000 .00002 .00005 .00005 .00006 12000 .00003 .00007 .00006 .00008 13000 .00004 .00009 .00008 .00011 14000 .00006 .00013 .00009 .00015 15000 .00007 .00018 .00012 .00020 16000 .00009 .00024 .00016 .00025 17000 .00011 .00034 .00020 .00034 18000 .00014 .00045 .00021 .00043 19000 .00016 .00060 .00031 .00056 20000 .00019 .00095 .00038 .00073 21000 .00023 .00137 .00048 .00101 22000 .00028 - .00061 .00136 23000 .00036 - .00077 .00181 24000 .00043 - .00099 .00266 25000 .00053 — .00127 — 26000 .00065 - — — 27000 .00084 - - — 28000 .00104 - - — 29000 .00131 - * — — 252 PEOPEKTIES OE METALS FOE CANNON, Table comparing extension of specimens cut from near the axis of trial cylinders A, B, O and D. EXTENSION, PER INCH IN LENGTH. Weight pee Squaee Inch of Section. A B c D 1000 libs. .00003 .00005 .00004 .00002 2000 .00007 .00010 .00008 .00008 3000 .00011 .00015 .00012 .00012 4000 .00016 .00020 .00018 .00020 5000 .00021 .00024 .00022 .00029 6000 .00026 .00030 .00027 .00037 7000 .00029 .00041 .00033 .00047 8000 .00035 .00052 .00038 .00057 9000 .00039 .00059 .00045 .00069 10000 .00050 .00070 .00051 .00081 11000 .00057 .00081 .00058 .00096 12000 .00064 .00091 .00065 .00114 13000 .00071 .00105 .00073 .00132 14000 .00079 .00119 .00081 .00156 15000 .00087 .00139 .00091 .00184 16000 .00096 .00161 .00099 .00223 17000 .00105 - .00109 — 18000 .00115 - .00120 — 19000 .00129 - .00133 — 20000 .00142 - .00147 _ 21000 .00159 - .00163 — 22000 .00175 - .00183 _ 23000 .00193 - .00207 — 24000 .00217 - .00232 — 25000 .00245 - .00269 — 26000 .00277 - .00336 _ 27000 .00334 ~ .00382 — AND QUALITIES OF CANNON POWDEB. 253 Table comparing the restoration from extension of specimens cut from near the axis of trial cylinders A, B, and D. Weight pee Squabe EBSTOBATION, PER INCH IN LENGTH. Ihoh of Seotioit. A • B c i> 1000 Its. .00003 .00005 .00004 .00002 2000 .00007 .00010 .00008 .00008 3000 .00011 .00015 .00012 .00012 4000 .00016 .00020 .00018 .00020 5000 .00021 .00023 .00022 .00029 6000 .00026 .00029 .00026 .00036 7000 .00029 .00039 .00032 .00045 8000 .00034 .00049 .00036 .00053 9000 .00038 .00055 .00043 .00062 10000 .00047 .00065 .00048 .00070 11000 .00053 .00072 .00054 .00080 12000 .00059 .00078 .00060 .00091 13000 .00065 .00086 .00066 .00100 14000 .00072 .00094 .00072 .00115 15000 .00079 .00104 .00080 .00122 16000 .00085 .00112 .00085 .00136 17000 .00091 _ .00092 — 18000 .00097 — .00098 - 19000 .00108 - .00106 - 20000 .00115 - .00114 - 21000 .00124 — .00122 — 22000 .00131 — .00131 — 23000 .00138 - .00137 — 24000 .00147 — .00146 — 25000 .00156 — .00158 - 26000 .00163 — .00188 — 27000 *~ — .00180 — 254 PKOPEETIES OF METALS EOE CANNON, Table comparing the permanent set from extension of specimens ctd from near axis of trial cylinders A, B, O and D. Weight pee Sqttaee PERMANENT SET, PEE INCH IN LENGTH. Inch op Section. A B * c D 1000 lbs. 2000 — _ _ _ 3000 — _ _ _ 4000 — — — _ 5000 — .00001 — _ 6000 - .00002 .00001 .00001 7000 - .00002 .00001 .00002 8000 .00001 .00003 .00002 .00004 9000 .00001 .00004 .00002 .00007 10000 .00003 .00005 .00003 .00011 11000 .00004 .00009 .00004 .00016 12000 .00005 .00013 .00005 .00023 13000 .00006 .00019 .00007 .00032 14000 .00007 .00025 .00009 .00042 15000 .00008 .00035 .00011 .00062 16000 .00011 .00049 .00014 .00087 17000 .00014 _ .00017 — 18000 .00018 _ .00022 _ 19000 .00021 _ .00027 _ 20000 .00027 _ .00033 _ 21000 .00035 — .00041 __ 22000 .00044 _ .00052 — 23000 .00055 _ .00070 _ 24000 .00070 — .00086 _ 25000 .00089 _ .00111 _ 26000 .00114 _ .00148 „ 27000 "" — .00202 - AND QUALITIES OF CANNON POWDEE. 255 Table comparing compression of outer specimens from trial cylinders A, B, and D. Weight pee Squaee COMPBESSION, PES INCH IN LENGTH. Iitoh of Section. A B C D 1000 lbs. .00003 .00016 .00011 .00008 2000 .00008 .00025 .00021 .00013 3000 .00013 .00032 .00031 .00017 - 4000 .00019 .00038 .00037 .00022 5000 .00023 .00045 .00041 .00028 6000 .00029 .00050 .00045 .00034 7000 .00034 .00056 .00045 .00040 8000 .00038 .00061 .00049 .00046 9000 .00044 .00066 .00056 .00052 10000 .00049 .00072 .00064 .00059 11000 .00054 .00078 .00067 .00064 12000 .00059 .00083 .00073 .00070 13000 .00063 .00089 .00085 .00076 14000 .00068 .00094 .00089 .00082 15000 .00073 .00101 .00098 .00089 16000 .00080 .00108 .00108 .00095 17000 .00085 .00114 .00113 .00101 18000 .00091 .00119 .00116 .00108 19000 .00096 .00127 .00125 .00116 20000 .00102 .00136 .00131 .00123 21000 .00109 .00145 .00135 .00130 22000 .00116 .00155 .00145 .00139 23000 .00120 .00166 .00151 .00149 24000 .00126 .00178 .00161 .00160 25000 .00136 .00192 .00171 .00170 26000 - .00142 .00211 .00179 .00182 27000 .00149 .00227 .00191 .00196 28000 .00161 .00256 .00200 .00213 29000 .00169 .00275 .00217 .00232 30000 .00181 .00297 .00231 .00258 31000 .00191 .00334 .00257 .00287 32000 .00202 .00390 .00278 .00326 33000 .00214 .00473 .00310 .00367 34000 .00227 .00542 .00336 .00420 35000 .00249 .00621 .00369 .00496 36000 .00267 — .00403 .00571 37000 .00288 — .00526 .00639 38000 .00314 - .00574 .00739 394)00 .00346 - .00622 — 40000 .00393 - .00683 — 41000 .00457 — •" ~~ 256 PEOPEETIES OE METALS FOE CANNON, Table comparing restoration from compression of outer specimens from trial cylinders A, B, O and D. Weight pee Square RESTORATION, PER INCH IN LENGTH. Inch of Seotiok. A B C D 1000 lbs. .00003 .00016 .00011 .00008 2000 .00008 .00025 .00021 .00013 3000 .00013 .00032 .00031 .00017 4000 .00019 .00038 .00036 .00022 5000 .00023 .00044 .00038 .00028 6000 .00029 .00048 .00041 .00033 7000 .00034 .00053 .00038 .00039 8000 .00038 .00057 .00042 .00044 9000 .00044 .00061 .00049 .00049 10000 .00049 .00066 .00056 .00056 11000 .00054 .00072 .00059 .00060 12000 .00058 .00076 .00065 .00065 13000 .00061 .00081 .00076 .00070 14000 .00066 .00085 .00078 .00075 • 15000 .00070 .00090 .00086 .00082 16000 .00077 .00096 .00094 .00087 17000 .00081 .00100 .00099 .00092 18000 .00087 .00104 .00101 .00097 19000 .00091 .00110 .00109 .00105 20000 .00097 .00117 .00114 .00111 21000 .00103 .00122 .00115 .00116 22000 .00107 .00127 .00120 .00122 23000 .00111 .00134 .00124 .00128 24000 .00114 .00141 .00134 .00132 25000 .00122 .00147 .00138 .00135 26000 .00125 .00155 .00141 .00140 27000 .00127 .00159 .00150 .00143 28000 .00135 .00168 .00152 .00147 29000 .00139 .00172 .00156 .00150 30000 .00145 .00173 .00157 .00152 31000 .00150 .00180 .00168 .00156 32000 .00154 .00183 .00172 .00159 33000 .00157 .00186 .00177 .00162 34000 .00161 .00195 .00181 .00165 35000 .00167 .00201 .00186 .00172 36000 .00171 — .00187 .00176 37000 .00172 — .00199 .00176 38000 .00178 - .00205 .00179 39000 .00181 — .00213 _ 40000 .00192 _ .00213 _ 41000 .00146 — — — AND QUALITIES OF CANNON POWDER. 257 Table comparing permanent set from compression of outer specimens from trial cylinders A, B, C and D. Weight pek Square PERMANENT SET, PEE INCH IN LENGTH. Ihoh op Sectiou. A B C D 1000 lbs. 2000 _ _ _ _ 3000 — - — — 4000 — — .00001 _ 5000 — .00001 .00003 — 6000 — .00002 .00004 .00001 7000 - .00003 .00007 .00001 8000 — .00004 .00007 .00002 9000 — .00005 .00007 .00003 10000 - .00006 .00008 .00003 11000. - .00006 .00008 .00004 12000 .00001 .00007 .00008 .00005 13000 .00002 .00008 .00009 .00006 14000 .00002 .00009 .00011 .00007 15000 .00003 .00011 .00012 .00007 16000 .00003 .00012 .00014 .00008 17000 .00004 .00014 .00014 .00009 18000 .00004 .00015 .00015 .00011 19000 .00005 .00017 .00016 .00011 20000 .00005 .00019 .00017 .00012 21000 .00006 .00023 .00020 .00014 22000 .00009 .00028 .00025 .00017 23000 .00009 .00032 .00027 .00021 24000 .00012 .00037 .00027 .00028 25000 .00014 .00045 .00033 .00035 26000 .00017 .00056 .00038 .00042 27000 .00022 .00068 .00041 .00053 28000 .00026 .00088 .00048 .00066 29000 .00030 .00103 .00061 .00082 30000 .00036 .00124 .00074 .00106 31000 .00046 .00154 .00089 .00131 32000 .00048 .00207 .00106 .00167 33000 .00057 .00287 .00133 .00205 34000 .00066 .00347 .00155 .00255 35000 .00082 .00420 .00183 .00324 36000 .00096 - .00216 .00395 37000 .00116 - .00327 .00463 38000 .00136 - .00369 .00560 39000 .00165 — .00409 - 40000 .00201 - .00470 - 41000 .00311 — — ~ * 258 PEOPEETIES OF METALS EOK CANNON, Table comparing compression of specimens cut transversely 13 in. from lower ends of trial cylinders A, B, C and D. Weight per Squaeb COMPBESSION, PER INCH IN LENGTH. Ihch of Sectiok. A B C D 1000 lbs. .00012 .00010 .00011 .00000 2000 .00022 .00021 .00017 .00006 3000 .00032 .00029 .00025 .00016 4000 .00042 .00034 .00030 .00029 5000 .00051 .00041 .00039 .00036 6000 .00053 .00047 .00039 .00047 7000 .00060 .00052 .00048 .00051 8000 .00068 .00059 .00054 .00059 9000 .00074 .00066 .00061 .00067 10000 .00080 .00073 .00067 .00075 11000 .00086 .00079 .00074 .00084 12000 .00093 .00086 .00079 .00091 13000 .00102 .00093 .00086 .00099 14000 .00108 .00101 .00091 .00106 15000 .00114 .00108 .00098 .00115 16000 .00121 .00117 .00106 .00124 17000 .00127 .00123 .00111 .00128 18000 .00135 .00130 .00118 .00140 19000 .00142 .00137 .00124 .00149 20000 .00151 .00146 .00132 .00159 21000 .00159 .00154 .00139 .00170 22000 .00163 .00164 .00147 .00179 23000 .00172 .00173 .00155 .00191 24000 .00181 .00184 .00163 .00202 25000 .00191 .00194 .00172 .00213 26000 .00200 .00207 .00181 .00224 27000 .00208 .00219 .00191 .00240 28000 .00217 .00232 .00201 .00253 29000 .00227 .00247 .00213 .00275 30000 .00237 .00267 .00227 .00297 31000 .00249 .00294 .00240 .00319 32000 .00263 .00307 .00257 .00355 33000 .00276 .00330 .00272 .00379 34000 .00294 .00358 .00294 .00412 35000 .00309 .00396 .00316 .00457 36000 .00330 .00434 .00341 .00500 37000 .00351 .00475 .00370 .00550 38000 .00378 .00525 .00409 .00615 39000 .00408 .00583 .00444 .00675 40000 .00441 .00666 .00496 .00750 41000 - .00745 .00554 _ 42000 - .00845 .00602 _ 43000 *■* "" .00650 - AND QUALITIES OF CANNON POWDER. 259 Table comparing restoration from compression of specimens cut transversely 13 in. from lower ends of trial cylinders A, B, O and D. Weight fee Square RESTOBATION, PEE INCH IN LENGTH. Inch of Section. A B C D 1000 lbs. .00012 .00010 .00011 .00000 2000 .00022 .00021 .00017 .00006 3000 .00032 .00029 .00025 .00016 4000 .00042 .00034 .00030 .00029 5000 .00051 .00041 .00038 .00036 6000 .00055 .00046 .00037 .00046 7000 .00060 .00050 .00044 .00050 8000 .00068 .00056 .00050 .00057 9000 .00073 .00062 .00057 .00064 10000 .00078 .00068 .00062 .00069 11000 .00083 .00074 .00069 .00076 12000 .00089 .00030 .00073 .00081 13000 .00096 .00085 .00079 .00087 14000 .00101 .00092 .00082 .00092 15000 .00106 .00097 .00088 .00099 16000 .00112 .00104 .00095 .00105 17000 .00114 .00108 .00098 .00106 18000 .00120 .00112 .00103 .00114 19000 .00125 .00117 .00103 .00119 20000 .00131 .00124 .00111 .00124 21000 .00137 .00130 .00117 .00130 22000 .00138 .00136 .00120 .00136 23000 .00143 .00141 .00126 .00141 24000 .00148 .00147 .00129 .00146 25000 .00155 .00152 .00133 .00151 26000 .00161 .00158 .00141 .00152 27000 .00165 .00162 .00146 .00159 28000 .00169 .00167 .00150 .00160 29000 .00173 .00171 .00154 .00167 30000 .00177 .00177 .00160 .00172 31000 .00181 .00194 .00165 .00174 32000 .00186 .00187 .00169 .00184 33000 .00191 .00192 .00168 .00181 34000 .00197 .00195 .00169 .00182 35000 .00199' .00204 .00176 .00189 36000 .00204 .00214 .00190 .00196 37000 .00206 .00214 .00193 .00200 38000 .00209 .00217 .00202 .00199 • 39000 .00215 .00225 .00201 .00207 40000 .00224 .00227 .00210 .00218 41000 — .00228 .00224 — 42000 _ .00228 .00218 - 43000 — — .00231 "■* 260 PEOPEETIES OF METALS FOE CANNON, Table comparing permanent set from compression of specimens cut transversely 13 in. from lower ends of trial cylinders A, B, C and D. PERMANENT SET, PER INCH IN LENGTH. Weight peb Square Inch of Section. A B C © 1000 lbs. _ 2000 - - - - 3000 - — — — 4000 - — - — 5000 - - .00001 — 6000 - .00001 '.00002 .00001 7000 - .00002 .00004 .00001 8000 - .00003 .00004 .00002 9000 ' .00001 .00004 .00004 .00003 10000 .00002 .00005 .00005 .00006 11000 .00003 .00005 .00005 .00008 12000 .00004 .00006 .00006 .00010 13000 .00006 .00008 .00007 .00012 14000 .00007 .00009 .00009 .00014 15000 .00008 .00011 .00010 .00016 16000 .00009 .00013 .00011 .00019 17000 .00013 .00015 .00013 .00022 18000 .00015 .00018 .00015 .00026 19000 .00017 .00020 .00021 .00030 20000 .00020 .00022 .00021 .00035 21000 .00022 .00024 .00022 .00040 22000 .00025 .00028 .00027 .00043 23000 .00029 .00032 .00029 .00050 24000 .00033 .00037 .00034 .00056 25000 .00036 .00042 .00039 .00062 26000 .00039 .00049 .00040 .00072 27000 .00043 .00057 .00045 .00081 28000 .00048 .00065 .00051 .00093 29000 .00054 .00076 .00059 .00108 30000 .00060 .00090 .00067 .00125 31000 .00068 .00100 .00075 .00145 32000 .00077 .00120 .00088 .00171 33000 .00085 .00138 .00104 .00198 34000 .00097 .00163 .00125 .00230 35000 .00110 .00192 .00140 .00268 36000 .00126 .00224 .00151 .00304 37000 .00145 .00261 .00177 .00350 38000 .00169 .00308 .00207 .00416 39000 .00193 .00358 .00243 .00468 40000 .00217 .00439 .00286 .00532 41000 - .00517 .00330 — 42000 - .00617 .00384 — . 43000 "*" — .00119 - AND QUALITIES OF CANNON POWDEE, 261 8 O is 8 8 a ^ so ft* 2 «s> .8 hH M CO t- •* CM CO CO t- t- OS CO 00 1—1 to CM CO GO lO CO CO CO *4^ CO CM T-H o -+ r-H Ol 1 — 1 r^ A CM O o o O O o O ^j CM O o o o o o o ICL t^ r ~ l CM CO lO CM ■* oo CO CO CM •* o © 00 CM Id CO CO t- CM T-H o o ■* f-H CM ■* ■ — i CO CM lO A oo "■* CM »-H T-H CO T-H T-H CM o lO o o o o o o C2 t- U CM o o o o o o O CO t-^ o CO CM CM T-H CM CO CO -* CM CO 1— < O CO T-H ■# 00 o CO CO CM T-H T— f o CO CM T-H CO M O o o o O o o o CM o o o o o o CO CO CO t^ co m CM f-H tH t- T-H T-H o CO if CO CO 6 CO oo ks co CM o CI . 1 fc- CM T-H T-H CO CM -* o A fc- O o o o o o CO »— ' O o o o o o CO t^ r-H CO CO T-H CO -H CO -* CD Oi o o i—5 CO CO CO r-H o CD r— ( 1>- T-H ■* CO T-H T-H CO T-H T-H CO CO t- o o o o o o CO CO ««, CM O o o o o o r-H t- t^ T-H i— 1 i— I o CO CO CT T-H CD t~ CM o © KS o t- CO ■* CO CO *o CM CO CO T-H CM r-H o t~ CO o o o O o o CM CO < CM o o o o o o CM T-H en t-^ TO -O! o . , , a 'cG p . TO o o o CO 4J c3 p" o 'to TO TO o CD >H cS fens H • • cd -T^» CD a a o 'to o o CO o o lO CO ■a =3 II CO a p CM a CD •4^ la 2 o o on in * M rd a o o P HJ CD m (h o p _o CD a _p O 'to TO CD O P 1 -fa K 'm a c8 o fH a p CD cS 03 C4 © CD Ph P ^o tH Ph H-3 TO *0Q -s +3 02 -+j eft a 'to 02 'w g CD <3 OP TO TO CD tH Sh a *-+3 c*. TH o TO a o o -T^> e3 g o a P a "■+3 .9 *-+3 .1 -12 a %-3 TO P c3 p P P p o O CD CD CO. P Sh P •-C3 p ■73 a 60 P Ph CO sS 66 262 PROPERTIES OF METALS FOB CANNON, A, o., broke at the Repetition of Strain. 2301st repetition of 22000 lbs. per square inch. A, o., duplicate, broke at the 282d B, o., broke at the 252d B, o., duplicate, broke at the 150th O, o., broke at the 651st C, o., duplicate, broke at the 457th D, o., broke at the 172d 26000 " it tt 20000 " it a 20000 " it it 22500 " a it 23500 " a a 21600 " et a Water Test, and Tangential Resistance. A B C D Water forced through the pores, at (per square inch), . Tangential resistance, or bursting force, per square inch, lbs. 25464 42908 lbs. 15276 30176 lbs. 21845 43162 lbs. 19099 37815 The cylinders for water test and tangential resistance were 5 inches long, 1 inch bore, open at both ends, and 1 inch thick ; they were burst, by pressure, upon a composition of three parts of bees-wax to one of tallow. The following ratios serve to compare the general properties of one variety of iron with those of another : — Mean tenacity X mean extensibility X mean restoration _ Compression at 35000 lbs. X mean set Mean tenacity X mean extensibility X mean restoration Compression at 35000 lbs. X mean set Mean tenacity X mean extensibility X mean restoration _ Compression at 35000 lbs. X mean set Mean tenacity X mean extensibility X mean restoration _ Compression at 35000 lbs. X mean set 38762, for cylinder A. 8764, for cylinder B. 29205, for cylinder O. 11809, for cylinder D. Cylinder (A) was the only one of the four found to be entirely sound in all its parts, all the others being more or less defective in their ulterior parts. The curves on Plates 10 to 17 are constructed from the foregoing tables, and give an ocular exhibition of the properties of the iron tested, and an ocular comparison of the properties of the different irons. PLATE X' Curves comparing extensibility, compressibility, and set of *outer Specimens Jrom Oyfindftr A . 0100 009S ■0090 0C85 0080 CC/S cote 0065 0060 ooss 0045 C'040 O035 0030 002.5 0020 00 iS 0010 , ooos /ewcoiis. 20CCO V/PCCC J>LATU XL Curves comparing extensibility: comjjTessib'ihirfjjid set of outer Specimens item Cylinder JB. Ci 0S - COK cose fcoeoihs 400CO .50000 61 Ooii 10000 TLATUXn. ( iv yes comparing eMeitsIbj7itT,cowj)ressilaBty,aDcl set of outer Specimens from Cylinder C 0. CandD. VfCC 609o ■eo&c -60 #5 <'0S6 trOiS A-C70 6C65 €fs>c etss &0-50 CX4-5 cclc ■e<3;7 4030 V0Z5 40ZO V015 G0 10 VOCc wooo SO&O erect. ■X etc tlate^tj: Company/? compressibility and set of outer specimens from irial Cyimder.r A. J3. f and 21 • CC.9V COSS cc*c 'cor ~corc rccc-j 006C OOS5 00SO ocis 6C4-C oojs 0030 .OOZS fOOOO 16s 2ffOCCP 30(300 . co/s CCfC ' cot... 'ccec JPLsiTEjm- ( omp<5trbi<£ extensibility- and set of specimens Irvm new dxes of CybiidersAfi, CancLD. costs 00.90 ooss 4— L 00 *C 007-i 6670 006S -CWSS -eeso 0C4fS f>C4fj >-- J 4- 0C3S 003c OOZS .(CIO -fioos ■ OC-/0 .ooos /OCOCCi.y 2CCCC joocc lecoo go tec 76CCC AND QUALITIES OF CANNON 20WDEB. 263 FABRICATION OF 15-INCH GUN. The foregoing experiments having shown that from our present knowledge of the requisite properties of gun iron, re-melted Bloomfield iron would be more likely to give a durable gun of this size than any other tested, it was determined that the gun should be made of this iron. Accordingly, on the 22d of December, 1859, the furnaces Nos. 1, 2, and 3, were charged with this quality of iron. Furnace No. 1 received 16046 lbs. " No. 2 " 32055 " No. 3 " 28069 " Total charge, 76170 lbs. The gun mould was placed in the pit on the same day, the pit having been previously well dried by fire, and the grate bars arranged in order for heating the mould, so as to prevent the gun from cooling from the exterior. Casting. The furnaces were lighted on the 23d; No. 2 at 5h. 30' A. M., and Nos. 1 and 3 at 6h. A. M., and the iron in all the furnaces was melted at lOh. A M. Furnace No. 2 was tapped at lOh. 51' ; No. 1 at lOh. 57', and No. 3 at llh. 2' ; and the whole time of running was 21'. Furnace No. 1 was tapped a little before No. 2 had run out, and No. 3 a little before No. 1 had run out ; so that the iron flowed from but one furnace at the same time. The iron was conducted from the furnaces, in runners, to a pool near the gun mould, from which it ran in two runners to the gun mould, which it entered through side gates cut into the walls of the mould, the main gates having branch gates inclined upward, at intervals of 12 inches from bottom to top of the mould. The iron entered in directions towards the axis of the mould, and not so as to produce a swirl. The first metal that entered the mould had a considerable distance to run, along a comparatively cold runner, and was so cold, by the time it reached the gun mould, as to require assistance to make it flow. 264 PEOPEETIES OF METALS FOE CANNON, This was very soon corrected, as the runners became heated by the passing metal, after which the casting proceeded, with perfect regularity, to completion. It was intended that the coldest metal should enter first, and remain in the breech of the gun ; the hottest metal, that in No. 3 furnace, which was close to the gun mould, was reserved for the chase and sprue head, in order that it might "feed" as long as possible. The " swirl " was not used ; and it is believed never should be in casting cannon, for the reason that it forces the scoria and lightest metal in around the bore, where the heaviest and best metal is required. Cooling. Water circulated through the core barrel at the rate of about 40 gallons per minute, entering at 36°, and leaving at 60° Fahrenheit. AND QUALITIES OF CANNON POWDER. 265 Cooling ' Table. Hours after Casting. Water entered at Left at Change of Temperature. ih. 36° 58° 22° = 7- for 3 hrs. flow. 4 36 56 20 7 36 54 18 10 36 52 16 13 36 51 15 16 36 49 13 18 36 48 12 21 36 47 11 24 36 47 11 The core barrel was then removed, 24h. 30m. 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76 79 82 85 88 91 94 97 100 103 106 109 112 115 118 121 124 127 130 142 145 168 and water circulated through the cavity thus left* at the rate of 43 gallons per minute. 36° 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36 36. 36 36 36 36 36 36 36 36 36 36 36 86° 82 76 67 62 59 56 52 50 49 48 47 47 47 47 46 45 44 43 43 43 43 43 42 42 42 42 42 41 40 39 38 38 38 38 38 37 37 36° 5' 36° 5' 50° 46 40 31 26 23 20 16 14 13 12 11 11 11 11 10 9 8 7 7 7 7 7 6 6 6 6 6 5 4 3 2 2 2 1 1 0°5' 0° 5' 690° 67 266 PEOPEETIES OF METALS EOE CANNON, The water was then stopped, and the flask and mould removed from the chase of the gun ; just one week being occupied in cooling. The gun and mould, as far as stripped, were perfectly cold, and the temperature of the water that remained in the bore of the gun only reached 50° at three hours after the water had ceased to flow, and only 60° at twenty-four hours after ; this gun having been cooled in less time than is required for the proper cooling of an '8-inch solid cast gun. Temperature of Pitt. Fire was lighted in the pit at 6 P. M. on the 23d (day of casting), and kept up till 12 M. on the 26th ; after which time no more fuel was added, and the fire gradually burned out. Up to this time, the temperature of the pit had been such that the iron cover would char, and ignite dry wood in 10' to 15' ; the lower part of the flask being at a dull red heat. The pit cover was removed at 6 P. M. on the 28th, at which time the temperature of the pit was 115°. The pattern for this gun was 50 inches in diameter at the largest part, and 38 inches at the muzzle ; and there were about 9 inches thickness of mould around the gun below the trunnions, and about 8 inches above that point. The flask was of circular cross section in all its parts. Rate, Extent, and Effects of Internal Cooling. Previous to the removal of the core barrel, water circulated at the rate of 40 gallons per minute. Taking the weight of one gallon = 8.33 lbs., and we have for the weight of water that passed through the gun per hour = 19992 lbs., and for that which passed in three hours = 59976 lbs. And since the temperature was (with one or two exceptions) taken at intervals of three hours, if we suppose the temperature to remain constant for intervals of three hours, and equal to that at the ends of these intervals, then 59976 lbs., multiplied by the sum of the temperatures, up to any given time after casting, will give something less than the number of pounds of water which the heat carried off by it would raise one degree in temperature. This number, divided by the number of pounds of water which the heat given out by one pound of melted iron would raise one degree, in cooling AND QUALITIES OF CANNON POWDEE. 267 down to 105°, will give the number of pounds of iron which the heat carried off by the water would raise from 105° to the casting temperature. Major Wade found (Reports of Experiments on Metals for Cannon, p. 303) that one pound of iron, in cooling from the common casting temperature down to 105°, gave out sufficient heat to raise 455 lbs. of water one degree. Calculations based upon the foregoing hypothesis and data show that, up to the time the core barrel was removed, the circulating water had carried off sufficient heat to cool 12879 lbs. of iron from the temperature of casting down to 105° ; and that, at fifty-five hours after casting, when the change of temperature became constant, there had been sufficient heat abstracted by the water to cool 55673 lbs. from the casting temperature to 105° ; and, consequently, since the total weight of the casting was 76170 lbs., that over two-thirds of all the heat in the gun at the time of casting had, at fifty-five hours after, been carried off from the interior of the gun. When the change of temperature in the circulating water becomes constant, it indicates that the gun is receiving as much heat from the exterior as the water is carrying off from the interior. There must, consequently (at this time), be a regular decrease in the temperature of the metal from the exterior to the interior, and a greater amount of contraction yet to take place in the exterior than in the interior portions of the gun. In undergoing this greater amount of contraction, the exterior portions of the gun must, when cold, hug or bind tightly upon the interior ; which, in addition to placing the strongest and most compact metal around the bore of the gun, is the principal object of internal cooling. Mechanical Tests. Specimens taken from rings cut from the chase of the gun gave, as tested by me, — For inner ring, Density = 7.204, Tenacity = 34.187. For outer ring, Density = 7.216, Tenacity = 33.758. As tested by founders, Density = 7.234, Tenacity = 36.163. Specimens from the trial cylinder, made of the same composition of iron (re-melted Bloomfield), gave, — ■ 268 PKOPEETIES OF METALS FOE CANNON, For inner specimen, D. = 7.274, and S. = 31.681. For outer specimen, D. = 7.266, and S. = 30.117. From which it appears, that while the density has been diminished, the tenacity has been increased, by casting in the larger mass. The metal in this gun is fine grained, presenting an uniform and rather finely mottled appearance ; the fractured surface is rough, jagged, and sharp to the touch, with strongly adhering fragments, which are indications of toughness. The elasticity, extensibility, and compressibility of the iron in the gun have not been determined ; but, as far as any judgment can be formed from the tests to which it has been subjected, and from the fracture and appearance of the metal, it is superior in quality to what was expected in so large a casting. Of the Rate of Application of Force. The remarks following this point (*) in the discussion of this subject, at page 42, Report of 1857, require modification. The excess of strain due to the rate of application of any force, above that due to its statical equilibrium, is caused by the momentum or living force developed in both the straining and resisting bodies, up to the time when they attain their position of statical equilibrium, or by the momentum with which they arrive at that position. To illustrate : suppose the sum of the masses of the resisting body [a b), and of the weight (p), to become infinitely , ^ J _,,. - ...__^ I I . ^ small, as compared with that assigned them in a[Z /r ■n^^w the discussion above referred to ; and the force **- y J>" of gravity to be so increased as to cause their weight to remain constant, and the resisting power of (« I) to remain the same. These hypotheses would not change the position of statical equilibrium, and the moving and resisting bodies would reach that position with the same velocity as before ; but their mass being, by hypoth- esis, infinitely small, their momentum at that position would also be infinitely small, as compared with its value under the former hypothesis, and they would consequently be carried, by that momentum, only an infinitely small AND QUALITIES OF CANNON POWDER. 269 distance beyond the position of statical equilibrium. The ultimate strain would, consequently, under this hypothesis, be independent of the rate of application of the straining force. The statical pressure exerted upon that portion of the surface of the bore around the seat of the charge, in firing a 10-inch gun with service charges and solid shot, cannot be less than 50000 lbs. per square inch. The weight of a body that would produce this amount of statical pressure, per square inch, on the area of a cross section of the bore of that gun, would = 78.54 X 50000 = 3927000 lbs. This would be the weight of the moving or straining mass necessary to render the remarks, in the discussion above referred to, applicable to a 10-inch gun ; whereas, in the discharge of cannon, the charge of powder is the moving mass, and that portion of the gun around the seat of the charge is the resisting mass. The extensibility of gun iron is, at the highest estimate, not over .004 in. per inch in length. The increase in diameter of the bore of a 10-inch gun would therefore be, at the moment of interior rupture, = .04 in., and the extent of radial motion of the surface of the bore would = .02 in. The surface of the bore would have a greater extent of motion than any other part ; and if there were no other resistance to motion than the inertia of the mass of the metal around the seat of the charge, the velocity devel- oped in that mass, in passing over a space of .02 in., would be very trifling indeed, and the momentum correspondingly small. The sum of the moving and resisting masses, in the case of a 10-inch gun, as compared with that of a body whose weight = 3927000 lbs., would be very small ; nor can the radial velocity of the charge, at the moment when the bore attains the diameter due to the statical pressure exerted upon it, be so great as to render its momentum of any considerable magnitude ; from which it follows that, in firing cannon, the excess in strain upon the gun, above that due to statical pressure, caused by the most rapid rate of applica- tion, or development of that pressure, is a very small percentage of the total strain. This reasoning, and the conclusion to .which it leads, must not, however, be construed into a disregard of the rate of combustion of the charge, for this is of primary importance ; but from causes entirely different from that discussed above. 270 PEOPEETIES OF METALS FOE CANNON, Experiments indicate that the pressure due to the combustion of gun- powder, in a space equal to it own volume, is not less than 200000 lbs. per square inch ; a pressure far greater than can be resisted by any known material. With a given mass of shot free to move, and a given charge of powder to move it, the more rapid the rate of combustion of that charge, the less dis- tance will the shot have moved during the time of combustion, and the nearer will the pressure developed approach that due to the combustion of powder in its own volume. And the fulminates, and very fine-grained quick powder, behind heavy projectiles, so nearly approach this condition, as to burst the gun before the shot has had time to move ; not from the momentum with which the straining and resisting masses reach their position of statical equi- librium, but from absolute statical pressure. Of the Difference in Effect due to Difference in the Times of Action of a Given Force. This subject has already been more than once referred to, but not in such a manner as to attach to it that degree of importance to which it is believed to be entitled. It is well known and understood in architecture and practical mechanics, that a given beam of wood, or bar of iron, will sustain for a limited time a weight which would be certain, ultimately, to break it ; and, in general terms, that the rupturing force is a decreasing function of the time required for it to produce rupture. It is believed, however, that we have not heretofore properly appreciated the effect of time on the resistance which a body can offer, where the absolute difference in the times of action is small, but where the ratio of the maximum to the minimum time of action is very great. For example, the time required to rupture a tensile specimen of cast iron on the testing machine is, say five minutes. This is a small absolute space of time, and the difference between this and any smaller space must be still less ; but as compared with the length of time during which the maximum pressure is exerted upon the bore of a gun at a single discharge, it becomes very great ; probably as great as the ratio of the time of existence of any known structure of either wood or iron to that required to test the strength of a single specimen of either material. And if so, why should not the resistance of a gun or shell, to a single dis- AND QUALITIES OF CANNON POWDEB. 271 charge, be as much greater than indicated by the test specimen, as the perma- nent architectural load required of any material is less than that indicated by the test specimen ? The results of different experiments which I have made, indicate that such is the fact. For example, in bursting cylinders with powder, (see page 192, Report of I860,) set No. 1, with a thickness of metal of. 5 inches, gave a burst- ing pressure per square inch =37842 lbs., and requiring a tensile strength of iron = 75684 lbs. per square inch, while the tensile strength of the iron by the testing machine was only 26866 lbs. And in set No. 4, (same page and Report,) with two inches thickness of metal, the bursting pressure was 80229 lbs. per square inch, while the most that it could have been by the testing machine would be twice the tensile strength, or 53732 lbs. These same results, as well as others, show important differences in resist- ance due to differences in time of action, when the greatest duration was so small as to be entirely inappreciable to the senses. Take, for example, sets Nos. 1 and 2, of the same cylinders just referred to. These sets were both of the same interior capacity, same metal, as near as could be, and were burst by equal charges of powder of the same quality. Set No. 1 was .5 in. thick, and set No. 2 was 1 inch thick. The mean bursting pressure of set No. 1 was 37842 lbs. per square inch, while set No. 2 was only 38313 lbs. One cylinder of set No. 2 required two charges to burst it, the indication of pressure being something less for the second than for the first charge. Now the only and true explanation of these results is believed to be, that 38313 lbs. was the pressure due to the combustion of the charge of powder used, in the space in which it was burned ; that it did not greatly exceed the resisting power of the cylinder of set No. 2, and required a greater, though still inappreciable, length of time to produce rupture, (as indicated by the fact of one cylinder forcing the whole products of combustion of one charge out through a hole one-tenth of an inch in diameter, without bursting,) while it greatly exceeded the resisting power of set No. 1, and consequently burst that set in much less time, but not before almost the full pressure due to the charge of powder used, had been developed. Another example occurs in bursting shells with powder, (see page 206, Report of I860,) where a shell of 5 inches interior diameter, and 6.5 in. thick, burst 272 PKOPEETIES OF METALS FOE CANNOU, at 42500 lbs. per inch, before the gas evolved from a volume of powder equal to half its interior capacity could escape through a hole one-tenth of an inch in diameter; while another shell 3.85 in. interior, and 12 inches exterior diameter, resisted the pressure due to the combustion of its entire capacity of powder, up to 185000 lbs. per square inch. Now the difference in the times of action of the forces in all these examples was entirely inappreciable to the senses, yet the ratio of the greatest to the least must have been very considerable. And in the ordinary discharge of cannon the gun is subjected, at each dis- charge, to a force which would inevitably burst it, if permitted to act for any appreciable length of time ; so that it may be said that cannon do not burst because they have not time to do so before the bursting pressure is relieved. The very short duration of the maximum pressure also renders important the inertia of the mass of the gun around the seat of the charge ; as I have no doubt that an increase of mass in this part of the gun, provided it introduce no injurious strain, would increase its endurance. The above results and conclusions indicate that it is highly desirable, if practicable, to adopt a standard of time for the action of the rupturing force, in testing the strength of any material ; as I doubt not that many apparent discrepancies in the strength of the same material, by different experimenters, may be due to the difference in the length of time occupied in producing rupture. Experiments with Powder of variable Diameter of Grain. For the purpose of further determining the relation between the maximum pressure of gas, and the velocity of shot, due to equal charges of powder, of the same quality in all respects, except in diameter of grain, the following series of fires was made with the 11-inch gun, prepared as heretofore explained, and the same cylindrical shot as was used in the previous experi- ments with that gun. AND QUALITIES OF CANNON POWDER. 273 The following table exhibits the results obtained : — Number of Size of Grain. Weight of Charge. Weight of Shot. Velocity of 8hot. PEESSOBE OP GAS, IN POUNDS. Fires. At bottom of bore. At 14 in. At 28 in. 1 2 3 4 5 .6 in. .6 .6 .6 .6 12.67 lbs. 12.67 12.67 12.67 12.67 186.3 lbs. 186.3 186.3 186.3 186.3 956 1073* 956 953 870 24970 20600 20340 20080 20850 9270 11580 9270 10300 11330 6430 11330 7720 8240 6430 933 21370 10350 8030 1 .5 in. 12.67 lbs. 186.3 lbs. 991 21100 12360 6430 2 .5 12.67 186.3 890 22910 12360 7210 3 .5 12.67 186.3 899 21100 11330 7460 4 .5 12.67 186.3 950 20850 9270 7710 5 .5 12.67 186.3 Failed. 20080 10550 7710 932 21210 11170 7300 1 .4 in. 12.67 lbs. 186.3 lbs. 890 28840 10800 7210 2 .4 12.67 186.3 811 24720 10300 7210 3 .4 12.67 186.3 955 24720 10300 7210 4 .4 12.67 186.3 889 22910 13900 8240 5 .4 12.67 186.3 860 26780 8490 6430 881 25590 10750 7260 1 .3 in. 12.67 lbs. 186.3 lbs. 942 33990 10300 6430 2 .3 12.67 186.3 882 30900 8490 6430 3 .3 12.67 186.3 847 24720 10300 6680 4 .3 12.67 186.3 877 " 39140 11070 6180 5 .3 12.67 186.3 904 47890 13390 7710 890 35330 10710 6680 1 2 3 .3 in. .3 .3 12.67 lbs. 12.57 12.67 186.3 lbs. 186.3 186.3 1037® 1087" 912 61800 67980 67980 10800 16480 16990 7210 8240 9520 912 65920 14750 8320 * Target wire cut by cartridge block ; not included in mean. 274 PROPERTIES OF METALS EOR CANNON, ETC. Table showing the velocity of shot, in feet, per second, and the pressure of gas, per square inch, in pounds, due to equal charges of powder, of the same composition, and differing only in size of grain — each result being the mean of five fires with the 11-inch gun, the same shot being used in all the fires. Diameter of Weight of Charge. Weight of Shot. Telocity of Shot. PRESSURE OP GAS, IN POUNDS. Grain. At bottom of bore. At 14 in. At 28 in. .6 in. .5 .4 .3 .3* 12.67 lbs. 12.67 12.67 12.67 12.67 186.3 lbs. 186.3 186.3 186.3 186.3 933 932 881 890 912 21370 21210 25590 35330 65920 10350 11170 10750 10710 14750 8030 7300 7260 6680 8320 * Powder of 1859, but not so hard pressed as that of 1860. Mean of three fires, housing having blown out of gun at the 3d fire, thus preventing a greater number. Respectfully submitted. T. J. RODMAN, Capt. Ord. Watertown Absenal, April 23, 1860. REPORT OS THE T H I A. L OF 10-INCH GUNS, Nos. 362 AND 363, [CONTINUED FROM PAGE 126.] REPORT TEIAL OF 10-INCH GUNS, NOS. 362 AND 363. [Continued from page 126.] In addition to the 24.50 service and 2 proof charges, to which each of these guns had been subjected at Pittsburg, they have each been fired, at the Fort Monroe Arsenal, by Capt. A. B. Dyer, 1632 rounds with 18 lbs. of powder and a solid shot, making a total of 4082 service and 2 proof charges for each gun ; neither gun broken. The cracks referred to in the solid cast gun (page 125) were considerably increased by the additional firing; one, at the conclusion of the firing, extending from the interior of one of the vents almost to the axis of the gun. The solid gun continued to deteriorate more rapidly than the hollow one, being much more, and more irregularly, enlarged and cut away by the gas, around the seat of the shot. The accompanying diagram, received from Capt. A. B. Dyer, (Plate 1,) shows the profiles of the elements of the bores, in vertical planes, after the above named number of fires. The right lines, in black, show the original lines of the bores. The irregu- larly curved lines, in red, show what would be the intersection of the surface of the bore of the hollow cast gun by a vertical plane through its axis, in its present deteriorated condition, and the irregularly curved lines in blue show the same for the solid cast gun. Owing to the greater enlargement and consequent increase of windage of the solid, than of the hollow cast gun, the latter has, for the last 2000 rounds, been subjected to a greater pressure of gas at each discharge, than the former. A short time before the suspension of the trial a mean of 19 70 278 PBOPEETIES OF METALS FOE CANNON, ETC. rounds, from each gun, with equal charges (18 lbs.) of the same quality of powder, and one solid shot, showed the maximum pressure at each discharge to be for the hollow gun 39793 lbs. per square inch, and that for the solid gun 35306; or that the hollow cast gun was subjected, at each discharge, to a pressure 4487 lbs. per square inch greater than that to which the solid cast gun was subjected. These pressures are both far below those due to the same charges, fired in a new unenlarged gun ; recent experiments having shown that with 18 lbs. of Dupont's No. 10 experimental powder, common cannon, and one solid shot, from a 10-inch gun, the maximum pressure per square inch was, for a mean of 4 fires, 71950 lbs. REPORT OF THE INSPECTION, TRANSPORTATION, MOUNTING AND TRIAL OF THE 15-INCH GUN. REPORT E THE INSPECTION, TEANSPOETATION, MOUNTING AND TEIAL OF THE 15-INCH GUN. Inspection. This gun was completed early in May, 1860, when it was carefully in- spected. The casting appeared to be perfectly sound in all its parts, the finished surfaces, both of the exterior and of the bore, being remarkably perfect. The dimensions of the gun are as follows, viz. : — Lengths. Total length of gun, . . . . . . 190 inches. From axis of trunnion to rear of breech, . . 71.3 Length of trunnions, ...... 6.5 " of cylinder of bore, .... 156. " of ellipsoidal termination of bore, . . 9. " from face of muzzle to maximum diameter, 155. Distance between rimbases, . . . . 48.1 Diameters. Maximum diameter, . Diameter of muzzle, . " of bore, " of cascable, " of trunnions, " of vent, Width of ratchet, Depth of Estimated weight of gun, Preponderance at ratchet about, 48.1 25. 15. 41.5 15. . .2 6. 1. 49099 lbs. 1200 71 282 PROPERTIES OF METALS FOE CANNON, The bore of this gun was remarkable for the smoothness of its surface, and the uniformity of its diameter, the star gauge not showing a variation, from 15 inches, of so much as one-thousandth of an inch in the whole length of bore. Transportation. For this purpose two strong trussed beams, 50 feet long, were prepared. These beams were placed parallel to each other, and about 36 inches apart, their ends resting upon two bolsters placed transversely across the middle points of two 8- wheeled platform cars. The gun was suspended under the two trussed beams, and between the cars ; so that its weight was equally distributed over the 16 wheels of the two cars. Thus mounted, it passed over the Pennsylvania Central Railroad to the Susquehanna River, thence on the Northern Central to Baltimore, and thence by the Washington Branch to Washington City, where it arrived in perfect order without change of cars. Speed of cars from 12 to 15 miles per hour. The gun was hauled from the Washington Depot to the Potomac River on a truck wagon, used for hauling heavy stones, where it was shipped on board a vessel used for carrying heavy stones for the Capitol and Treasury Extensions. From this boat it was landed at Fort Monroe, Old Point Com- fort, Va., where it was mounted for trial. Mounting. This gun was mounted for trial on a wrought iron carriage, designed by Capt. A. B. Dyer, and built under his supervision at the Fort Monroe Arsenal. The carriage was of the form and style of the Columbiad barbette carriage, having its pintle in the middle transom of the chassis. The platform, on which the carriage was placed, was formed by excavating the sand on the sea-beech to the depth of about 2 feet, and filling in the cavity thus formed with concrete ; on this concrete, after it had perfectly set, was placed a layer of cut stone, about 15 inches thick, to which the traverse circles were bolted. The chassis was inclined, so that the gun and carriage, in recoiling, ascended at an angle of three degrees with the horizon. The maximum elevation of the gun, attainable on this carriage, was 28° 35'. The first trials were made with but one pair of traverse wheels under the front end of chassis, the wheels being of cast iron. AND QUALITIES OF CANNON POWDEE. 283 A few fires at the maximum elevation, and with the gun pointed in the same direction, sufficed to split both the front traverse wheels in two. These were replaced by another pair of cast iron wheels, banded with inch thick wrought iron bands; and an additional pair of Wrought iron wheels was placed as close as practicable to, and in rear of the others, an inner traverse circle being laid down to receive this pair of wheels. Diagonal braces were also added to the chassis and to the top carriage. With these additions this carriage stood all the subsequent firing of this gun, working perfectly well, and with much greater facility than had been expected. Trial The firing was commenced with 25 lb. charges of large grained (.6 in. diameter) powder, and shells weighing 320 lbs., strapped to oak sabots weigh- ing 11 lbs. The charges were increased, as the trial proceeded, by incre- ments of 5 lbs. up to 40 lbs. of the large grained powder, and as high as 50 lbs. of the perforated cake ; the maximum pressure of gas being deter- mined for each increment of the grained powder, and that due to 40 lbs. and to 50 lbs. of the perforated cake. The following table shows the mean results obtained : — Number of Fires for each result. Kind of Powder. Weight of Charge. Weight of Shell and Sahot. Pressure of gas per square inch, in lbs. 5 .6 in. grain. 25 lbs. 330 lbs. 8940 5 .6 30 311 10020 3 .6 35 330 13133 3 .6 40 311 18833 5 Perforated cake. 40 330 3280 3 u tt 50 325 8000 The charges of perforated cakes had generally from 3 to 5 lbs. of large grained powder in the bottom of the cartridge, to render the ignition of the charge more certain, which is included in the weight of charge given in the table. For the purpose of comparing the ranges and maximum pressures due to different kinds of powder, with each other, and with those obtained from the 15-inch gun with .6 in. grain, and with the perforated cake, 22 rounds were fired with solid shot of 126 lbs. weight from a 10-inch gun, with the mean results recorded in the following table : — 284 PROPERTIES OF METALS FOR CANNON, Name of man- ufacturer of powder. Kind of powder. No. of fires for each result. Weight of charge. Elevation. Maximum pres- sure of gasper square inch, in pounds. Range in yards. Time of flight. Dupont. No. 2. 4 16 lbs. 10° 14475 2423 10.05" a " 10. 4 16 10 71950 2882 11.16 a " 2. 16 30 15000 4335 23.00 a " 10. 16 30 65000 5283 24.53 Perf'dcake. 20 30 9000 4845 21.30 Dupont. .6 in. grain. 16 30 14800 4820 21.65 Hazard. No. 2. 16 30 31000 4860 Lost. a Cannon, 1859. 16 30 57000 5130 it ti tt 4 16 10 52000 2847 11.09" n No. 2. 4 16 10 43725 2796 10.67 Corresponding Results from the 15-inck Gun. o Maximum pr of gas per s inch, iu pou a bo § =M o a .3 'o PS - Perforated cake. 3 50 lbs. 345 lbs. 28° 35' 8000 5208 26.43" 60. Dupont. .6 in. grain. 3 40 330 28° 35' 18833 5088 26.37" 61. " .6 " 3 35 328 6° 00' 13133 1976 6.93" 70.5 tt .6 " 1 35 330 28° 35' - 5070 27.25" 58. The maximum pressures recorded in the last of the foregoing tables were not all determined from the identical same fires, from which the ranges and other results were obtained, but from the same number of other fires with the same charges of the same kind of powder. The No. 2 powder, of both Hazard's and Dupont's make, was of about the same size of grain as that used in firing the 15-inch gun (.6 in. diameter). The No. 10 powder was of about the same size of grain as common cannon powder. The .6 in. grained powder, used in firing the 15-inch gun, was of Dupont's make, and was purposely harder pressed than the ordinary powder; his experimental powder used in firing the 10-inch gun for the foregoing results was also much harder pressed than that of Hazard's make. The greatest range yet obtained with the 15-inch gun was with 40 lbs. of AND QUALITIES OF CANNON POWDEE. 285 .6 in. grained powder, and = 5730 yards. This, however, is not the greatest range attainable from this gun ; but as guns of this calibre are intended rather for direct fire at short ranges, than for curved fire at great distances, it was deemed advisable to first determine whether or not it was capable of enduring a satisfactory number of such charges as would likely be fired from it in actual service. The results in the foregoing tables show the ranges of the 15-inch gun, with 35 lbs. of .6 in. grain powder, to be about the same as those of the 10- inch gun, with 18 lbs. of powder, at corresponding elevations. These ranges require as high velocities, it is believed, as will be ordinarily required from guns of this calibre. After this gun had been fired 40 rounds, with charges varying from 25 to 40 lbs., a mixed Board of Engineer, Ordnance and Artillery Officers, was appointed by the Secretary of War to witness the farther firing, and to "report as to whether or not the efficiency of our present Armament for harbor defence would be improved by the addition of a judicious proportion of guns of this class." The Board was composed of the following named Officers, viz. : — General J. G. Totten, 1 Major J. G. Barnard, J- of Engineers. Captain H. G. Wright, J Major J. Symington, Captain A. B. Dyer, V of Ordnance. Captain J. Gorgas, J Colonel J. Dimick, 1 Major R. Anderson, }■ of Artillery. Captain J. H. Carlisle, J Lieutenant G. Tallmadge, 4th Artillery, Recorder. The firing before this Board embraced 49 rounds from the 15-inch gun, with full charges, and the 22, before referred to, from a 10-inch gun. Ten rounds were fired for accuracy at 2000 yards, with the results given in the following table, viz. : — 72 286 PEOPEETIES OF METALS EOE CANNON, Deviation. Number Weight and kind of charge. Weight of projec- tile and sabot. Elevation. Range, in yards. Time of flight. Recoil, in of fires. inches. Right. Left 1 35 lbs., .6 in. grain. 317 + 11 lbs. 6° 2017 7.00" 1 yard. 72 2 35 .6 " 317+ 11 6 1937 7.00 3 _ 70 3 35 .6 " 317 + 11 6 1902 6.30 1 - 69.5 4 35 .6 " 317 + 11 6 1892 6.35 5 _ 67.5 5 35 .6 " 318+ 11 6 1873 6.55 - 5 yards. 70 6 35 .6 " 317 + 11 6 1973 7.00 2 yards. - 69 7 35 .6 " 317 + 11 6 1970 6.75 i _ 68 8 35 .6 " 317 + 11 6 1979 7.00 6 _ 77 9 35 .6 " 317 + 11 6 1888 6.80 1.5 _ 70 10 35 .6 " 318+ 11 6 1930 7.00 4 — 72 The target at which this firing was done, was simply a stake driven into the ground, with a small red flag attached to its upper end, to render it visible. The elevations were measured with a gunner's quadrant, having a spirit level attached, and the horizontal sighting was done with a straight edge, held as near as could be judged by the eye, in the vertical plane of the axis of the gun, and a chalk mark on the upper element at the muzzle. The axis of the gun was about 8 feet above the plane on which the shells struck. When firing at the target, the ranges and lateral deviations from a line joining the gun and target, were measured on the ground. All ranges above 2500 yards were plotted from observations, on the columns of spray thrown up by the plunge of the shell into the water, made with two plane-tables placed at the extremities of a base line of 1900 yards. The times of flight were measured with a stop chronometer, reading hundredths of a second. The mean of the above ten ranges is 1936.1 yards, and the mean time of flight 6.775"; the mean velocity in feet would therefore — 1986 " 1 X 3 - — 857 6.775" "~ ' feet per second. Some trials were made to determine the initial velocities of the shells fired from this gun, with M. Navez's electro-ballistic pendulum ; but the results were so irregular as to be wholly unreliable, and its further use was aban- doned. An approximation to the initial velocity was then made by firing the gun with the bore horizontal, and measuring the fall of the shells in passing over a known distance. The shells were fired through a board screen at 885 feet AND QUALITIES OF CANNON POWDER. 287 from the muzzle of the gun. This distance, divided by the time required for a body to fall from the height of the axis of the gun to the centre of the hole in the screen made by the shell, would give the mean velocity of the shell in passing over this distance. The velocity thus determined was, for a mean of 2 fires, with 40 lbs. of large grained powder, and shells of 300 lbs., and rope grommets instead of sabots = 1328 feet per second ; and for a mean of 3 fires with 50 lbs. of perforated cakes, and shells of 315 lbs., and rope grommets instead of sabots, the velocity was 1282 feet per second. One fire with 35 lbs. of large grains, shell 318, and 11 lb. sabot, gave a velocity of 1135 feet per second. Four rounds were then fired with 40 lb. charges of large grained powder, at 10° elevation, with the following results, viz. : — Order of fire. Weight of shell and sabot. Range, in yards. Time of flight. Becoil, in inches. 1st, 2d, 3d, 4th, 315 + 11 lbs. 337 + 11 319 + 11 319 + 11 2700 2910 2754 2760 11.48" 11.30 10.80 11.06 78 85 78 78 Mean, 2781 11.16" 79.75 The above results give the mean velocity of the shell = 747 feet per second. The greatest recoil obtained was, with 40 lbs. of .6 in. grained powder, gun horizontal, =90 inches. Trials for Ricochet on Water. In these trials the gun was 13.5 feet above the surface of the water, and was fired 5 times with 40 lb. charges of large grained powder, and 318 lb. shells, with 11 lb. sabots, with the following results, viz. : — 1st fire, 5° elevation, range 2085 yards, no ricochet. 2d fire, gun horizontal, range 1785 yards, 3 ricochets. 3d fire, gun horizontal, range 1781 yards, 4 ricochets. 4th fire, 5° elevation, range lost, no ricochet. 5th fire, 1° elevation, range lost, 4 ricochets. The surface of the water was not smooth, nor very rough, during this trial ; 288 PEOPEETIES OF METALS FOE CANNON, but the results indicate that lighter projectiles, moving with higher velocities, are better adapted to ricochet firing. Loading and Manoeuvring. The method of loading was as follows, viz. : — One man placed the shell hooks upon the shell, the hooks taking into .6 in. holes bored .6 in. deep at the extremities of a diameter perpendicular to that through the fuze hole, another inserted a handspike through the ring of the shell hooks ; then four men, two at each end of the handspike, raised up the shell, carried it up the ramp, and held it opposite the muzzle of the gun, while a fifth man revolved it upon the shell hooks and inserted the sabot into the muzzle, pushing the shell in till the hooks came against the face of the muzzle ; the hooks and handspike were then taken away by one man, while the other four rammed the shell home. Three men could load this gun, two carrying the shell, and holding it oppo- site the muzzle, while the third inserted it, as above described ; but five is the proper number for this purpose. The time occupied by seven men in running the gun to battery, depressing it from maximum elevation, sponging, loading, and elevating ready to fire at maximum elevation, was for the first trial 4', and for the second 3' 10". With the gun horizontal, the time of running it to battery, sponging and loading, was, for the first trial, V 52"; for the second, V 28"; for the third, V 10"; for the fourth, V 15"; for the fifth, V ; and for the sixth, V 3". The time occupied in traversing the gun through an angle of 90° by the same seven men, (one sergeant and six negroes) was 2' 20" ; and in traversing back 45°, the time was 1'. Two men could run the gun to battery, but four is the proper number for this purpose ; and four could run it from battery, but eight is the proper number. To determine the effect of the explosion of a loaded shell from this gun, one round was fired with 8 lbs. large grained powder, and shell of 117 lbs., containing 17 lbs. of cannon power, elevation 28° 35'. The shell fell on the beech 1430 yards from the gith, and exploded in the sand, forming a crater 40 inches deep, 12 feet long, and 10 feet wide. This concluded the firing before the Board, and the bore of the gun was then carefully examined with AND QUALITIES OF CANNON POWDEE. 289 the Star Gauge, but no enlargement above its original diameter could be detected in any part. The following is an extract from the Report of the Board : — " Considering, therefore, the inappreciable injury which the 15-inch gun has sustained from the trials to which it has been subjected, and the facility and rapidity with which it was manoeuvred and fired, the Board is decidedly of opinion that the introduction of guns of much larger calibre than any now in the service, is desirable and practicable. " Before, however, the actual adoption of such a class of guns, the Board think that a series of experiments should be made to test fully their endur- ance and their effect upon targets made after the manner of ships' sides and floating batteries, with and without the iron clothing. " Qualified by the above remark, the Board, therefore, in the words of the order under which they are acting, report that, in their opinion, ' the efficiency of our present armament for harbor defence would be improved by the addition of a judicious proportion of guns of this class.' " On the receipt of the Report of the Board, the Secretary of War directed the proof of the gun to be continued, with service charges, till the number of rounds fired should reach 500. Accordingly the firing was resumed on the 18th of December, 1860, and continued with 35 lb. charges, .6 in. grained powder, till a total of 200 rounds had been fired, when the bore was again carefully examined, without detecting any enlargement. An impression of the interior of the vent was taken, which showed it to be of a nearly regular elliptical form, the greatest diameter being .4 in., and the least .3 in. The exterior of the vent was not perceptibly enlarged. The firing was then continued with the same charges and kind of powder, till 77 additional rounds had been fired under my supervision; when, in order to enable me to proceed with the preparatory tests of metal for, and to supervise the casting of the 12-inch Rifle Gun at Pittsburg, Captain A. B. Dyer kindly volunteered to continue the firing for me. The firing was continued with 35-lb. charges, and a shell, till 356 rounds had been fired. Several series of firing were then made, for the purpose of determining the initial velocities of shells, the maximum pressure of gas, the ranges and times of flight, at different elevations, due to charges of .6 in. grained powder varying from 35 to 50 lbs. 290 PEOPEETIES OF METALS EOE CANNON, The pressures of gas were determined by means of the internal pressure gauge, and the initial velocities by Captain J. G. Benton's electro-ballistic pendulum, which Captain Dyer represents as giving very satisfactory results. The results obtained by Captain Dyer from these series, are given in the following tables : — Table showing the initial velocity and maximum pressure of gas, due to different charges of .6 in. grain powder. S a 03 O CO a> as 1° U IS o o a o F 0> o % a a .2 A -** a> CD 6b CO _H oT be S ba ■3 C4H o CO a EH Initial velocity shell, in feet, second. Pressure of gas square inch, pounds. 'u Cm o 'o ti CD M 11 .6 in. grain. 35 lbs. 300 lbs. - - - 902 - - 10 .6 40 300 - - - 949 - - 5 .6 45 315 25° 00' 4595 23.24" - 12528 58 in 4 .6 45 315 - - - 1064 - - 4 .6 50 315 - - - 1118 - - 5 .6 50 315 25° 00' 4686 23.29" - 12568 - Table of ranges of shells fired from the 15-in. gun, at different elevations. No. of fires Tf\r aorin Kind of Weight of Weight of Elevation, in Range, in Time of Recoil of iUi. C£LU11 result. powder. charge. shell. degrees. yards. flight. carriage. 5 .6 in. grain. 40 lbs. 301 lbs. 0°00' 273 .88" _ 5 .6 40 302 1 00 484 1.66 - 5 .6 40 302 2 00 812 2.49 — 5 .6 40 300 3 00 1136 3.43 - 5 .6 40 300 4 00 1310 4.37 - 5 .6 40 302 5 00 1518 5.07 - 4 .6 40 302 6 00 1760 5.96 - 5 .6 40 298 7 00 1948 7.11 - 5 .6 40 315 8 00 2194 8.17 35.5 in. 5 .6 40 315 9 00 2236 8.87 30.5 6 .6 40 315 10 00 2425 10.00 32.95 5 .6 40 315 12 00 2831 12.07 30.5 5 .6 40 315 15 00 3078 13.72 47.25 5 .6 40 315 20 00 3838 17.82 ' 46.6 5 .6 40 315 25 00 4528 22.03 43.6 6 .6 40 315 28 00 4821 24.18 40.75 6 .6 40 315 30 00 5018 26.71 23.16 2 .6 40 330 30 00 4832 25.55 25.5 5 .6 45 315 25 00 4595 23.20 62.45 5 .6 50 315 25 00 4686 23.29 65.65 AND QUALITIES OF CANNON POWDER. 291 This gun has been fired 505 rounds with full charges, and 4 rounds with smaller charges. It will be seen that the maximum pressures, and the ranges, for correspond- ing charges and elevations, are something less in the foregoing tables than those obtained from the firing before the Board. This is due to the greater density of the powder used by Captain Dyer than that used by me in firing before the Board. After 500 rounds had been fired the gun was again carefully washed out and examined with the Star Gauge, and by means of light thrown into the bore, and Captain Dyer informs me that " no wear or enlargement was dis- covered," and that " the interior of the vent was enlarged, but less than vents usually are after 500 rounds." These results leave no doubt of the ability of this gun to endure an almost indefinite number of fires, with charges equal to the maximum hitherto fired from it. It is also certain that much higher velocities, and much greater ranges, than any yet reached with this gun, may be safely attained, as I have the utmost confidence in its ability to endure 1000 rounds with charges giving a maximum pressure of gas double as great as the greatest to which it has yet been subjected ; and if the perforated cakes be used in such charges as to give this pressure, the resulting ranges, at the maximum elevation, must be considerably over four miles. The perfect reliability of this gun, for all the requirements of service, having been fully established by the trials to which it has been already sub- jected, it is now deemed of the highest importance that the effects of its projectiles, upon masonry, and especially upon iron-clad and steel-clad targets, should be experimentally determined, with the least possible delay : after which it is recommended that it be subjected to a series of firing, with increasing charges, for the purpose of determining the greatest safely attain- able range of its projectiles, and the proper diameter of grain powder, and the proper thickness of walls in the perforated cake cartridge, for its ordinary service. Of the Perforated Cake Cartridge. This form of cartridge was arrived at from the following principles and theoretical considerations : — 292 PEOPEETIES OF METALS FOE CANNON, The projecting charge should be so related, in its rate of combustion, to the form of the gun from which it is fired, that, with a given convenient thick- ness of metal and length of bore, the maximum velocity of shot attainable from such gun should be produced ; or that a given velocity of shot should be produced with the minimum strain upon the gun. This requires that every part of the gun should be subjected to the same proportional strain, or "bursting tendency," at each discharge. If the resistance which the gun could offer were independent of the length of surface subjected to pressure, as would be the case if the tangential resist- ance alone were brought into play, the foregoing conditions would require that the pressure of the gas should be uniform throughout the entire length of the bore. Now let us see in what manner this condition is fulfilled by our present form of cartridge. The most favorable hypothesis that can be made with grained powder, is that the grains are spherical, and of uniform size. If, now, we suppose the powder to be so hard pressed that the gas cannot permeate the grains, and that the diameters of the grains undergo equal reductions in equal successive portions of time, we shall, at the end of half the time required for its total combustion, have consumed seven-eighths of the whole charge, the volumes of spheres being to each other as the cubes of their diameters, while the shot will have traversed only something more than one- fourth of the bore, supposing the time required for the combustion of the charge to = that required for the shot to traverse the entire length of bore. From which we see that with this kind of charge the gas is evolved in the inverse order of what it should be ; the evolution being greatest while the velocity of the shot is least, and least while that velocity is greatest, and giving rise to excessive pressure at and near the seat of the charge, and a too rapid diminution of pressure from that point forward. This highly objectionable property of grained powder may be remedied, in some degree, by increasing the size of grain, if it be sufficiently hard pressed to render it impermeable to gas under the pressure to which it is to be subjected in firing; for it is clear that the initial burning surface, the maximum pressure, the difference between the mean and maximum pres- sures, and the difference between the initial and the terminal burning surfaces, will all diminish as the diameter of grain increases. But to bring the maximum pressure within proper limits, requires such an increase in diameter of grain as to require either an inconveniently long gun, AND QUALITIES OF CANNON POWDER. 293 or an increase of charge, in order to produce the requisite velocities : since for a gun of ordinary length the grains would not be entirely consumed in the gun ; and the larger the grain, the more of each would be blown out un- burned. And in soft, lightly pressed powder, we lose all control over the rate of combustion, the larger grain sometimes giving the greater pressure, owing to the larger interstices causing a more rapid rate of inflammation. The condition of uniformity of pressure along the entire length of the bore, would be theoretically fulfilled by a cartridge so constructed that ignition should take place only on the inner surfaces of elementary cylinders of constant lengths, and that the combustion of the charge should increase the radii of these cylinders equal quantities in equal successive portions of time ; for since, by hypothesis, the radii of the cylinders increase, directly as the times, and since the areas of the cross sections of the cylinders, and consequently the volume of the charge consumed, and of the gas evolved, are to each other as the squares of the radii of the cylinders, it follows that they will also be to each other as the squares of the times. But the dynamical equation, expressive of the circumstances of motion due to a constant accelerating force, is S=\ gt 2 ; in which S= space passed over, g = constant force, and t = the time. Or the spaces passed over, (which, in this case, are the volumes of gas, or the spaces behind the shot,) are to each other as the squares of the times. But experiments on bursting cylinders by pressure exerted upon un- equal lengths of bore (page 145) show that, owing to the transverse resist- ance developed, for short lengths of surface pressed, the resistance which a cylindrical gun could offer to a bursting force would be greater for a length of two calibres from the bottom of the bore, than for one of seven calibres, or any greater length, in the ratio of 3 to 2 ; and the pressures of the gas, at these points of the bore, ought to be to each other in the same ratio. And this requirement may be almost exactly met in practice by establishing the proper relation between the initial burning surface, or between the number and diameter of the cylindrical holes, and the thickness of the walls between them. The initial burning surface, and the ratio of the maximum to the mean pressure, may also be varied by varying the number and thickness of the cakes in a given weight of charge ; the initial burning surface and the 294 PEOPEETIES OF METALS FOE CANNON, maximum pressure both increasing with the number of cakes, since the burn- ing surface extends over the whole surface of the cakes. The thickness of walls between the cylinders, should be such as to be burned through, or consumed, before the projectile leaves the gun ; and for ordinary velocities we should economize in weight of charge, by making the walls of such thickness as to burn through by the time the projectile has traversed two-thirds or three-fourths of the bore, and allowing the gas to act expansively from there to the muzzle. It will readily be seen, from the foregoing, that this form of cartridge gives us entire control over the rate of combustion of the charge, a fact of which the importance can hardly be overrated ; for, taken in connection with the hollow mode of casting cannon, it removes all limit, as regards safety, to the calibre, of which even cast iron guns may be made. In practice, the axes of the cylindrical holes through the cakes are parallel to that of the bore, when the cartridge is in position, and the cakes vary, as hitherto constructed, from one to two inches in thickness ; and those used in firing the 15-inch gun were made of Hazard's cannon powder of 1857. Manner of Forming the Cakes. The cakes are formed by placing either mealed or grained powder, moistened with about three per cent, of water, in a mould of the proper form, and subjecting it to a pressure such as to render it impermeable to gas, under the pressure to which it will be subjected in the gun. The mould is placed on a bed piece, pierced with the number and size of holes to be in the cake, the holes in the bed piece being covered by a sheet of paper, to keep the powder in the mould from entering them. A piston, of the same cross section as the interior of the mould, and having screwed into its lower end, and parallel to its axis, the same number of cylindrical wires with conical points, and of the same diameter, as the holes in the bed piece, is mounted at its lower end with a follower, having the same cross section as the mould, and pierced with the same number of holes. This follower is slipped on, over the teeth, till it comes against the end of the piston, and is of such length as to penetrate the mould far enough to give the proper thickness and degree of pressure to the cake. The teeth project through this follower far enough to pass entirely through the loose powder in the mould, and enter the holes in the bed piece, before the follower enters PLATE fJ. Perforated fa. hi far/jisft/f: ^ * *, fc * \, * * fc ** \ fc *► * ** * •«*«ii,*'«/ — ' *D ka ** • \i- ** *• < i. *" «> ft- * » S ■ ■■■# ft ** * «*■'*' ''.«>, *? *» «. * «V-"*-fti*L *